DEV Community: OleksiiTsymbal

IoT in Manufacturing: 6 Industrial IoT Trends in 2022

OleksiiTsymbal — Wed, 30 Mar 2022 14:48:51 +0000

The Internet of Things (IoT), is all about connecting devices into networks that work together. This follows a shift in design from monolithic machines to segmented ones. In other words, IoT is all about micro processing and splitting larger systems into various smaller ones. By doing this, managing the network, upgrading it, and maintaining it become substantially easier.

In manufacturing contexts, industrial Internet of Things (IIoT) is important for a number of reasons. One of the most important is IoT analytics. Manufacturers need data to understand what the status of their performance is and what’s slowing them down. From there, they can make decisions to improve the efficiency of their business. This process can even be automated by combining IIoT systems with artificial intelligence.

However, it’s important to think carefully about how IIoT technologies apply to your business. Combining technologies together in new and innovative ways is one of the best ways to remain competitive as a business. It’s important to consider the existing manufacturing IoT trends this year in 2022 to get ideas about how exactly these technologies can benefit you.

Trend #1: Connectivity Technologies

One of the greatest bottlenecks of IoT technology in manufacturing is bandwidth. How much data can all of these devices in a network exchange at a time? The more data that can be communicated at once, the faster and more efficient these systems will be within alternative setups. Network speed is essential for the success of real-time applications and edge computing. However, if data does not need to be communicated as fast as possible, then extremely high bandwidth may not be needed. It’s important to evaluate the needs of your manufacturing business so that you can choose the best options.

WIRED IIOT SOLUTIONS

For the most consistent and fastest solutions, wired IIoT connections are the best options for manufacturers looking to connect multiple IoT devices together into a network. These use EtherCAT, Ethernet/IP, and Profinet protocols. USB connections are limited in their speed and range, so typically category cables are used to extend their range. In more distant scenarios, fiberoptic cables can be used to connect industrial facilities miles apart from each other.

Being more mature than wireless technology, wired solutions offer proven reliability and noise immunity. However, the greatest downside of these connections is the physical component. Wiring can take up space, have environmental limitations, and is more time consuming to set up. Wired connections also add the additional costs of cabling.

WIRELESS IIOT SOLUTIONS

For versatility and ease of setup, wireless connections are far more effective. Although wireless connections are inherently vulnerable to inconsistency, this can be mitigated depending on the type of wireless connectivity used. There are several technologies in use and on the rise today in 2022 for manufacturing IIoT systems:

Source

Bluetooth Low Energy is a technology that is more efficient than Wi-Fi and Zigbee. It is a great choice for devices that are portable and battery powered. However, less data can be transferred at once due to the energy restrictions. It also is a subject to interference in the 2.4 GHz range both Bluetooth and old-school Wi-Fi networks may use, affecting how the devices work.

Zigbee is an alternative connection type that works best with battery-operated sensors with low data throughput. This relies on nodes that interconnect multiple pathways. A central hub coordinator is needed which increases the complexity of the system.

Wi-Fi has much more potential for machine sensors in factories. With 5 GHz access points, these points can provide high speed connections to devices up to 190 feet away. To provide maximum coverage, multiple beacons can be spread throughout the factory floor. This can be the best way to support IIoT sensors that are cable powered with fixed locations. Since additional communications cabling is not necessary, setup of IIoT sensors that can monitor machinery is simple. High speed Wi-Fi connections are one of the best solutions for industrial Internet of Things applications for factories in 2022.

Trend #2: Predictive Maintenance and Analytics

Breakdowns in manufacturing centers are extremely costly. With predictive maintenance provided by artificial intelligence, organizations can save millions. However, industrial machine learning algorithms cannot function without high quality data about the machinery they are evaluating. Industrial Internet of Things sensors can collect data across a network of machines. This data can then be used to identify which machines need to be preemptively scheduled for maintenance and when.

These sensors can also measure temperature, vibration, and electricity usage of machines to estimate potential future points of failure.

Trend #3: Automated Quality Assurance

Thanks to Industrial Internet of Things networks, quality assurance monitoring can be done remotely and automatically. This can improve productivity and efficiency of manufacturing businesses greatly. Real-time alerts can be sent to allow for more rapid response to issues like unexpected machine failures and other disruptions. Real-time video connectivity through IIoT devices also supports artificial intelligence efforts like automated visual inspection. This allows for AI to detect defectives and remove them from the assembly line before they can be shipped off. AI-driven visual inspection solutions would not be possible without IoT sensors and cameras to give eyes and ears to the decision-making process.

Trend #4: Edge Computing

One of the most interesting trends in the sphere of Industrial Internet of Things technology is edge computing. Many industries and businesses have shifted toward offloading processing from local devices to far-away servers that do data processing for them. While this reduces the amount of processing that the local device like a cell phone or PC has to do, it is costly when it comes to time and bandwidth. The goal of edge computing is the opposite, keep processing as close to the ‘edge’ as possible.

In manufacturing contexts, several devices in the local edge network at a factory can handle the processing without having to send data elsewhere for processing. Not only is this faster and more efficient, but it is also inherently more secure. Since the data never leaves the factory, there is no risk of it being intercepted or recovered by a third party.

Forward-looking industrial companies leverage new opportunities by fusing edge computing and AI into Edge AI. The Edge AI concept allows AI computation to be done near the user at the edge of the IoT network, instead of a cloud. That helps bring real-time intelligence to industrial processes, increase privacy and enhance cybersecurity, at the same time reducing costs and securing persistent improvement of the manufacturing processes.

Trend #5: Location Tracking

Location tracking has various applications to manufacturing, and all of them rely on Industrial Internet of Things technologies. While GPS is well known to be extremely effective in most outdoor contexts, indoor positioning systems and areas with GPS interference like dense cities with tall buildings can be more challenging to work with. Outdoor solutions are generally in the domain of logistics, but indoor solutions are within the realm of manufacturing.

Real-time location systems (RTLS) are based on wireless technologies, such as Wi-Fi, BLE beacons, UWB, and RFID. They can help identify where products are on the factory floor which can monitor their progress from beginning to end of the production process. This not only can help verify quality assurance, but it can also provide additional data to support digital twins applications.

Trend #6: Energy Optimization

You have probably at some point wandered into a dark room, only for the lights to turn on by themselves. It may not seem like it at first, but we have a lot to learn from motion-sensing light switches. Their goal isn’t just to make it easier for us to turn on the lights. Their underlying goal is to make it so that if the room isn’t in use, the lights turn off by themselves. Where this crosses over into the Industrial Internet of Things is the concept of energy optimization. We can optimize lights in our factory, but how can we optimize energy consumption of other devices?

Energy optimization can be achieved through various ways, including paying close attention to temperature control systems, industrial machines, and more. A report by EIA reveals that in 2020, 33% of the total United States energy consumption can be attributed to the manufacturing sector. Not only is energy optimization better for the environment, but it can result in significant cost savings.

By using IIoT energy optimization sensors to monitor the electrical status and usage of devices and machines in a factory, operators can fine tune the process and automatically optimize energy usage by various devices. However, this is only one piece of the puzzle. Creating more eco-friendly manufacturing processes will take much more than just IoT sensors and electrical data processing.

The Future of IoT in Manufacturing

Ultimately, the future of the Industrial Internet of Things depends on a number of factors. Thanks to a chip shortage that began last year and that is still plaguing the market, there is a limitation on the number of devices that businesses can field at more affordable prices. With sanctions against Russia intensifying due to the recent invasion of Ukraine, Bloomberg reports that the shortage is going to get worse.

However, this doesn’t mean that IIoT devices are impossible to acquire and implement, and existing systems can be improved as well. In times of disruption, innovation is extremely important. Although the chip shortage is straining expansion of the market, those who can solve problems with existing hardware and resources will remain competitive throughout the rest of the shortage. These businesses will continue to thrive after the shortage ends as well, resulting in even more gains.

Future of IoT Technology: 8 Trends for Businesses to Watch in 2022

OleksiiTsymbal — Thu, 17 Mar 2022 14:43:15 +0000

In a world dominated by artificial intelligence, data, and ever-advancing connectivity technologies, it’s hard to leave the ‘Internet of Things’ out of a list of innovative and game changing technologies. In fact, IoT may be one of the most important technologies out there right now, being responsible for the success of many other technologies like machine learning.

As the market landscape evolves over the next several years, it’s critical for businesses to monitor how things are changing. Some of the most successful businesses are the ones who think creatively about evolving technologies. Coming up with ideas for innovative ways to use and combine these technologies together isn’t possible without keeping an eye on these trends. Let’s talk about the future of IoT technology and IoT trends that are shaping markets around the world in 2022.

IoT Technology: A More Connected World

As machines get smaller, it becomes easier to produce. A shift in tech philosophy has occurred recently. Industries have shifted away from a monolithic, single-device model in favor of a more modular, microservices approach. Instead of one device handling every calculation and every measurement, a network of devices can do this instead. Each device may have its own individual utility as well, and it may benefit the entire network. This is the core of IoT technology.

IOT TECHNOLOGY’S GROWTH

According to Mordor Intelligence, the IoT technology market value is expected to rise to $1.39 trillion by 2026. This incredible growth is likely due to a number of factors:

The COVID-19 pandemic accelerated the advancement of remote monitoring, smart home devices, and data analysis solutions.
Businesses are racing to develop better artificial intelligence solutions. These often require a network of advanced sensors and edge computers within the scope of IoT.
IoT networks can accomplish some tasks more efficiently than centralized solutions.

Don’t get too excited just yet though. There’s more to the story.

LIMITED PRODUCTION: THE CHIP SHORTAGE

Due to high demand and low supply of valuable semiconductor chips, IoT solutions have become more expensive to produce. The situation has only worsened due to the pandemic. Although production of these chips has increased, it will take a long time before the chip shortage is over. Some businesses are choosing to order their semiconductor inventories far in advance instead of using a just-in-time solution.

According to Forrester, the semiconductor shortage in 2022 will limit the growth of IoT markets by 10 to 15%. That sounds like bad news for businesses looking to invest in new IoT solutions, but it’s not the whole story.

There is good news on the horizon. In February of 2022, the European Union passed the European Chips Act, a combination of public and private investments to support efforts to resolve the supply chain shortage.

IOT IS POWERED BY BANDWIDTH

The main bottleneck of Internet of Things technology is bandwidth. The higher the bandwidth, the lower the latency of an IoT network. The faster a device can communicate with another device, the more fluid and efficient IoT technology will be. If there’s too much latency, there’s not much of an advantage for IoT over its centralized alternatives.

In the past, high-speed networking was limited by cable and fiber connections. An IoT network can be severely limited if every device needs to be connected with an Ethernet cable. Wireless connections can be inconsistent and have slower data rates. However, IoT has surged over the past several years due to advancements in 5G networks and Wi-Fi 6 (802.11ax). With faster wireless data rates than ever, the potential of IoT can be realized.

Trend #1: AIoT – Artificial Intelligence & IoT Technology

One of the most fascinating use cases of IoT technology is to support artificial intelligence software. Artificial intelligence and IoT have a mutualistic relationship. AI is benefitted by IoT with distributed data, and IoT is benefitted by AI with advanced management.

A NETWORK OF DATA

Since artificial intelligence technologies are heavily data driven, IoT sensors are an immense asset to the machine learning data pipeline. Research and Markets reports that AI in IoT technology will reach a value of $14,799 million by 2026. High quality data is extremely important for the success of machine learning techniques. For example, live data from IoT sensors monitoring factory equipment can help manufacturing-based machine learning algorithms determine when equipment needs to be serviced in the future, a practice called predictive maintenance.

VISUAL INSPECTION: AI AND IOT WORK TOGETHER

Visual inspection is another area where IoT technology and machine learning synergize to enrich industrial and distribution industries. Check out our video about AI-driven visual inspection technology to see this in action.

Although machine learning is extremely adept at identifying patterns, it can’t do that without high quality data. With rising bandwidth and increased opportunities for IoT networks to take center stage in various industries, machine learning’s use will increase over the next few years.

Trend #2: IoT Connectivity — 5G, Wi-Fi 6, LPWAN, and Satellites

The main challenge that IoT networks have had to overcome in recent years is wireless data rates. As these technologies improve, so too will the aspects of IoT technology, including sensors, edge computing, wearables, smart homes, and more. Recently, more infrastructure has been developed for newer connectivity types that make IoT solutions more feasible. These are connectivity technologies like 5G, Wi-Fi 6, LPWAN, and satellites.

5G: ADVANCED MOBILE NETWORKS

In many IoT technology solutions, connectivity infrastructure needs to be set up before an array of edge devices, sensors, or other devices can be maintained. However, mobile networks like LTE can be a potential alternative for certain situations such as outdoor settings. However, 4G LTE is limited by bandwidth. 5G networks however are much faster and can support data processing needed for IoT networks much more efficiently.

WI-FI 6

For indoor settings, Wi-Fi operating in the 6 GHz band increases the bandwidth potential of IoT technology greatly. The faster that a network of devices can communicate, the more reliable that the system will be. As a bonus, Wi-Fi 6 can be used in households, meaning that this can offer great benefits for smart home IoT networks.

LPWAN

Low-power wide-area network connectivity is an emerging technology that is effective for connecting devices with low-bandwidth usage with low bit rates over larger areas. This makes it a good choice for IoT devices that communicate with one another on a machine-to-machine basis. LPWANs are more energy efficient, making them more cost-effective. If you need to use a large number of devices over a large area, LPWAN technology is a good choice.

SATELLITE

In some cases, IoT technology can be powered by satellites for geographically separated networks. Powered by Globalstar satellites, Traksat’s satellite-powered IoT devices enable humanitarian staff to report emergency incidents to immediately request assistance. GPS information is immediately recorded and sent to a headquarters for rescue preparations.

Trend #3: Edge Computing – Low Latency & Security

Real time applications depend on edge computing. Instead of calculating everything at a central source, edge networks process information closer to the user and lighten the load of the entire network for all users.

Not only can edge computing reduce latency of IoT technology, but it also has potential to increase the security of data processing. If data can be processed on an edge device instead of being transmitted to a central server, there are fewer opportunities for it to be intercepted by hackers. All that needs to happen is for the information to be exchanged with the edge device and then handed right back to the user. The information doesn’t need to be stored using memory in this case.

Edge computing is useful for any situation where decisions need to be made quickly. This is especially true in contexts like security and safety. Automatically stopping machinery from operating while someone is in a restricted area at a factory is one way that IoT edge computing can be used to protect people from harm. Autonomous vehicles need to receive data to make critical decisions in real-time, and this can mean the difference between life and death on the road.

Trend #4: Wearable IoT Technology

While sensors and edge devices are important for many IoT technology solutions, wearable IoT devices shouldn’t be overlooked. Smart watches, earbuds, and extended reality (AR/VR) headsets are important wearable IoT devices that are making waves in 2022 and will only continue to evolve.

THE FUTURE OF IOT SMARTWATCHES AND WRISTBANDS

In the past, it was expected that wearable technology, smartwatches in particular, would eventually replace the smartphone and desktop computing. However, this prediction doesn’t look like it will become reality anytime soon. The limited feasibility of wearable devices like smartwatches makes them much less likely to be useful for tasks you could complete on a smartphone or laptop.

However, wearable IoT technology has immense potential to aid in medical roles due to its ability to keep track of patient vitals. These devices can perform tasks such as automatically alerting others in case of emergencies and gathering of continuous health records.

AR AND VR

If IoT technology can provide devices with real-time contextual data, then it’s clear how this can greatly benefit the performance of AR and VR devices. Being provided with sensor data and real-time network information on a head mounted display can be very helpful for professionals in these situations:

Doctors performing surgery aided by head-mounted AR, where wearables and other sensors provide data to the surgeon.
A manufacturer reviewing a digital twin simulation of a factory layout in virtual reality. Data from the real factory floor’s IoT sensors can be shown alongside the digital twin for performance comparison.
A distribution center worker using AR on a mobile device to locate the position of an item in a warehouse, where IoT devices help locate the item and send directions to the worker.

Trend #5: Smart Homes

The rise of Google Assistant, Amazon Echo, Apple’s Siri and other digital assistants have transformed the smart home industry. With at-home IoT technology in 2022 now able to manage innumerable devices like lights, appliances, and even home security systems, the technology seems to have reached an apex. However, there is still much more room for growth that we will see over the next several years. According to Mordor Intelligence, the smart home market will have a CAGR of 25% between now and 2025, allowing the industry to reach a size of $246 billion.

One of the next steps forward is a focus on smart home automation. Smart home IoT networks are now expanding their ability to automate tasks like lighting, temperature control, and security. These can be configured manually by consumers, or they can be tuned automatically using AI algorithms that review sensor and usage data.

On the security side of IoT smart home devices, at MobiDev we collaborated with CUJO AI to build a mobile app that allows users to monitor the security of their home network.

Trend #6: Smart Cities

There are many applications for IoT technology when developing smart city networks. Traffic monitoring is one of the most pressing issues. Being able to monitor traffic with sensors placed city-wide enables better control over intersections for better traffic optimization. Water level monitoring can also be useful to help detect and inform citizens about potential and active flooding. This information can also guide additional action to prevent flooding in the future.

Smart streetlights have a variety of possibilities. As platforms for sensors, they can be used to detect weather conditions, monitor traffic, host public Wi-Fi, and conduct surveillance. These streetlights can also be used as an edge device that can control digital street signs. They may even have the streetlights mounted on the light pole itself.

Smart Columbus in Ohio has experimented with self-driving shuttles, a futuristic experiment which shows us that in the future, smart cities can help reduce the need for private vehicles.

In the video below, you can learn more about MobiDev’s approach to IoT smart parking system development.

Trend #7: IoT Technology in Healthcare

Earlier on we talked about wearables and healthcare, but there are other use-cases for IoT technology in this sector. For example, WebRTC unified with IoT networks can provide more efficient telemedicine in certain areas where it may be more efficient to use edge devices than a traditional Internet connection. There are also plenty of traditional IoT use-cases such as hospital-room sensors that can monitor patient vitals over the course of the day to aid doctors in diagnosis and treatment. The evolution of this technology has accelerated due to the spread of COVID-19.

Trend #8: Connected Networks Aid Manufacturers

IoT technology has a great deal of potential to advance the manufacturing industry. With arrays of sensors on factory floors, this industry has become more automated than ever before. One of the most important results of the expansion of IoT sensors in manufacturing is that these networks are powering advanced artificial intelligence applications. AI can’t provide solutions like predictive maintenance, defect detection, digital twins, and generative design without critical data provided by sensors.

What’s Next for IoT Technology in 2022 and Beyond?

The future of IoT technology lies in each of the previous trends and beyond with possibilities extending into many areas like the retail industry and indoor navigation. It’s also important to realize that the Internet of Things is not a technology that functions alone in many cases. Synergizing technologies like IoT and AI is what enables businesses to cultivate innovation and remain relevant against their competitors. All it takes to get started is an idea or a vision to modernize your business’s infrastructure to take the future head on.

How to Mitigate IoT Security Threats in 2022

OleksiiTsymbal — Tue, 22 Feb 2022 12:30:38 +0000

The Internet of Things plays a key role in digital transformation. However, in many cases, organizations realize that they already have a large fleet of legacy IoT devices that have been gradually deployed over the years. Many of these devices may not have been designed with security in mind.

One of the biggest concerns of IoT is managing the risks associated with a growing number of IoT devices. Information security and privacy issues related to IoT devices have attracted global attention, because of the ability of these devices to interact with the physical world. IoT vulnerabilities continue to emerge, making it critical for manufacturers to emphasize IoT security by design.

IoT vulnerabilities have been discovered and exposed across many industries. These vulnerabilities threaten sensitive data as well as personal safety. Without a doubt, the Internet of Things is a prime target for hackers in the 2022, and any organization that produces or uses these devices needs to be prepared.

IoT Security Threats

Below we briefly review some of the common cybersecurity threats facilitated by IoT devices.

IOT BOTNETS

IoT devices are attractive targets for botnet builders—these are hackers who compromise millions of devices, connecting them into a network they can use for criminal activities. IoT devices are a good candidate for botnets due to their weak security and the large number of virtually identical devices, which attackers can compromise using the same tactics.

Attackers can use unprotected ports or phishing scams to infect IoT devices with malware, and enlist them into botnets that can be used to launch large-scale cyber attacks. Hackers can use readily available attack toolkits, able to detect sensitive devices, penetrate them, and avoid detection. Another module in the toolkit then instructs the device to launch an attack or steal information on behalf of the botnet owner.

Threat actors often leverage IoT botnets during distributed denial of service (DDoS) attacks; see the example attacks section below.

DATA EXFILTRATION

When hackers use malware to infect IoT devices, they can do more than just enlist the device into a botnet. For example, attackers can access the device data and steal any sensitive information stored there. Attackers also leverage IoT to harvest credentials from device firmware. Using these credentials, attackers can gain access to corporate networks or other systems storing sensitive data. In this way, an attack on a seemingly innocent device can turn into a full-scale data breach.

SHADOW IOT

Shadow IoT arises because IT administrators do not always have control over devices connected to the network. Devices with IP addresses, such as digital assistants, smart watches, or printers, frequently connect to corporate networks, and do not always meet security standards.

Without knowledge of shadow IoT devices, IT administrators cannot ensure that hardware and software have basic security features, and find it difficult to monitor malicious traffic on devices. When hackers compromise these devices, they can leverage the connection to the corporate network and escalate privileges to access sensitive information on the corporate network.

Notable IoT Security Breaches and Hacks

Since the IoT concept was born in the late 1990s, security experts have warned that devices connected to the Internet will pose a risk to society. Since then, numerous large-scale attacks have been publicized, in which attackers compromised IoT devices and created a real threat to public safety and corporate security. Here are a few examples.

STUXNET

In 2010, researchers discovered that a virus called Stuxnet caused physical damage to nuclear centrifuges in Iran. The attack began in 2006, with the primary stage of the campaign in 2009. The malware manipulated commands sent from programmable logic controllers (PLC). Stuxnet is often considered an IoT attack, amongst the earliest targeting a supervisory control and data acquisition (SCADA) system, used in industrial environments.

FIRST IOT BOTNET

In 2013, Proofpoint researchers discovered what is now considered “the first IoT botnet”. Over 25% of the botnet was composed of non-computer devices such as smart TVs, home appliances, and baby monitors. Since then malware like CrashOverride, VPNFilter, and Triton, have been used extensively to compromise industrial IoT systems.

COMPROMISING A JEEP

In 2015, two security researchers hacked into a Jeep vehicle wirelessly, via the Chrysler Uconnect system deployed in the car, and performed remote actions like changing channels on the radio, turning on the wipers and air conditioner. The researchers said they could disable the breaks, and cause the engine to stall, slow down, or shut down altogether.

MIRAI BOTNET

In 2016 Mirai, one of the largest IoT botnets ever discovered, began its activity by attacking the websites of security researcher Brian Krebs and a European hosting company, OVH. The attacks were of a huge magnitude—630 Gbps and 1.1 Tbps. Afterwards, the botnet was used to attack Dyn, a large DNS provider, and high profile websites including Twitter, Amazon, Netflix, and the New York Times. The attackers built their network from IoT devices like routers and IP surveillance cameras.

ST. JUDE CARDIAC DEVICE VULNERABILITY

In 2017, the Food and Drug Administration (FDA) announced that implantable cardiac devices manufactured by St. Jude Medical, including pacemakers implanted in living patients, were vulnerable to attack. Billy Rios and Jonathan Butts, security researchers presenting at the Black Hat Conference, proved their ability to hack into a pacemaker and shut it down, which if it were done by hackers, would kill the patient.

IoT Security Best Practices

As you start to consider an IoT security strategy for your organization, here are a few best practices that can improve your security posture.

USE IOT SECURITY ANALYTICS

A security analytics infrastructure can significantly reduce vulnerabilities and security issues related to the Internet of Things. This requires collecting, compiling, and analyzing data from multiple IoT sources, combining it with threat intelligence, and sending it to the security operations center(SOC).

When IoT data is combined with data from other security systems, security teams have a much better chance of identifying and responding to potential threats. Security analytics systems can correlate data sources and identify anomalies that might represent suspicious behavior. Security teams can then investigate and respond to anomalies, preventing attackers from compromising corporate IoT devices.

NETWORK SEGMENTATION

Network segmentation is a technique that enables isolation of specific components from others, to improve security. In the case of IoT, segmentation can help prevent attackers or malicious insiders from connecting to IoT devices, or can prevent compromised devices from infecting other parts of the network. You can implement this technique into your strategies or use a network security solution.

To begin a segmentation effort, create a comprehensive list of IoT devices currently in use, their connection methods (VLAN or LAN), how and what type of data they transmit, and which other devices on the network each device really needs to connect to. In particular, check if each category of device needs to have access to the Internet, and if not, disable it.

One suggestion for segmentation is to designate specific categories of devices, such as data collection, infrastructure, or personal employee-owned devices. You can create a segmentation strategy based on the connectivity requirements of each IoT endpoint, and act to isolate or block network access to endpoints that don’t really need it.

ENABLE DEVICE AUTHENTICATION

Another way to reduce the vulnerability of IoT devices to attacks by enforcing full authentication on all devices. Whether your IoT devices have simple password authentication, or more advanced measures like digital certificates, biometric, or multi factor authentication (MFA), use the most secure authentication available on the device and ensure you never use the factory default password.

AI AND ML FOR IOT SECURITY

An expanding network of IoT devices produces tremendous amounts of data, which are useless without proper analysis. Massive sets of data are analyzed with the help of Artificial intelligence (AI) and machine learning, allowing machines to teach themselves, retain what they learnt, and hence improve the capabilities of IoT systems.

Being the one of the recent IoT trends, AI-based Intrusion Detection Systems (IDS) continuously monitor the network, collecting and analyzing information from previous attacks. They can predict an attack based on the historical data and suggest a solution to fight the threat. Even new hacking techniques are made up, they still may include previously used patterns, which can be recognized with ML algorithms in real-time.

In general, there are two types of ML-based IDS.

Anomaly IDS detects attacks based on recorded normal behavior, comparing the current real time traffics with previous recorded normal real time traffics. These systems are capable of detecting a new type of attack, so widely used even in spite of a large amount of false positive alarms.

Misuse or signature IDS compares the similarity between the patterns recognized in the current real time traffics and the already known patterns of various types of previous attacks. It shows a smaller amount of false positive alarms, but at the same time, the new type of attack can pass-through undetected.

ML algorithms like Linear Discriminant Analysis (LDA), Classification and Regression Trees (CART) and Random Forest can be used for attacks identification and classification.

MobiDev offers innovative IoT development services to ensure IoT ecosystem security, using artificial intelligence and machine learning to combat and mitigate IoT cyber threats.

Artificial Intelligence in Marketing: Boost the Growth in 2022

OleksiiTsymbal — Fri, 21 Jan 2022 15:04:37 +0000

Industry leaders around the world are using artificial intelligence to enhance their business with marketing technology. Whether it’s analyzing consumer interests and data, guiding sales decisions and social media campaigns or other applications, artificial intelligence is changing the way we understand marketing in many industries. Let’s talk about the latest ways that businesses can utilize these powerful tools to achieve their marketing goals.

Photo by Francesco Gallarotti on Unsplash

Core Concepts of AI in Marketing

Technology changes every day. A lot can change over several years, especially in trending artificial intelligence technologies. The same goes for AI in marketing applications. Understanding the basic ideas behind applications of AI in marketing solutions can generate unique ideas that can break new ground in various industries.

AI can help automate projects to make businesses more efficient. According to Accenture, productivity of businesses can be improved by 40% when utilizing AI. This not only can save time and money but can enable your company to focus their efforts on providing quality experiences for customers rather than spending too much time moving things from one spreadsheet to another.

AI can also help minimize errors in marketing processes. Artificial intelligence can complete specialized tasks with greater efficiency than humans can so long as supervision and guidance is involved. Often in cases where AI fails to provide the right results, human error was involved in setting up the AI program with appropriate data or it was used in a way that was not intended.

Because AI can dramatically speed up the process of marketing campaigns, reduce costs, and improve efficiency, artificial intelligence is much more likely to result in an increased return on investment (ROI).

If you have a unique idea for how AI can help your business’s marketing strategy, we encourage you to contact our AI consulting team to start a conversation.

Conducting Market Research with Artificial Intelligence

Artificial intelligence is a strong tool when used alongside high quality data. Many companies have had positive results in the real world when combining their market research data with artificial intelligence. This enables them to do all sorts of things. A big part of this trending use case is target group segmentation. AI is far quicker and more efficient at performing this task than humans are.

By investigating their target audiences more deeply, businesses can make more personalized offers to them that they are more likely to accept.

When we examine how this looks up close, we can get a better understanding of how it works. A nationwide department store can take a look at the data they’ve collected on their customers and narrow down their search to those interested in food. Using artificial intelligence, we can identify customers that have a strong preference for organic foods. By quickly using AI to analyze the habits and preferences of these consumers, campaigns can be tailored toward them with greater efficiency to improve sales.

AI in Predictive Marketing Analytics and Personalization

Target group segmentation is one of the keystone elements of personalizing a marketing campaign, but there are many other ways that artificial intelligence can help businesses personalize experiences for their audiences and customers. According to Salesforce, 76% of customers want businesses to have a clear understanding of their personal expectations.

One way that businesses do this with AI is to use predictive marketing analytics. By having AI analyze data of past events, it can reasonably and accurately infer how performance will look in the future based on a variety of factors. More importantly, analyzing what users like most can be useful when looking to suggest products to them.

For example, Amazon is the champion of this strategy. When browsing on their site, Amazon’s artificial intelligence knows about what you have bought in the past. Based on this, it can suggest products to you in your feed. It also knows what other users like you are interested in, meaning that they can provide suggestions based on that activity. This results in very personalized suggestions that can lead to higher conversions.

Spotify also takes advantage of this to make more effective music suggestions to you. It also uses this data to invest in artists to create new music that will be generally liked by a wider audience on a broader scale.

However, most personalization methods with AI tend to start from the top down and personalize to the individual instead of an entire group. The more that the system can understand the individual user, the more likely that conversions can be made. Every user has variations that differentiate them from the larger group, so no group marketing campaign will ever be as effective as a campaign that targets specific individuals and their own interests.

The ability to use artificial intelligence to predict the success of marketing campaigns and to better personalize experiences for users is a powerful technological trend that will continue for years to come. Adaptation to include this tool in your arsenal is critical for relevancy at scale.

MobiDev AI engineers developed a recommendation system that offers relevant accompanying goods that customers tend to buy together with any selected product. Read more in the case study.

Dynamic Pricing & Demand Forecasting

One of the most difficult challenges of the onset of the 2020 COVID-19 pandemic was a surge in sales of various products by stockpilers. Shortages of toilet paper became a notorious meme on the Internet as stores struggled to maintain stock in the face of the buying panic. Eventually stock would be controlled by buying limitations. However, there was an important lesson to be learned here: demand forecasting and dynamic pricing could have prevented a great deal of this struggle.

Earlier we established that artificial intelligence is a powerful tool for analyzing past data in order to predict future activity. The same principle can be applied here. It’s possible that AI can be used to analyze consumer interests, world events, and other sources to determine if there will be a rise in demand for certain products.

Using the pandemic as an example, BlueDot is a program that already can analyze the likelihood of a disease spreading across the world. If worldwide or nationwide emergencies can be predicted in this manner, stores can automatically begin ordering more products like toilet paper, medicine, and more. Not only can this help maintain stock and improve sales for stores, but it can also help the public better manage the disaster and lead to a swifter recovery.

This can also be used to dynamically and automatically raise prices. This can be used to better control stock during times of high demand and panic buying, naturally dissuading customers from bulk buying beyond reasonable amounts, as well as optimizing revenue for your business.

Dynamic pricing and demand forecasting for every business is unique. From the types of items that you carry to the types of consumers that you are serving, a custom solution made by your team or by an external vendor may be the best options for creating a system that can accomplish your goals.

Read more about machine learning approach to sales forecasting system development.

AI in Content Creation

Providing unique and engaging content can be challenging. While AI can automatically generate content, it often can be more trouble than it’s worth.

Although this technology is improving and can be very effective in some contexts, a more widely accessible and reliable possibility is for AI to offer intelligent suggestions to human writers. AI-guided suggestions for writers forms the basis of features in applications like Grammarly, Microsoft Editor, Google Docs, Microsoft Word, Yoast, SEMRush, and more.

Adobe Premiere Pro uses AI for a variety of purposes, such as automatically matching colors and managing sound mixing against voiceovers. What’s great about content creation is that humans can create unique and interesting content that AI cannot, but AI can help us augment our talents to improve the quality of the final product.

AI can also help us with image generation and manipulation tasks:

Creating copyright-free images of non-existent people, animals, and objects
Turning photos into sketches and painting of various styles
Inpainting and restoring removed parts of images
Upscaling images
Face frontal view and new human poses generation

All this can be done with the help of generative adversarial networks (GANs) that learn the structure of the complex real-world data examples and generate similar synthetic examples.

Read more about the GAN technology and its business use cases.

Does it mean that robots can replace designers? Absolutely not. The power of AI in design is mostly about optimization and speed. Designers armed with AI tools can work faster and more effectively.

Language Optimization for Email Marketing

One particular avenue of AI in content creation comes from its role in marketing campaigns via email. eBay is a particularly good example of AI email marketing, utilizing a third party service called Phrasee and natural language processing to improve email open rates by 15.8% and improve clicks by 31.2%.

This technology is used to optimize the subject text and headline copy automatically to find the most effective variation to use with eBay’s audience. The AI-generated portions of the emails are attentive to tone of voice to maximize their success.

Aside from the fact that natural language processing is improving as years go by, AI-based email marketing at its most basic can be automated with a series of A/B testing. However, the more demographic data and natural language processing that can be incorporated into the project, the better the results. Advanced artificial intelligence algorithms can improve the dynamic optimization of email marketing greatly, as seen in eBay’s case.

Future of AI in Marketing

Ultimately, the future of AI’s role in marketing technologies will be determined by imagination and innovation. Combining different technologies together can result in businesses outcompeting other leading players in the market for years. At the bare minimum, understanding what’s already in use is important for bringing your company up to speed to remain relevant and competitive in the market.

The first step toward achieving your objectives is to start a conversation. At MobiDev, our AI engineers have been helping businesses internationally for years to accomplish their goals. If you’re ready to take your business to the next level, contact us today.

7 Retail Technology Trends Reviving the Stores in 2022

OleksiiTsymbal — Thu, 13 Jan 2022 15:38:57 +0000

Businesses around the world were rocked two years ago by the beginning of the COVID-19 pandemic. However, with disruption comes great innovation in both physical and digital storefronts. It’s important to consider how technological advancements in the retail industry can affect large enterprises and smaller businesses alike.

Trend #1: PoS Systems and Ecommerce

In addition to selling products in brick and mortar stores, many storefronts are now offering buy online pick up in store (BOPIS) services, same day delivery, and online shipping. With so many new ways to order goods, point of sale systems need to be updated in an intelligent manner. When a guest orders a product online, how does that affect the store’s sales floor quantity in the database? If a product is on hold for a pick up order, does that reduce the count on the sales floor? These are all questions a modern POS system must be able to answer.

Not all businesses were able to survive the shift in emphasis to digital retail options. Microsoft’s physical stores were shut down around the country. However, those who were able to adapt to these changes, like Walmart and Target, were able to thrive.

The key to this is more than just having systems in place to handle both online and in-store purchases, but a unified and connected system that integrates with other technologies the business uses. Integrating online and offline transactions, inventory, and promotions across all in-person locations and online stores are some of the greatest hurdles that software engineers seek to overcome with developing PoS systems for retail businesses.

Trend #2: Indoor Positioning — Not Just For Store Maps

Indoor positioning systems (IPS) have been a trend of interest for a few years now for helping users navigate indoor areas where GPS is not available or not accurate enough. However, there are many more applications for this technology than first meet the eye.

On the surface, indoor navigation does have a great deal of potential for user navigation that has been realized in stores, office buildings, airports, and hospitals. Hong Kong International Airport uses this technology in their “HKG My Flight” mobile application. In the retail industry, Lowes and Target are utilizing this technology as well.

IMPLEMENTING IPS SYSTEMS IN STORES

The challenge comes from implementing in-store navigation solutions for your business. The first step is hardware. Having the right infrastructure in place is essential, whether it be Bluetooth beacons, Wi-Fi RTT, or ultra wideband (UWB). Target chose to implement Bluetooth IoT lighting systems at many of its locations. When guests shop with the Target app on their phone in stores, they can get access to a map that helps them find their position in the store and find the items that they need.

This particular solution placed the Bluetooth beacons in the lighting systems above the sales floor. However, other solutions may be possible based on the layout and size of the store. After the infrastructure is set up, all that’s left is to develop the software that can take advantage of it.

IPS systems have much more potential than just helping guests shop. Having data on where guests shop in the store has three beneficial implications: targeted suggestions, tracking customer traffic, and item tracking.

TARGETED DEALS

When guests are in a particular part of the store using the store’s official app, the IPS system can track where they are and offer them targeted suggestions on products and deals on their smartphone. This can improve conversions as guests shop.

CUSTOMER TRAFFIC DATA

Perhaps more profoundly, data on where shoppers are in the store can lend itself useful for improving the placement of products on shelves. Where do guests go when they enter the store? Do they tend to prefer one location in the store over another? Are there any locations in the store that guests tend to avoid? These are all questions that IPS tracking can answer. Of course, anonymized data is important to protect user privacy, but these analytics could be crucial for helping business owners make decisions on where to place certain products.

In fact, this data could be a valuable bargaining chip when negotiating with vendors on how much they should pay to have their product included in a particular high-traffic area of the store.

ITEM TRACKING

RFID technology used for tracking down products in retail locations is useful but relatively short range. As these technologies improve, IPS can offer more detailed data on the locations of products on the sales floor. This can help not only to improve the efforts of assets protection but it can also make it easier to return moved items to their correct locations.

IoT devices are becoming smaller and cheaper as time goes on, but for right now this technology may need to be limited to a few high-value items if it’s to be implemented. It can also be used to help track work equipment like personal devices.

For tasks like order fulfillment, retail workers can use their work devices to find locations in the store using IPS maps to find the item’s location. If the item has IPS tracking set up, they could navigate to the item instead. This would be useful in instances where the item has been moved from its original location by mistake.

Trend #3: Augmented Reality for Consumers and Workers

Augmented reality shopping experiences are becoming the new norm in 2022. With virtual fitting rooms, enhanced in-store AR navigation, and other AR experiences spreading across the market, those without these features are falling behind. This is one of the key technological advancements bridging the gap between digital ecommerce storefronts and brick and mortar shopping.

TRY BEFORE YOU BUY

Some of the most useful applications of VR that help digital customers in the retail industry are the ‘try before you buy’ applications. One of the best examples of this is virtual fitting room technology. By using AR technologies, shoppers can see how they look when trying on different kinds of products on themselves. Similarly, they can also see what other kinds of products look like, such as furniture. Virtual fitting room technology is already being implemented by stores like Sephora, Target, Ikea, and more.

In fact, Ikea’s version of this lets you do much more than just see the size and shape of furniture. Ikea Studio can use the unique LiDAR technology on iPhones to display rooms, measurements, windows, doorways, and more to visualize interior design like never before.

AR NAVIGATION

Although indoor positioning was already a trend we discussed, when combined with AR it can enrich the experiences of customers and workers even further. Augmented reality-based indoor navigation technology can help shoppers find the items that they need in store just by following directions on their phone’s screen. This can be very engaging for shoppers if implemented correctly and can provide for more opportunities to show guests targeted suggestions on-screen.

AR navigation can help retail workers too. Some of the most popular devices for workers in the industry, Zebra’s TC52, TC57, and TC77 portable computers are ARCore-capable. With the potential of these devices to use their cameras for AR navigation, order fulfillment workers can simply look for directions on their screen to find the aisle that they’re looking for.

AIDING RETAIL WORKERS IN OTHER TASKS

Augmented reality developers can help retail workers visualize shelves before they are set up. By using a planogram as a base, AR-enabled enterprise devices can display a 3D planogram on the sales floor to serve as a guide for the setup process. This can make it easier for workers to set pegs, shelves, and fixtures in their appropriate locations.

Augmented reality could also help workers identify problems with a shelf. Combining AR with AI-based object detection technology, a worker could hold their device’s camera up to a shelf to find areas for improvement. By comparing the shelf with its corresponding planogram, AR rendering can show the worker which parts of the shelf do not align with the planogram and need to be fixed. This could be out of place product, improperly zoned items, and even identification of damaged products.

Trend #4: Artificial Intelligence in the Retail Industry

In nearly every one of the previous trends that have been discussed so far, artificial intelligence has played a role in some fashion. Modern AR is powered by AI scene analysis. However, AI can serve much more benefit in the retail industry than this. AI can be used in inventory management and other kinds of analysis on consumer behavior.

INVENTORY MANAGEMENT POWERED BY AI

The retail industry is driven by computers and rich databases more than ever before in 2022. However, the complexity of these systems has resulted in a great deal of inefficiencies from errors. Retail workers may find that the system has incorrect sales floor quantities of certain SKUs. Tackling these issues is critical to saving time and money for businesses.

Artificial intelligence can help us manage these inventories more effectively with a number of different methods. For example, AI can measure consumer spending analytics to predict when certain types of items may be more likely to move around on the sales floor. This can drive the system to require manual auditing of certain areas over others. While auditing the entire store is a challenging task, auditing only the areas that need attention is much more effective.

Amazon takes it a step further with its Amazon Go Grocery model stores powered by Just Walk Out. The project uses computer vision, sensor fusion, and deep learning. The store’s rich network of cameras and IoT sensors can detect when a consumer has taken an item off the shelf and placed it into their cart. When the guest leaves the store, their credit card is charged for the items that they took.

This requires not only a deep understanding of what the guest has placed into their cart but also where items are in the store. If an item moves away from its shelf, the store must still be able to keep track of how much of it that it has on hand.

If a more advanced network of cameras isn’t the solution, perhaps it can be done manually while still using machine learning algorithms. Instead of auditing empty spots on a shelf one by one, a retail worker could take a photo of each shelf section with their device. The image could be compared to a planogram with object recognition deep learning software to see which items are missing, speeding up the process greatly.

CONSUMER BEHAVIOR ANALYSIS

Earlier we touched on this with analysis of in-store foot traffic with indoor positioning systems. However, the potential of analyzing consumer behavior doesn’t stop there. In addition to their spending habits, we can measure their interest in what they’re spending money on alongside their demographics with artificial intelligence. This can give businesses a better idea of how they should be marketing to their audiences.

The focus so far has been analyzing consumers that are already in the store, but what about drawing in new ones? A store using Kimola’s platform decided to remodel a store for soccer fans using data analyzed by AI. By using social media accounts as a base for the data, the algorithm was able to determine the most optimal location for the store.

This idea extends to e-commerce as well. A user’s behavior can be tracked while they visit an online storefront. What items their cursor hovers over, how much time they spend on a particular page, as well as where they come from. This type of information is familiar to any web marketer using Google Analytics.

However, behind that technology is powerful artificial intelligence that can search for connections and trends faster than we can on our own.

Trend #5: AI-Driven Demand Forecasting

This is a powerful tool used by some of the largest brands in the world. A few years ago it became apparent that many businesses weren’t ready for the sudden panic buying that resulted from the beginning of the COVID-19 pandemic. As products flew off the shelves, stores that were able to adapt and get their hands on high-demand items were able to thrive.

To evolve to serve the needs of the public during the pandemic, demand forecasting powered by machine learning took off. Amazon is again another example of a company that leverages the power of machine learning for this purpose. This technology can also improve inventory planning, relationship management for both customers and suppliers, logistics, manufacturing, and marketing.

Demand forecasting can also lead to more sustainable consumption and production. When demand is predicted much more accurately, items can be produced and ordered only according to how much is needed by the consumer.

ML-based demand forecasting approaches are much more versatile and adaptive than their traditional counterparts. Since machine learning can be implemented much more quickly, it can better follow customer demand trends.

Read the case study to see an example of how machine learning-based sales prediction feature was implemented in the ERP system.

Two years into the pandemic, demand forecasting technology continues to thrive. With ever shifting customer behaviors, it’s more important than ever that your business is deeply interested in data on customer spending trends. Technologies to analyze that data to get meaningful insights are just as important for maintaining a competitive foothold in the market.

Trend #6: Robots and Automation in Retail

Closely related to artificial intelligence is the hardware of robotics. This has a multitude of applications, such as delivery, inventory management, and customer service.

AUTONOMOUS DELIVERY

As artificial intelligence improves, so too have autonomous vehicles. Delivery is evolving in 2022 and autonomous delivery is becoming the new norm. The Safeway cart, developed by Tortoise, was introduced last year as an autonomous delivery vehicle. Serve Robotics, formerly called Postmates X, is also making a delivery robot for Uber. With Grubhub successfully rolling out autonomous food delivery robots to college campuses like the Ohio State University, their potential cannot be understated.

Drone retail delivery is also being explored. Verizon and UPS Flight Forward announced last year that they were working to leverage 5G technologies to improve drone delivery in Florida.

CUSTOMER SERVICE ROBOTICS

Customer service robotics have also been explored. In January of last year, Hyundai introduced a robot called DAL-e that would aid in automotive showrooms. The robot can greet customers and help them find the vehicle that meets their needs. The robot also uses facial recognition and AI to detect if the customer is wearing a mask and advise them appropriately about masking rules.

ROBOTIC INVENTORY MANAGEMENT

Special hardware can sometimes aid with inventory management where ceiling and shelf-mounted cameras cannot. Machines like SmartSight can automate the process of identifying misplaced items on shelves and sales floor quantities and alert workers when certain items are running low.

Trend #7: Voice Commerce

As artificial intelligence improves in 2022, so does natural language processing (NLP). Smart assistants like Google Assistant, Alexa, Siri, and Bixby are becoming more and more advanced in their voice recognition and responsiveness. Their ability to serve customers in the retail industry has improved tremendously. There is a great deal of nuance to be explored with the process of ordering products from home with your voice with or without a screen to aid you visually.

Smart home devices are expected to be used in more than 77 million homes in the United States by 2025, meaning that this is a huge opportunity that businesses can take advantage of. Other kinds of IoT devices in customers’ homes can prove valuable too, such as AR-capable smart mirrors. The most important part of the whole process is making sure that these devices are not intrusive and are genuinely helpful for the consumer.

Walmart’s Voice Ordering is a great example of this technology in action. By asking smart speakers to add products to a cart, they can place orders entirely by voice. Those items can then be made available for pickup by Walmart’s workers.

Future of the Retail Industry: Embracing, but Managing, Technology

As technological advancements thrive in the retail industry, legacy technologies like older PoS systems, inventory management solutions, etc are being replaced. However, during these transitions many hiccups tend to bog down the process. There are sometimes oversights and unforeseen issues that require special attention. More importantly, it is critical to manage these new systems to ensure that they are suiting the business correctly.

Problems can be seen as opportunities, and 2022’s challenges for the retail industry are no different. Companies must adapt and use retail technology innovations to succeed in the market.
As technology advances, there are many new niches waiting to be monetized and opportunities to be taken. Digital transformation and data-centric cultures will continue to disrupt businesses that cannot adapt to rapidly changing customer preferences, new technologies, and supply chains. We must bring consumers experiences that they find engaging, accessible, and valuable.
Alex Vasilchenko
AI Engineer

The most important part about introducing new technologies into the retail space is the relationship between the business and the vendor developing the technology. How they work together to develop and maintain the solution determines the success of the product, no matter if the technology is intended for the consumer directly or if it’s intended to aid the business’s own tasks first and foremost.

Healthcare Technology Trends and Digital Innovations in 2022

OleksiiTsymbal — Tue, 11 Jan 2022 14:29:12 +0000

With 2021 well behind us, COVID-19’s presence still continues to linger around the world. Of all the industries that have been forever changed by the impacts of the global pandemic, healthcare arguably has changed the most. With significant advances in technology and methods necessary to support the high demand of access to healthcare and growing digitization of protected health information, the healthcare industry has evolved in novel ways to continue to deliver the same exceptional quality of service.

As we press on into the future, it’s critical to remain mindful of the trends driving healthcare technology in 2022. Although legacy software and infrastructure is critical to the success of modern hospitals and care centers, it’s important that we consider how those systems can integrate with newer technologies or how they may eventually be replaced with more reliable systems. The focus should be on improving performance, productivity, efficiency, and security without sacrificing reliability or accessibility.

If you’re ready to explore the tech innovations driving the healthcare industry towards digital transformation this year, let’s take a look at the most important technologies that have the potential to transform your organization.

Trend #1: Artificial Intelligence (AI) in Healthcare

Across multiple industries, artificial intelligence has made great waves as a useful technology in 2022, especially for healthcare.

AI’S ROLE AGAINST COVID-19

The global pandemic has had a profound impact on our society, but cutting-edge technology has helped us stay ahead of the curve. In fact, a company from Toronto, Canada was able to predict that COVID-19 would spread worldwide. By scanning 100,000 media sources in over 65 different languages daily, their application called BlueDot can ascertain dangerous outbreaks in what is nearly real time.

A great deal of progress was also made with machine learning advances in the development of vaccines. By using machine learning to help identify protein fragments, the vaccines for COVID-19 have been developed in much shorter periods of time than ever before.

AI has also been helpful in analyzing crowd temperature data. This makes thermal screening a much more viable option for identifying potentially symptomatic individuals. Advances in AI-powered facial recognition have also made it viable toward identifying individuals even if they are wearing a face mask. It can also detect if the user is wearing a mask in the areas where it’s mandatory.

AI IN DIAGNOSIS & DRUG DEVELOPMENT

Artificial intelligence has plenty of applications outside of treating and responding to the pandemic. AI is incredibly helpful for improving efficiency with information processing and decision making. In the healthcare industry, machine learning is extremely helpful for the development of new pharmaceuticals and the efficiency of diagnosis processes.

For those being treated for the effects of COVID-19, AI is helping analyze CT scans to detect pneumonia. Microsoft developed Project InnerEye, a radiotherapy AI tool. This dramatically speeds up the process of 3D contouring of the patient, bringing time to completion down to minutes instead of hours. The project is open source on GitHub. Project Hanover is another Microsoft AI system meant to catalog biomedical research papers from PubMed. This helps reduce time for cancer diagnosis and assists with deciding on which drugs should be used for each patient.

AI IN MENTAL HEALTH

Artificial intelligence innovations don’t just apply to physical health. MIT and Harvard University researchers have utilized machine learning to track trends and mental health in correlation to the COVID-19 pandemic. By using an AI model, they were able to analyze thousands of online Reddit messages to find that topics of suicidality and loneliness had nearly doubled over a period of time. This has the potential to transform our understanding of the mental health of larger populations.

NATURAL LANGUAGE PROCESSING

Chatbots have the potential to improve the efficiency of telehealth. Researchers at UCLA combined chatbot technologies with AI systems to create a Virtual Interventional Radiologist (VIR). This was intended to help patients self-diagnose themselves and for assisting doctors in diagnosing those patients. Chatbots powered by Natural Language Processing aren’t ready to provide primary diagnosis, but they can be used to assist in the process. They are also well equipped to help obtain information from patients before proper treatment can begin.

THE KEY TO AI IN HEALTHCARE: DATA

The most important element that powers artificial intelligence’s success in healthcare is data. More specifically, training data. Software powered by machine learning will never outperform the quality of its training dataset. The higher the quality and breadth of the data we give to the model, the better it will perform. It’s essential that your AI team is composed of experienced software developers and data scientists that can work together to produce the best results.

Trend #2: Telemedicine and the Evolution of Remote Care

Telehealth has come a long way since the beginning of the pandemic in 2020. In 2022, care providers regularly have video conference meetings with patients over the Internet to discuss concerns and give advice. The infrastructure to support this has vastly improved. Telehealth is expected to grow to $185.6 billion by 2026. With that in mind, what lies in store for the future?

COMPLYING WITH REGULATIONS

Before we consider that possibility, one of the most important issues with telehealth is HIPAA compliance. Although some restrictions were relaxed during the height of the pandemic in 2020, it’s important for care providers to consider the applications that they are using to communicate with their patients. Are they secure and certified to handle private health information?

WEBRTC FOR VIDEO CONFERENCING

In many instances, a more dedicated solution is required that can more specifically adhere to legal privacy requirements. If you need a dedicated telemedicine app, one of the most important technologies that will be needed is WebRTC, an open-source API-based system that connects web browsers and mobile applications and allows for transmitting audio, video, and data. This is especially helpful for teleconferencing features.

CLOUD HOSTING AND DATA STORAGE

Storing data in most cloud storage services is relatively secure, but not necessarily compliant with government regulations on protected health information. HIPAA compliant cloud hosting solutions are critical for maintaining functionality and efficiency for any healthcare operation needing electronic health records (EVR).

However, teleconferencing and data hosting aren’t the only features that may be useful for your organization. Other features like security, location services, appointment management, secure messaging, healthcare provider reviews, visit history, and wearable integration are all potentially useful features.

Some applications may need to store fitness data from consumer devices such as Google Fit and Apple HealthKit. Being able to maintain these integrations in a secure and efficient manner can benefit the patient and caregiver greatly.

Trend #3: Extended Reality in Healthcare Settings

Extended reality, a blanket term including augmented reality, virtual reality, and mixed reality, has a great deal of potential in the healthcare industry. From assisting surgery to aiding telehealth applications, AR and VR technologies can improve the healthcare industry substantially.

AUGMENTED REALITY AND MIXED REALITY IN HEALTHCARE

Augmented reality and mixed reality are useful in a variety of healthcare settings. One of the most popular and useful forms of this technology is the use of mixed reality headsets like Microsoft Hololens 2 by surgeons. The headset can provide heads up information to the surgeon while allowing them to use both of their hands during the procedure.

Not only can these surgeries be enriched by this heads up information, but it can be a collaborative and remote effort and assist in training purposes. The head-mounted camera view of the headset can enable other doctors to observe the surgery and offer advice. The ‘holographic’ nature of the device can be used to enrich training as well. Similar applications are possible with on the rise AR headsets. More specialized software solutions will also be necessary to expand its use to different types of surgeries into the future.

AR isn’t just restricted to headsets and operating rooms. The technology can also be used to help nurses find veins to draw blood from.

Augmented reality development relies heavily upon artificial intelligence and specialized sensors to function. Whether you’re developing for mobile devices or other kinds of hardware, appropriate data and software expertise will be required. AR developers focus heavily upon leveraging AI with the software frameworks of target hardware to make these products successful.

METAVERSE: FUTURE OR HYPE?

There is a big debate about whether Facebook’s rebrand to Meta and the focus on social virtual reality experiences is justified. Ultimately whether or not this is something that you are willing to invest in is up to you. Even if the metaverse is vastly overblown, there is some potential for virtual reality in healthcare settings.

One of the most useful applications of VR in healthcare that is in use right now is training. Creating virtual training situations for doctors can help them improve their skills and prepare for operations. VR can also be used in some contexts for treatment. For example, The Virtual Reality Medical Center uses VR therapy to help those suffering from phobias like fear of heights and PTSD.

Maplewood Senior Living in Connecticut also has a VR therapy program for the elderly that can help them unlock past memories and improve emotional wellbeing.

Meta’s shift toward cartoon-style meetings might become useful for VR therapy, but the effectiveness of this in place of traditional teleconferencing remains to be seen. However, some technologies in this space such as spatial audio have potential to improve the effectiveness of telehealth systems by providing a more immersive digital experience.

Trend #4: IoT and Wearables in Healthcare

With wearables and IoT technologies becoming more popular, their potential in the healthcare industry has grown significantly. For applications in telemedicine and telehealth technologies many have come to call this trend in micro processing the Internet of Medical Things.

There were 11.3 billion IoT devices connected at the start of 2021. The global IoT medical devices market is projected to reach USD 94.2 billion by 2026 from USD 26.5 billion in 2021. With the healthcare industry becoming increasingly more connected through these technologies, IoT cannot be ignored.

WEARABLES

One of the most important innovations in the healthcare industry is the advancement of wearable technology. The ability to monitor the status of a patient throughout the day remotely or for an individual to monitor their own status is incredibly valuable. A survey conducted by Deloitte found that 39% of users had a smartwatch. With consumer smartwatches becoming more widely available, their potential to be used for healthcare applications should be noted.

One of the most basic things that a smartwatch can provide that can be useful for monitoring a person’s health is heart rate. However, this is not the only thing that a smartwatch can measure. These devices can also monitor physical health with pedometers and blood oxygen saturation. Low blood oxygen saturation is difficult to detect without specialized sensors. Since this can be a life-threatening condition, smartwatches with this sensor can save lives.

Smartwatches are also improving in their ability to measure blood vitals in their users. Photoplethysmography (PPG) is an optical technology that can measure variations in blood volume and composition. Since it has been miniaturized for use on smartwatches, it can provide users with more data than ever about their blood vitals. Healthcare providers can use this data to help advise patients and complete diagnoses.

Smartwatches aren’t the only wearable that has potential for the healthcare industry. Bio patches and smart hearing aids have similar levels of impact. Bio patches can provide a better understanding of a person’s vitals without the use of a smartwatch. Artificial intelligence can also be used to improve noise isolation of hearing aids.

SMART PILLS

One of the most profound applications for IoT technology in healthcare is the concept of a smart pill, which transforms The Internet of Things into The Internet of Bodies. Smart pills are edible electronics that not only serve as pharmaceuticals, but can provide care providers with valuable information about patients. The first smart pill approved by the FDA was released in 2017.

CREATING IOMT SOLUTIONS

With the industry trending toward using numerous microcontrollers in unison, getting all of these computers to communicate with each other can be challenging. Another obstacle that needs to be overcome is that almost every manufacturer uses their own proprietary protocol to get their devices to talk to each other. This can make integration difficult.

Connectivity can also be an issue, as many environmental factors can disrupt communications. In order to overcome this, buffering methods on local microcontrollers need to become more robust. Security is always a concern as well.

Trend #5: Healthcare Privacy and Security in 2022

Beyond the scope of efficiency and quality of care, privacy and security take critical priority in the healthcare industry. In February of 2021, Kroger pharmacy data was exposed in a data breach through Accellion’s FTA secure file-transfer service. They’re not the only ones. According to HealthITSecurity.com, over 550 organizations suffered data breaches last year that affected over 40 million people.

Ensuring your organization is HIPAA compliant is an essential first step toward avoiding costly data breaches. If you are serving patients internationally, it may be a good idea to consider the regulations of the General Data Protection Regulation (GDPR) in the European Union.

The HIPAA regulation covers a great range of activities and it even covers some that are not mentioned in the act directly. Electronic signatures that are used in online forms quite often are a great example of that. They are not mentioned in the regulation but covered entities have to ensure that they are using HIPAA compliant e-signature services as these services will store data that is considered PHI for authorization and authentication purposes.
Erman Ergun, Healthcare Content Manager, JotForm

Some healthcare providers are able to use software like Facetime and Skype to communicate with patients that may not be fully compliant with government regulations. Although the Office for Civil Rights (OCR) of the Department of Health and Human Services (HHS) in the United States has previously announced relaxed enforcement policies for organizations during the public health emergency, it’s important not to rely on that discretion. Eventually as the pandemic continues to subside, enforcement will go back to normal. Getting ahead of the curve can help avoid steep fines in the future.

Although compliant video conferencing software already exists, sometimes a more customized solution needs to be created. This is especially the case if existing data infrastructure does not integrate well with available options. More importantly, if a healthcare provider wants to use an existing system to exchange ePHI with patients through third party software, they will have to obtain a business associate exception with the vendor which can be tedious and difficult.

There still is no guarantee that the third-party program can fully protect patient data. In addition, keeping information secure through remote doctor calls is difficult. ePHI data must be transmitted in structured formats, and these calls can make the process complicated.

Trend #6: Organ Care Technology & Bioprinting

With the world’s transplantation market size predicted to reach $26.5 billion by 2028, organ transplants are certainly an important part of the healthcare industry. According to Matthew J Everly, about 2,000 heart transplants take place in the United States every year. However, it’s estimated that over 50,000 people need a heart transplant. What can be done to help all of these people suffering from heart disease?

IMPROVING ORGAN CARE TECHNOLOGY: EXTENDING TIME FOR ORGAN EVALUATION AND TRANSPORT

One of the approaches to this issue is to improve organ care technology. This means taking care of the organ while it’s outside of the body. The Organ Care System developed by Transmedics is a great example which is in use by the Ohio State University’s Wexner Medical Center. This device can keep a heart, lung, or liver outside of the body for several hours through proper care, heat, and provision of important nutrients.

It’s possible that the future of this technology may depend on artificial intelligence to automatically take action without a doctor’s intervention to preserve the organ for longer periods of time.

Perhaps more importantly, machine learning may be able to better determine if an organ being preserved is suitable for transplantation or not. The faster that this can be determined, the more quickly a life could be saved.

BIOPRINTING: CREATING NEW ORGANS

In addition to keeping organs alive outside of the body, other options should also be explored. Although it may sound like science fiction, 3D printed organs are a very real, although developing, technology that has already made its way into clinical testing. Ears, corneas, bones, and skin are all organs in clinical testing for 3D bioprinting.

The process is not too different from traditional 3D printing. First a digital model of the tissue must be created. Careful attention needs to be paid to the resolution and matrix structure, as the materials used in the printing process are quite literally living cells called bioink. They then need to test the organ’s functionality with stimulation.

One of the ways that rejection of an organ can be prevented is by using the cells of the patient requiring transplantation. These cells can be grown in a culture and then cultivated into the bioink necessary for printing.

Bioprinting has been done in the past but has not yet hit the mainstream. It’s possible that through AI analysis of organs and of recipient patient characteristics, organs can be better engineered to be compatible with their new hosts.

Future of Healthcare Technology

As 2022 rolls forward, healthcare technology will continue to improve in every area. Although security will improve across the industry, threats are always evolving that must be dealt with through prevention rather than response. Quality and efficiency of care will continue to improve due to groundbreaking and evolving technologies like artificial intelligence, machine learning, and extended reality.

Artificial Intelligence Technology: AI Trends That Matter for Business

OleksiiTsymbal — Thu, 18 Nov 2021 14:31:59 +0000

According to 2020’s McKinsey Global Survey on artificial intelligence (AI), in 2020 more than 50% of companies have adopted AI in at least one business unit or function, so we witness the emergence of new AI trends. Organizations apply AI tools to generate more value, increase revenue and customer loyalty. AI leading companies invest at least 20% of their earnings before interest and taxes (EBIT) in AI. This figure may increase as COVID-19 is accelerating digitization. Lockdowns resulted in a massive surge of online activity and an intensive AI adoption in business, education, administration, social interaction, etc.

This article aims to overview new and current AI trends that emerged in 2020 and are still increasing in 2021. Based on trends, companies can make projections of the AI future in 2022 and successfully mitigate risks. Saying AI, we mean neural networks, machine learning, deep learning, computer vision, and other subfields of Artificial Intelligence.

Image credit

AI Adoption Trends

AI Adoption level differs depending on the industry. Using the data mentioned in the McKinsey Global Survey on AI, we can highlight four leading sectors: high-tech, telecom, automotive, assembly.

Companies apply AI for service operations, service or product design, advertising, and sales. Regarding investments, the area of drug discovery and development received the highest amount of money — in 2020, the total sum of investments exceeded 13.8 billion dollars, 4.5-fold higher than the year before.

AI drives the highest revenue growth if applied in inventory and parts optimization, pricing and promotion, customer-service analytics, sales, and demand forecasting. Use cases that reported cost decrease are related to optimization of talent management, contact-center automation, and warehouse automation.

AI Technology Trends

In 2021 and the following years, AI will be leveraged to simplify operations and make them more efficient. Businesses should try to benefit from the commercial application of Artificial Intelligence through improving IT infrastructure and data management. But not each deployed AI model could be useful for companies and appropriate for performance monitoring. We’ll focus on AI trends 2021-2022 that are likely to become mainstream.

Trend 1. AI for Security & Surveillance

AI techniques have already been applied to face recognition, voice identification, and video analysis. These techniques form the best combo for surveillance and biometric authentication. So, in 2021, we can foresee the intensive exploitation of AI in video surveillance.

Artificial Intelligence is beneficial for a flexible setup of security systems. Previously, engineers spent a lot of time configuring the system because it was activated when a specific number of pixels on a screen changed. So, there were too many false alarms. These alarms were caused by falling leaves or a running animal. Thanks to AI, the security system identifies objects, which contributes to a more flexible setup.

AI in video surveillance can detect suspicious activity by focusing on abnormal behavior patterns, not faces. This ability enables creating more secure spaces, both public and private, through identifying potential threats. Such AI-driven video solutions could be also useful for logistics, retail, and manufacturing.

Another niche that provides promising perspectives for the AI application is voice recognition. Technologies related to voice recognition are able to determine the identity. By identity, we mean the age of a person, gender, and emotional state. The principles on which voice recognition for surveillance is based can be the same as in the case of Alexa or Google Assistant. A feature that is suitable for security and surveillance is a built-in anti-spoofing model that detects synthesized and recorded voice.

One of the most crucial technologies for security is biometric face recognition. Different malicious applications try to trick security systems by providing fake photos instead of real images. To defend against such cases, multiple anti-spoofing techniques are presently being developed and used at large scale.

Trend 2. AI in real-time video processing

The challenge for processing of real-time video streams is handling data pipelines. Engineers aim to ensure accuracy and minimize latency of video processing. And AI solutions can help to achieve this goal.

To implement an AI-based approach in live video processing, we need a pre-trained neural network model, a cloud infrastructure, and a software layer for applying user scenarios. Processing speed is crucial for real-time streaming, so all these components should be tightly integrated. For faster processing, we can parallelize processes or improve algorithms. Processes parallelization is achieved through file splitting or using a pipeline approach. This pipeline architecture is the best choice since it doesn’t decrease a model’s accuracy and allows for use of an AI algorithm to process video in real-time without any complexities. Also, for pipeline architecture, it’s possible to apply additional effects implying face detection and blurring. You can find more information on the subject in our article dedicated to AI in real-time video processing.

Modern real-time stream processing is inextricably linked to the application of background removal and blur. The demand for these tools has increased because of COVID-19 contribution to the emergence and popularization of new trends in video conferencing. And these trends will be actively developed because, according to GlobeNewswire, the global video conferencing market is expected to grow from USD 9.2 billion in 2021 to USD 22.5 billion by 2026.

There are different ways to develop tools for background removal and blur in a real-time video. The challenge is to design a model capable of separating a person in the frame from the background. The neural network that is able to carry out such a task could be based on existing models like BodyPix, MediaPipe, or PixelLib. When the model is chosen, the challenge remains for its integration with an appropriate framework and organizing the optimal execution process through the application of WebAssembly, WebGL, or WebGPU.

Trend 3. Generative AI for content creation & chatbots

Modern AI models are able to generate text, audio, images in a very high quality, almost indistinguishable from non-synthetic real data.

At the heart of text generation stands Natural Language Processing (NLP). Rapid advances in NLP have led to the emergence of language models. For instance, BERT model is being successfully used by Google and Microsoft to complement their search engines.

How else does the development of technologies related to NLP boost companies? First of all, combining NLP and AI tools allows the creation of chatbots. According to Business Insider, the chatbot market is expected to reach USD 9.4 billion in 2024, so let’s emphasize the ways businesses benefit from AI-driven chatbots implementation.

Chatbot tries to understand the intentions of people, instead of just performing standard commands. Companies working in different areas use the AI-driven chatbot to provide their clients or users with human-level communication. Applications of chatbots are widely observed in the following business domains : healthcare, banking, marketing, travel and hospitality.

AI-driven chatbots help to automate admin tasks. For instance, in healthcare they reduce the amount of manual work. Here, chatbots help to organize appointments, send reminders related to taking meds, and provide patients with answers to queries. In other areas, chatbots are introduced to deliver targeted messages, improve customer engagement and support, and provide users with personalized offers.

Besides chatbots, NLP lies at the heart of other cutting-edge technological solutions. One of the examples is NLP text generation that can be used in business applications. An NLP-based Question Generation system presented in the video below is used in a secure authentication process.

The recent arrival of the GPT-3 model allows AI engineers to generate an average of 4.5 billion words per day. This will allow a tremendous range of downstream applications of AI for both socially useful and less useful purposes. It is also causing researchers to invest in technologies for detecting generative models. Note that in 2021-2022 we will witness the arrival of GPT-4 — “artificially generally intelligent AI”.

Coming back to Generative AI, we want to pay attention to GANs, or Generative Adversarial Networks, that are now capable of creating images indistinguishable from human-produced ones. That could be images of non-existent people, animals, objects, as well as other types of media, such as audio and text. Now is the best moment to implement GANs gaining from their abilities because they can model real data distributions and learn helpful representations for improving the AI pipelines, securing data, finding anomalies, and adapting to specific real-world cases.

Trend 4. AI-driven QA and inspection

The most remarkable branch of Computer Vision is AI inspection. In recent years, this direction has been prospering because of the increasing accuracy and performance. Companies started to invest both computational and financial resources to develop сomputer vision systems at a faster rate. The intensive development of AI inspection is also connected with a rapid progress in the domain of object detection in video frames.

Automated inspection in manufacturing implies the analysis of products in terms of their compliance with quality standards. The methodology is also applied to equipment monitoring.

Here are few use cases of AI inspection:

Detecting defects of products on the assembly line
Identifying defects of mechanical and car body parts
Baggage screening and aircraft maintenance
Inspections of nuclear power stations
Use cases of AI inspection

Trend 5. Game-changing AI breakthroughs in healthcare

The next trend related to the implementation of AI in the healthcare industry has been intensively discussed over recent years. Scientists use AI models and computer vision algorithms in the fight against COVID-19, including areas like pandemic detection, vaccine development, drug discovery, thermal screening, facial recognition with masks, and analyzing CT scans.

To counteract the spread of COVID-19, AI models can detect and analyze potential threats and make accurate predictions. Also, AI helps to develop vaccines by identifying crucial components that make them efficient.

AI-driven solutions may be applied as an efficient tool in The Internet of Medical Things and for handling confidentiality issues specific to the healthcare industry. If we systematize use cases of AI in healthcare, it becomes clear that they are united by one aim – to ensure that the patient is diagnosed quickly and accurately.

Trend 6. No-code AI platforms in at least three areas

No-code AI platforms have enabled even small companies to apply powerful technologies that were previously available only to large enterprises. Let’s find out why such platforms are a key AI trend for businesses in 2021.

Developing AI models from scratch requires time expenditure and relevant experience. Adoption of the no-code AI platform simplifies the task because it reduces the entry barrier. The advantages are:

Fast implementation — compared with writing code from scratch, working with data, and debugging, time saving reaches 90%.
The lower cost of development — through automation, the businesses eliminated the need for large data science teams.
Ease of use — drag-and-drop functionality simplifies software development and enables the creation of apps without coding.

No-code AI platforms are in demand in healthcare, financial sector, and marketing — though produced solutions couldn’t be highly customized. Among the most sought-after no-code AI platforms, you can find Google Cloud Auto ML, Google ML Kit, Runaway AI, CreateML, MakeML, SuperAnnotate, etc.

Enterprise-sized companies, as well as mid-size businesses, leverage no-code platforms for software solutions aiming at image classification, recognizing poses and sounds, and object detection.

Trend 7. Diversity in AI

The lack of diversity in AI can contribute to the emergence of racial and gender biases. By diversity, we mean a variety of people who develop AI models. According to NYU’s research, 80% of professors involved in AI development are men, and only 10% of researchers who work with Artificial Intelligence at Google are women. The same research shows that not even 5% of staff at Google, Facebook, and Microsoft are Black workers.

The number of female graduates of AI PhD programs and computer science faculties has remained at a low level for a long time. But the need for diversity in AI should influence this situation, which is one of the emerging trends. Moreover, women in AI can make big decisions influencing the development and implementation of AI systems. So, if you want to know more about females in AI, read the article dedicated to the brilliant career path of an AI engineer at MobiDev.

Evolution and Future of AI

Trends show that the future of Artificial Intelligence is promising because AI solutions are becoming commonplace. Autonomous cars, robots and sensors for predictive analysis in manufacturing, virtual assistants in healthcare, NLP for reports in media, virtual educational tutors, AI assistants, and chatbots that can replace humans in customer service — all these AI-powered solutions are moving forward with huge steps.