<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Media Sharing</title>
    <description>The latest articles on DEV Community by Media Sharing (@masroficom).</description>
    <link>https://dev.to/masroficom</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4241335%2Fe226d48a-d38b-48cf-b794-05f43fc7b507.png</url>
      <title>DEV Community: Media Sharing</title>
      <link>https://dev.to/masroficom</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/masroficom"/>
    <language>en</language>
    <item>
      <title>10,600 AI-Assisted TB Screening Devices: Deploying Diagnostic AI at National Scale in Indonesia</title>
      <dc:creator>Media Sharing</dc:creator>
      <pubDate>Thu, 12 Nov 2026 15:14:51 +0000</pubDate>
      <link>https://dev.to/masroficom/10600-ai-assisted-tb-screening-devices-deploying-diagnostic-ai-at-national-scale-in-indonesia-322p</link>
      <guid>https://dev.to/masroficom/10600-ai-assisted-tb-screening-devices-deploying-diagnostic-ai-at-national-scale-in-indonesia-322p</guid>
      <description>&lt;p&gt;Every five minutes, two people in Indonesia die of tuberculosis. That is how Health Minister Budi Gunadi Sadikin opened the 40th Stop TB Partnership Board Meeting in Badung, Bali, on 5 November 2026 — and announced the government's most concrete step yet to break that chain: 10,600 molecular TB diagnostic devices for community health centres (puskesmas) in 2027.&lt;/p&gt;

&lt;p&gt;For anyone who builds or deploys AI systems, the interesting part is not the headline number. It is the architecture: a two-layer screening pipeline, a deliberate cost curve, and a genomics programme running in parallel. Indonesia is not piloting AI diagnostics. It is industrialising them.&lt;/p&gt;

&lt;p&gt;Layer One: AI Reads the X-Ray. Layer Two: The Molecule Confirms It&lt;br&gt;
The pipeline the Health Ministry designed has two stages. The first is a chest X-ray as the initial screen — photographed, then read with the help of artificial intelligence to flag indications of TB. The second is confirmation through a molecular test at the same health centre. The combination, according to the ministry, is what makes case-finding precise, fast, and able to reach remote island communities.&lt;/p&gt;

&lt;p&gt;The hardware is the near point of care (NPOC) molecular diagnostic device: a small machine that tests sputum samples and can sit in a puskesmas rather than only in a large hospital. Around 4,000 units are ready by the end of 2026, rising to 10,600 through 2027, for 290 million people across 7,000 islands.&lt;/p&gt;

&lt;p&gt;The X-ray side is scaling on a similar schedule. Some 800 portable and mobile X-ray units are already installed, with 4,000 more to be added in 2027, and a target of 10,000 puskesmas equipped with AI-integrated X-ray by 2028.&lt;/p&gt;

&lt;p&gt;The cost curve is explicit. With Global Fund support, one molecular test now costs US$3.60 in 2026, targeted under US$3.00 from 2027. The ministry targets national screening capacity of 10 million people a year — roughly ten times the estimated TB case burden.&lt;/p&gt;

&lt;p&gt;The Genomics Layer: 7,000 Patients, Three Regional Variants&lt;br&gt;
At the same Bali forum, Budi announced research results never previously disclosed: from 7,000 TB patients whose samples were analysed, researchers found three different TB genome variants, differing by region — western, central, and eastern Indonesia. The Global Fund-financed research used Illumina and Oxford Nanopore DNA sequencing technology.&lt;/p&gt;

&lt;p&gt;The finding opens a question that has no answer yet: whether genetic differences in the bacteria between regions affect drug efficacy and vaccine effectiveness. Budi called that the next research agenda. As of the 5 November announcement, the ministry had not published results answering it.&lt;/p&gt;

&lt;p&gt;The vaccine track runs separately. Indonesia is in phase 3 trials of the M72 candidate with South Africa, Kenya, Zambia, and Malawi, having enrolled 2,095 participants — more than 2,000 already injected. The GSK/Gates Foundation vaccine is targeted for production in 2029; early phase 2 data put efficacy at 54 to 57 percent, still an early estimate.&lt;/p&gt;

&lt;p&gt;Why Case-Finding Is the Bottleneck, Not the Drug&lt;br&gt;
The ministry is blunt about where the problem sits. TB drugs already exist and work. The problem is that many sufferers do not know they are sick, because screening comes too late or never happens at all.&lt;/p&gt;

&lt;p&gt;That is why screening is being taken to where people gather. Since 13 August 2026, the ministry has expanded active screening to Islamic boarding schools (pesantren) and dense settlements. The reasoning is measurable: general screening finds around 0.3 percent of cases, while screening at Nusakambangan prison found around 3 percent — ten times higher. There are around 42,000 pesantren with more than 7 million students, plus over 82,000 madrasah with around 10.5 million students; the settlement programme covers 3,373 villages in 38 districts and cities, targeted for completion by the end of 2026.&lt;/p&gt;

&lt;p&gt;The decline is slow. Based on the WHO global TB report presented by Budi on 23 October 2026, estimated cases fell from 1.092 million to 1.090 million, then to 1.080 million. His logic is simple: more screening means more people treated, and more treated means fewer transmitting.&lt;/p&gt;

&lt;p&gt;One discrepancy is worth stating plainly: the ministry cited around 120,000 TB deaths a year in August and October 2026, but 136,000 at the Bali forum on 5 November. No official explanation for the 16,000 difference has been published.&lt;/p&gt;

&lt;p&gt;Screening is integrated with the Free Health Check (CKG) programme, which had served 109.3 million people across 10,319 puskesmas by 22 October 2026. A cough that does not heal for more than two weeks is reason to get checked — the ministry's long-standing advice, and the practical takeaway: early detection is the difference between a disease cured completely and one that keeps spreading silently. Full details on the rollout, the costs, and the vaccine timeline can be read more at Media Sharing: &lt;a href="https://mediasharing.web.id/artikel/kemenkes-target-10600-alat-tbc-2027-skrining-ai-puskesmas" rel="noopener noreferrer"&gt;https://mediasharing.web.id/artikel/kemenkes-target-10600-alat-tbc-2027-skrining-ai-puskesmas&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Media Sharing — Indonesian news portal: national, technology, automotive, sports, economy, health, and fact-checks. &lt;a href="https://mediasharing.web.id/" rel="noopener noreferrer"&gt;https://mediasharing.web.id/&lt;/a&gt;&lt;/p&gt;

</description>
      <category>ai</category>
      <category>healthtech</category>
      <category>indonesia</category>
      <category>tuberculosis</category>
    </item>
  </channel>
</rss>
