DEV Community

Cover image for Lung Infection Quantification of COVID-19 in CT Images with Deep Learning
Paperium
Paperium

Posted on • Originally published at paperium.net

Lung Infection Quantification of COVID-19 in CT Images with Deep Learning

New deep learning tool measures COVID-19 lung infection from CT scans fast

Doctors use chest CT scan to spot and follow COVID-19, but reports often say only some or many areas are sick.
A new deep learning method now can find and measure the exact lung infection areas, and give a clear percent of how much lung is affected.
It was tested on 300 scans and matched human outlines very closely, so the numbers are reliable.
The system works automatic but lets doctors step in quickly when needed, cutting the time to draw regions down to about four minutes after a few rounds of training.
That means faster check-ups and easier tracking of progress or recovery.
This tool show where infection sits in different parts of the lung and can help connect scans with how the patient feels.
For patients and families, that could mean clearer updates and better decisions.
It may not replace a radiologist, but it gives them a fast, clear picture to work from.
COVID-19 care just got a bit more measurable, and follow-ups easier to compare.

Read article comprehensive review in Paperium.net:
Lung Infection Quantification of COVID-19 in CT Images with Deep Learning

🤖 This analysis and review was primarily generated and structured by an AI . The content is provided for informational and quick-review purposes.

Top comments (0)