The AI model developed at IIT Jodhpur can detect COVID in X-ray scans!

To detect Covid, researchers at IIT Jodhpur created the COMiT-Net, a deep learning-based algorithm. To distinguish between COVID-19 damaged lungs and normal lungs, the model can learn the anomalies evident in chest X-Ray images.

Furthermore, the AI model can identify the COVID-19-affected lung areas. The AI-based method, according to IIT Jodhpur, can also be used instead of traditional RT PCR.

Our researchers from @CSEIITJ1 have developed an AI-enabled software to detect #Covid_19 from lung x-ray images. It is capable of identifying the infected regions in the lungs!

This is a part of the RAKSHAK project under NM-CPS DST and iHuB Drishti at IIT Jodhpur. @IndiaDST pic.twitter.com/dsVAAby9l0

— Indian Institute of Technology Jodhpur (@iitjodhpur) January 29, 2022

The investigation used approximately 2,500 chest X-ray scans and achieved a sensitivity of around 96.80%.

“With the increasing number of COVID-19 cases around the globe, countries have faced challenges with the limited availability of testing kits and processing centers in remote areas.” This has been the key motivation for researchers to find alternate methods of testing that are reliable, easily accessible, and faster, ”IIT Jodhpur stated.

The AI solution used in this study, according to the institute, is explainable from both an algorithmic and a medical standpoint.

The journal ‘Pattern Recognition (Volume 122)’ released a full study paper on this project. The study is part of the RAKSHAK project at IIT Jodhpur, which is funded by NM-CPS DST and iHuB Drishti.

Aakarsh Malhotra, a visiting research scholar at IIT Jodhpur; Surbhi Mittal, a PhD scholar in Computer Science at IIT Jodhpur; Saheb Chhabra, a visiting research scholar at IIT Jodhpur; Puspita Majumdar, a visiting research scholar at IIT Jodhpur; and Kartik Thakral, a PhD scholar in Computer Science at IIT Jodhpur; and Kartik Thakra.