IIT Steps Up Again, Invents New Technique To Diagnose COVID Using Chest X-Ray

IITs many times make the right headlines, especially on innovation and research. Yet again, yet another IIT has stepped up with an innovation that people, especially the healthcare sector, could use in times of the pandemic. IIT Jodhpur researchers have designed an innovative automated Artificial Intelligence (AI) solution for coronavirus prediction from chest X-rays. The findings were spread over a huge sample and a high accuracy rate. This experiment was conducted with over 2,500 chest X-ray images and acquired nearly 96.80 per cent sensitivity.

According to the statement released by IIT Jodhpur, the scientists at the institution proposed a deep learning-based algorithm by the name o COMiT-Net (check name), which retains information regarding the abnormalities present in the chest X-ray images to find differences between a COVID infected lung and a non-COVID infected lung.

"The developed AI algorithm not only predicts whether the CXR has COVID-19 pneumonia or not, but it is also capable of identifying the infected regions in the lungs, thus making them explainable," ANI quoted an IIT Jodhpur release.

An Innovative Study By IIT Jodhpur Researchers!

There have been multiple analytical studies in COVID detection using CT scans and X-rays in the last year; most have largely failed to provide an explainable solution. However, this study is unique because it proposed a study that can visually point out the infected areas, and the technique interprets only from the lung region.

As per the IIT Jodhpur, the Artificial Intelligence (AI) solution utilised in this study is explainable from algorithmic and medical points of view.

The Brains Behind The Research

Furthermore, the team of researchers that contributed to this study are Surbhi Mittal, PhD Scholar, Computer Science, IIT Jodhpur, Aakarsh Malhotra, Visiting Research Scholar at IIT Jodhpur, Kartik Thakral, PhD Scholar, Computer Science, IIT Jodhpur, Saheb Chhabra, Visiting Research Scholar at IIT Jodhpur, Santanu Chaudhury, Professor and Director, IIT Jodhpur, Puspita Majumdar, Visiting Research Scholar at IIT Jodhpur, Richa Singh, Professor and Head of Department, Computer Science, IIT Jodhpur, Mayank Vatsa, Professor, Computer Science, IIT Jodhpur, Ashwin Pudrod, Consultant Pulmonologist, Ashwini Hospital and Ramakant Heart Care Centre, India, Anjali Agrawal, Consultant Radiologist, TeleRadiology Solutions, India.

The research paper on the above mentioned project has been officially released in the journal "Pattern Recognition (Volume 122)". The study is also part of the RAKSHAK project under iHuB Drishti and NM-CPS DST at IIT Jodhpur. The researchers also want to develop a full-scale prototype through the knowledge learnt in this project.

Also Read: Wuhan Scientists Warn About New COVID 'NeoCov' Virus With High Infection, Death Rate: Report