Ever since the Covid-19 disease started spreading, there has been a lot of research and medications that are being improved day by day. Several scientists have emphasized how the disease can cause damage to the lungs. During the second wave of Covid-19, cases surged due to the Delta variant which caused an increase in mortality rate, there was a rise in the number of RT-PCR tests taken, and the test-positivity rate also increased parallelly, in several countries. It is the time when taking CT and X-Ray scans went to peak, this was because CT and X-Ray scans showed how much the lungs were affected. 

With increasing Covid-19 cases affecting across the globe, countries have faced challenges with the limited availability of testing kits and health centres in remote areas. 

AI-based technology

This has motivated researchers to come up with many new technologies, likewise, IIT Jodhpur research team have developed an Artificial Intelligence-based X-Ray design to detect Covid-19. The research team has proposed a deep learning algorithm called “COMit-Net” which has the capability to learn the abnormalities present in the chest X-Ray images. The project will detect the presence of Covid-19 using chest X-Ray images. This new technique is capable of identifying infected regions in the lung area. 

https://twitter.com/iitjodhpur/status/1487253027400519681?s=20&t=xeSz5FzbuHjkzjqh6chZow

Accurate diagnosis

The AI-based technology is a part of the RAKSHAK project under NM-CPS DST and iHuB Drishti at IIT Jodhpur. A chest X-Ray through radiation has the capability to produce pictures of a person’s lungs, but the new technology that’s has been developed by IIT Jodhpur has the capacity to not only predict whether a person has Covid-19 pneumonia or not but also identifies the infected region in the lungs. 

There were numerous studies and researches that have taken place in detecting the Covid-19 using X-Ray or CT scans in the past years but most of them failed to provide explainable solutions, but this project “visually” showcases the region which is infected. The technique interprets only from the lung region.

The experiment was done with 2,500 chest X-Ray images and has achieved 90.8 per cent sensitivity.  The proposed study has the capability to visually showcase the lung region which is infected.  The AI solutions used is explainable from both algorithmic and medical point of views. 

Earlier, an AI-Based X-ray technique was developed by Scottish researchers which can replace the use of PCR tests in detecting Covid-19. This technology has the capability to accurately diagnose Covid-19 in just a few minutes, according to reports. 

Related Topics