In an important breakthrough in COVID research, two Class-XII students of Dhirubhai Ambani International School (DAIS), have devised a model that can successfully detect COVID-19 through the chest X-rays of a patient. Machine Learning Model developed by Sumeet Motwani and Sidharth Parekh will be able to successfully detect COVID-19 through chest radiographic images in less than a second with an accuracy of more than 90%.
It's important to note that, the students have decided against monetizing such technology in a time when it is desperately needed by the world and have instead decided to make it fully Open Source. With this approach, the Model is now freely available for interested Government Agencies, Public and Private firms to develop their own Product and rapidly deploy it across the nation to save lives.
Unlike common ways of testing like the nasopharyngeal swab technique, this technology will be able to detect the presence of the COVID virus in the lungs of a person just through their CT scans proving to be a much faster way to detect the novel virus. Sumeet Motwani and Sidharth Parekh are the first ones in India to come up with such technology.
"The technology works on a Convolutional neural network (CNN) and it acts as a classifier, so within one second it can tell whether the person is COVID positive or negative based on their CT scan," said Sumeet Motwani.
"When a person is COVID positive, there are changes in their lung structure. This will be able to detect those," he added.
For diagnosing COVID, doctors mainly employ blood/swab testing and/or chest X-rays. While looking through the CT scans of a potential patient through the naked eyes are one way, doctors usually wait to get the result of the blood test before confirming the diagnosis. However, this technology will be able to fasten the process of COVID testing since this model is solely dedicated to detecting COVID through their CT scans.