Now, lung cancer diagnosis just a click away

Tribune News Service

Chandigarh, November 8

The PGI, Chandigarh, in collaboration with the IIT, Kharagpur, has developed India’s first indigenous computer-aided detection (CAD) system by creating the hospital’s own database for the diagnosis of early lung cancer and various patterns of interstitial lung diseases (ILD). 

It will act as a learning tool for trainee radiologists and help physicians working in remote areas and peripheral health centres to make a confident diagnosis of these diseases. Additionally, it will work as a source of second opinion even for experienced radiologist. 

Dr Mandeep Garg, Professor, Department of Radiodiagnosis, PGI, did the research project on the topic, ‘Content Based Image Retrieval (CBIR)-based CAD”, in collaboration with the IIT. They were given Rs 1 crore by the Department of Electronics and Information Technology (DeiT), New Delhi. 

Dr Garg said their team collected medical and radiological data of 150 cases of early lung cancer i.e lung nodules (both cancerous and non-cancerous), 200 cases of interstitial lung diseases (ILD) and 250 normal cases. Initially, they manually marked various abnormalities on CT scans and the database was created. 

“All nodules were confirmed as benign or malignant either on biopsy or other tests. This entire data set was given to the IIT team that developed the CBIR-based CAD. This required multiple levels of our feedback and inputs to finally develop the software. Now, all one needs to do is to open any new CT scan in our software and just click the mouse on the abnormal looking area. Then our software will automatically bring out six similar cases from the database with diagnosis. Isn’t it amazing that doing nothing, computer straightaway gives you the diagnosis,” Dr Garg shared. 

“Many researches nowadays do not translate into the actual patient care and I hope that this one after running physical trial will be available for patient use,” opined Dr MS Sandhu, Head, Department of Radiodiagnosis.

Dr Garg elaborated, “In case there is no radiologist available, then physician can also check the status by a click of a button on the system. We will share this database with other hospitals. It can also be used commercially if the government wants.”