GUWAHATI: Researchers from the
Indian Institute of Technology
, Guwahati (IIT-G), and Cotton University (CU) have achieved a major breakthrough by developing an automated system to detect
with 95.24% accuracy.
The USP of the innovation, for which researchers from Harvard and some globally renowned universities also contributed during the study, is the use of
algorithms that have made cancer detection very efficient.
The researchers from the two leading higher education institutions based in Guwahati, have jointly designed an automated AI (Artificial Intelligence)-based system to detect colorectal cancer using colonoscopy images. Colonoscopy is a technique for early detection of colorectal cancer, which has become increasingly popular for early identification and prevention of colorectal carcinoma.
“In the current manual approach for colonoscopy examination by physicians, observation bias may sometimes lead to an erroneous diagnosis,” said Kangkana Bora, an assistant professor of computer science and IT department of CU, who carried out the work under the supervision of Prof MK Bhuyan of the electronics and electrical engineering department of IIT-G.
The research team believes that the work has a global impact on reducing the incidence of colorectal cancer, a formidable health problem worldwide. It is the third most common cancer in men (6,63,000 cases and 10.0% of all cancer cases) and women (5,71,000 cases, 9.4% of all cancer cases).
They have worked on real colonoscopy images generated by Dr Kunio Kasugai of Aichi Medical University, Japan. Prof Zhongming Zhou from the
University of Texas
, Health Science Centre, Houston, USA, and Dr Saurav Mallik of
, USA, has also contributed to the study.
The research work has been published recently in reputed journal Scientific Reports recognizing its authenticity. “This AI-based solution can reduce the laborious task of physicians by sieving out the normal cases so that they can concentrate more on suspicious cases, as doctors normally spend lots of time looking at normal cases,” Bhuyan said.
Now, they plan to commercialize the technology in the future as the market need is enormous, both in terms of societal and medical view. “This AI-based solution does not require any unique ecosystem for application. The proposed software can be easily integrated with the current methods of diagnosis,” Bora added.