The web-based platform integrates research and clinical workflows, enabling doctors to plan treatment strategies while factoring in patients’ socio-economic conditions. Photo | Express Illustration
Health

AI framework for early cancer diagnosis developed

The web-based platform integrates research and clinical workflows, enabling doctors to plan treatment strategies while factoring in patients’ socio-economic conditions.

Bosky Khanna

BENGALURU: Timely detection and treatment of cancer is often expensive and time-consuming. However, researchers say artificial intelligence (AI) can now help detect cancer at the very first consultation, enabling quicker initiation of personalised treatment. A team of researchers from across India has developed a first-of-its-kind AI framework capable of reading the molecular behaviour of cancer cells to assess their spread and aggressiveness for immediate clinical intervention.

Dr Debayan Gupta, Assistant Professor of Computer Science at Ashoka University and head of the research team, said the tool, named OncoMark, helps doctors determine the stage of cancer at the time of diagnosis and design patient-specific treatment strategies. He said data on 14 types of cancer has been fed into the software. Based on patient samples and diagnostic inputs, the system conducts a detailed analysis to identify the type of cancer and its level of aggression.

“Using AI for detection and treatment not only saves time but also reduces costs. Cancer treatment is highly individualised, and doctors often have to try multiple combinations before arriving at the right approach. With dedicated AI tools, treatment can be faster and more accurate,” Dr Gupta said.He added that the current focus has been on solid tumours, while work on blood and liquid cancers is underway.

The web-based platform integrates research and clinical workflows, enabling doctors to plan treatment strategies while factoring in patients’ socio-economic conditions. Accurate transcriptomic data analysis aids in understanding tumour biology, which is crucial for effective treatment planning. The findings were documented in a paper titled “OncoMark: A high-throughput neural multi-task learning framework for comprehensive cancer hallmark qualification”, published in the journal Communications Biology. The research has also been recognised by the Ministry of Science and Technology.

LIVE | Iran conflict: Iran warns attack on ships trying to pass Strait of Hormuz, US says it destroyed IRGC command posts

Over 200 Indian students shifted from Tehran to Qom amid strikes in Iran

Air India operates first flight from Dubai amid West Asia conflict; 149 stranded passengers reach Delhi

Rahul Gandhi questions PM Modi’s silence on Khamenei’s killing, calls for clarity on India’s stand

CBSE postpones board exams scheduled on March 5, 6 in West Asia amid Iran-Israel conflict

SCROLL FOR NEXT