Cancer may soon be diagnosed more easily, similar to COVID-19, as an AI based virtual model has been developed.
For years, artificial intelligence (AI) has remained the subject of debate and concern, but researchers are now using the technology to understand one of humanity’s greatest challenges: the mysteries of cancer.
At the center of this effort is Debarka Sengupta, Associate Dean of Innovation, Research and Development at the Indraprastha Institute of Information Technology Delhi, who is using AI and genomics to support the early detection of cancer, understand tumor behavior, and help doctors select treatments tailored to individual patients.
Rather than viewing cancer as a single disease or the result of a mutation in one gene, Sengupta’s laboratory approaches it as a complex biological system. Their research combines molecular biology, genomics, single cell analysis, microfluidics, and AI.
Sengupta said that the goal is to identify subtle signs of cancer that are often hidden in blood, tissues, or large biological datasets and convert them into information that doctors can use in clinical practice.
He explained that AI enables researchers to analyze thousands of genes, different cell types, and medical records simultaneously, revealing patterns that would be nearly impossible to identify manually.
AI helps researchers discover patterns that are extremely difficult to detect through conventional methods. Among the team’s key achievements is the development of an 11 gene blood test based on platelet RNA, which could become an effective low cost screening tool for various types of cancer in the future.
Unlike expensive genome sequencing technologies, the test has been designed to operate on RT-qPCR machines, which were widely used across India during the COVID-19 pandemic.
Sengupta said that this type of test can be performed in the same molecular laboratories equipped with qPCR machines, the number of which increased significantly during the COVID-19 testing period. He added that his team is also working on identifying circulating tumor cells in triple negative breast cancer, where the main challenge is detecting extremely rare cancer cells present in the bloodstream.
Researchers are also developing AI models capable of predicting how different drugs may affect different types of cancer. This could help move beyond the traditional trial and error approach to treatment.
Through a startup called Genesilico, the team is developing an “agentic digital twin.” This AI based virtual model integrates a patient’s molecular profile, medical history, tumor biology, treatment guidelines, and scientific data to help cancer specialists better evaluate different potential treatment options.
