9 months ago
AI Framework Predicts Cancer Behavior for Personalized Therapy
Scientists in India have created a smart computer program called OncoMark to help doctors understand cancer better.
Instead of just looking at how big the cancer is, this program looks at the tiny parts inside the cancer cells.
It can tell doctors which parts are active and might help the cancer grow or hide from treatments.
The program was tested on lots of cancer samples and worked very well.
It can help doctors choose the best medicine for each person and find cancers that might be more dangerous than they look.
This could make cancer treatments more personal and effective.
Scientists developed an AI framework called OncoMark to analyze cancer's molecular hallmarks.
OncoMark can predict cancer behavior and guide personalized therapy by identifying active hallmarks.
The framework was tested on 3.1 million single cells across 14 cancer types with high accuracy.
It was validated on 20,000 real-world patient samples from eight major datasets.
OncoMark can help doctors choose targeted drugs and identify aggressive cancers for earlier intervention.
- Who
- Scientists at S N Bose National Centre for Basic Sciences and Ashoka University
- What
- Developed an AI framework called OncoMark for personalized cancer therapy
- Where
- New Delhi, India
- When
- Not specified
- Why
- To better understand and treat cancer by analyzing its molecular personality and hallmarks
Key facts
- Framework Name
- OncoMark
- Developed By
- S N Bose National Centre for Basic Sciences and Ashoka University
- Data Analyzed
- 3.1 million single cells across 14 cancer types
- Accuracy
- Over 99% in internal testing, above 96% across five independent cohorts
- Validation
- 20,000 real-world patient samples from eight major datasets
- Publication
- Nature journal Communications Biology
Quotes
Ministry of Science and Technology
Government ministry responsible for science and technology in India
“OncoMark achieved over 99 per cent accuracy in internal testing and remained above 96 per cent across five independent cohorts. It was validated on 20,000 real-world patient samples from eight major datasets, showing broad applicability. For the first time, scientists could actually visualise how hallmark activity rises with advancing cancer stage.”
thehansindia.com
“Cancer is not just a disease of growing tumours -- it is powered by a set of hidden biological programmes called the hallmarks of cancer. These hallmarks explain how healthy cells turn malignant: how they spread, evade the immune system, and resist treatment.”
thehansindia.com
