If you are a hospital lab dealing with the high cost and risk of over-treating cancer patients — this project developed AI software that identifies low-risk patients in minutes. This can prevent up to 250,000 patients from receiving unnecessary chemotherapy.
AI-Powered Digital Pathology to Reduce Unnecessary Chemotherapy in Colorectal Cancer Patients
Imagine if doctors could tell exactly who needs chemotherapy and who doesn't just by looking at a digital photo of a tumor. Right now, many people get harsh treatments they don't actually need because the current tests aren't precise enough. This tool uses AI to scan those images and accurately predict if a patient will truly benefit from the medicine.
What needed solving
Up to 90% of colorectal cancer patients receiving adjuvant chemotherapy may not benefit from it, leading to severe side effects and unnecessary healthcare spending.
What was built
Deep learning algorithms applied to cancer tissue whole slide images (WSI) to stratify patient risk.
Who needs this
Who can put this to work
If you are an insurer dealing with the high cost of cancer care and side-effect hospitalizations — this project developed a biomarker that could save 4 billion EUR globally every year by reducing unnecessary treatments.
If you are a software provider dealing with the need for more clinical utility in digital pathology — this project developed a deep learning algorithm for whole slide images that provides a hazard ratio >10 for risk stratification.
Quick answers
How does this reduce healthcare costs?
Based on available project data, the tool can identify patients who do not need adjuvant chemotherapy, potentially saving 4 billion EUR globally every year.
Can this be scaled to existing labs?
Yes, the test can be run in local labs using existing digital pathology equipment and does not consume any tissue.
What is the intellectual property or licensing model?
Based on available project data, the technology is developed by DoMore Diagnostics AS, but specific licensing terms are not provided.
What is the clinical accuracy of the tool?
The biomarker provides risk stratification (low, intermediate, or high) with a hazard ratio >10.
What is the timeline for implementation?
The project period is from 2024-06-01 to 2026-05-31, focusing on broadening clinical validations to become the gold standard.
Who built it
The project is led by a single partner, DoMore Diagnostics AS, a Norwegian SME. This 100% industry-led structure suggests a strong focus on commercialization and rapid market entry rather than academic exploration.
Contact DoMore Diagnostics AS in Norway
Talk to the team behind this work.
Contact us to explore licensing or partnership opportunities with DoMore Dx.