Home testing kits that screen for cervical cancer risk could be a game-changer for reducing health inequalities for physically Disabled women, according to a new University of Sheffield study revealing that over 50% would prefer a self-test over a traditional clinic visit.

Researchers from the Technion – Israel Institute of Technology, in collaboration with international partners, have developed an artificial intelligence (AI) model that predicts both the risk of breast cancer recurrence and the likelihood of benefit from chemotherapy using standard pathology slides. The study, published in The Lancet Oncology and presented at ESMO, is the first to validate such a model using data from a large randomized clinical trial (TAILORx).

The AI system analyzes high-resolution digital images of tumor tissue and identifies complex visual patterns linked to cancer behavior and treatment response. It generates a score within minutes, offering a fast, low-cost, and widely accessible alternative to genomic tests such as Oncotype DX, which are expensive and not globally available.

The model was validated on thousands of patients across multiple countries and healthcare systems, demonstrating consistent performance. By enabling more accurate treatment decisions, the approach could reduce unnecessary chemotherapy and expand access to personalized cancer care, particularly in low-resource settings.

Researchers are now working toward clinical implementation and further development, including expansion to additional cancer types and treatments.

A machine learning model that analyzes patient demographics, electronic health record data, and routine blood test results predicted a patient’s risk of hepatocellular carcinoma (HCC), the most common type of liver cancer, with high accuracy.