Preeti
Artificial intelligence software developed by Pangea Biomed can predict a patient’s response to cancer treatment by analyzing tumour histopathology images. The software, called ENLIGHT-DP, uses a deep-learning framework to predict gene expression in tumours from slides with staining, eliminating the need for expensive RNA sequencing. Based on these predictions, the software can recommend targeted or immune-based treatments.
A study published in Nature Cancer tested ENLIGHT-DP’s ability to predict patient responses to treatment, finding that patients predicted to respond were two to five times more likely to do so compared to those predicted not to respond. The software matched patients to treatments with a 39.5% higher response rate than baseline.
Pangea Bio believes that existing methods for predicting treatment response from histopathology slides are limited due to the need for large datasets, making ENLIGHT-DP a more accessible and accurate option. The software could help clinicians recommend therapies more quickly, potentially improving outcomes, especially in low- and middle-income countries with limited access to cancer diagnostics.
The study involved researchers from Pangea, the Australian National University, and the US National Cancer Institute. The potential of ENLIGHT-DP to provide rapid and precise treatment recommendations has the potential to impact cancer care, particularly in situations where immediate treatment is necessary.
Source link
Source link: https://pharmaphorum.com/news/deep-learning-ai-predicts-response-cancer-treatment
GIPHY App Key not set. Please check settings