Improving Drug Safety with AI
Toxicologic pathology helps decide whether a new medicine is safe enough to test in people, and at what dose. Few people outside the field know it exists. This 90-second video explains the problem we work on, and how we are building AI to help.
Drug discovery is generating candidate compounds faster than ever. Each one still has to answer two questions before it reaches patients. Is it safe for humans, and at what dose does it become toxic?
Toxicologic pathologists manually examine large volumes of tissue, often hundreds of slides per study, and connect each finding to the controls, every dose group, the other organs and the clinical data. As drug development accelerates, that work is getting harder to scale.
TRACE analyzes pathology images, ToxScribe turns what the models see into language, and Tremont Studio brings these capabilities together, so pathologists can connect evidence across an entire study. Pathologists stay in control and make the call.
We are working with a small number of pharma and CRO partners and evaluating the system on their own studies. If that is relevant to your team, get in touch or write to contact@tremont.ai. You can also meet the team or read our launch announcement.
Watch on YouTube. Rat liver and kidney sections in the video: Open TG-GATEs (Toxicogenomics Project and Toxicogenomics Informatics Project), licensed under CC BY-SA 2.1 Japan. Other tissue in the video is illustrative.