As trustworthiness takes center stage in medical AI, a new paper in Nature Biomedical Engineering describes a solution for trustworthy AI cancer subtyping.

The artificial intelligence model Enformer was retrained with tissue-specific datasets; this "transfer learning" approach outperformed the base model in identifying disease-associated genes.

The new artificial intelligence tool aims to spare Vanderbilt Health patients and clinicians the back-and-forth that can delay answers to patients' electronic health care questions.

Large language models were tested on cancer patient notes from two academic medical centers and seven drug trials.

Risk prediction software using artificial intelligence can help determine whether pulmonary nodules detected during lung cancer screening need to be biopsied.

The collaboration will integrate Vanderbilt Health’s Molecular AI Initiative capabilities with Bertis’ deep proteomics and AI-enabled technologies to identify novel therapeutic targets and predictive biomarkers.

A novel AI framework found documented intolerance to statins in 6.4% of all adult patients, contraindications in 0.7%, and statin deferral in 2.9%.