Department of Biomedical Informatics

AI gets risk wrong in the clinic

Large language models have become far better than you at solving math problems but may still be bad at employing probabilistic terms in medical contexts.

Large-scale integration of genomics, proteomics and clinical records identifies cancer risk proteins and potential therapeutics

A new study identifies previously unreported protein biomarkers and candidate drug targets across six major cancer types and highlights approved drugs with potential cancer preventive effects.

AI recruited to lower leading cause of preventable hospital deaths

Risk of VTE rises in hospitals due to factors such as patient immobility, effects from major surgery, and use of indwelling medical devices like central lines.

(AdobeStock)

Artificial HIV patients do their part for science

To speed HIV research, VUMC scientists are drafting droves of people who do not exist to serve as research subjects.

New resources help people take action to prevent inherited diseases

To launch a connection search, someone with a specific gene variant signs up to meet others who share that variant, beginning a process of discovering new connections with unknown relatives and encouraging family members to get tested.

Three VUMC leaders elected to the National Academy of Medicine

With more than 2,400 members, the NAM (formerly the Institute of Medicine) collaborates closely with its peer academies, the National Academy of Sciences and National Academy of Engineering, to provide independent and authoritative advice on matters of science, technology and health nationally and globally.

1 2 3 26