JAMA Network Open (journal)

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.

Experts call for better adherence to guidelines for antiviral prescribing for children at higher risk of severe influenza

Researchers found a substantial gap between recommendations and real-world treatment practices in pediatric EDs (emergency departments) across the country.

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.

Behind the curtain: Secrets of the volatile, delusional brain 

The VUMC approach, based on the “Feeling Safe Program” developed by British psychologist Daniel Freeman, targets specific factors that contribute to persecutory delusions, including worry, safety behaviors, avoidance and hallucinations (such as hearing voices).

AI tools could shorten ‘diagnostic odyssey’ for patients with rare diseases

Large language models achieved diagnostic rates of 13.3% and 10.0%, compared to the historical clinical review rate of 5.6%, and they suggested next steps to evaluate the suggested diagnoses.

(AdobeStock)

Study demonstrates burden of potentially preventable hospitalizations for pneumococcal pneumonias among adults in Tennessee and Georgia 

Community-acquired pneumonia refers to a case of the disease contracted without prior exposure to a health care setting.

1 2 3 5