Doctors may soon be able to diagnose an elusive form of heart disease, coronary microvascular dysfunction, within seconds by using an AI model developed at University of Michigan, according to a recent study.The prediction tool significantly outperformed earlier AI models in nearly every diagnostic task and has potential to be a cost-effective diagnostic tool at non-specialty hospitals. 

A study led by NYU Langone Health researchers reveals a new strategy to study Hirschsprung disease in mice that better mimics how the disease manifests in humans. Previous HSCR animal models only looked at the role individual genes played in causing the disease, but the new approach is based on how interactions among multiple genes control the condition.

In an observational study of critically ill patients at UCSF, an AI analysis of medical records paired with a biomarker of lower respiratory infections correctly diagnosed the cause of respiratory failure 96 percent of the time. The method could distinguish between infectious and non-infectious causes more accurately than clinicians in the intensive care unit. Had the model been available when the patients were admitted, it could have cut inappropriate antibiotic use by more than 80%.