From patterns in cells to more accessible care. AI opens new questions about health, development and aging.
Researchers trained an open system to outline tau tangles in post-mortem brain tissue after experts marked them with single points. It may speed up pathology research, but it is not a test for diagnosing Alzheimer’s disease in living people.

A study in South Korea found that AI could distinguish behavioral patterns in brief family interactions. Whether it can help in everyday care remains unproven.

Researchers trained models to estimate aging signals for 11 physiological systems from DNA methylation in blood. The scores tracked several age-related conditions, but they are research measures, not a clinical diagnosis or a guide to treatment.

A system used during tumor profiling suggested a different tissue of origin for a small share of cases. Doctors checked those suggestions against pathology, imaging and other evidence before changing a diagnosis.

A chemical language model proposed molecules for a tuberculosis protein, and laboratory tests found 14 generated or closely related compounds that inhibited the target. They are early research hits, not medicines, and their safety and activity in living systems remain unknown.

SenseToKnow combined computer vision and machine learning in a prospective study of 475 toddlers. The results suggest a possible screening aid, with independent evaluation still needed.