AI system shows promise in early Alzheimer's risk detection

Researchers have developed an artificial intelligence system that may assist doctors in identifying individuals at risk of progressing to Alzheimer's disease earlier. Using a model called SigLIP, the system combines brain MRI images with clinical data, including age, sex, and memory test results.

In a study of 416 individuals, the method achieved an AUC of 0.91 for predicting individual risk of progression over the following four years. This performance exceeded other models, such as those based solely on MMSE scores (0.85) or CSF Aβ42 biomarkers (0.73). While the results are promising, researchers emphasize that the AI is not a replacement for medical professionals and that risk predictions must be interpreted alongside clinical examinations and medical history.