New Study Aims to Predict Susceptibility to Alzheimer’s Disease

A new study funded by the National Institutes of Health aims to harness artificial intelligence and machine learning to predict individual susceptibility to Alzheimer’s disease and aging.

The four-year study, led by researchers at Columbia University Irving Medical Center, will employ advanced artificial intelligence and machine learning methodologies, coupled with multi-omics and multi-organ approaches, to comprehensively model the neurobiological and genetic underpinnings of the disease.

“It is well-known that the disease pathogenesis of brain diseases such as Alzheimer’s is multifaceted, and modeling disease effects using neuroimaging data alone may be insufficient,” said Junhao Wen, PhD, assistant professor of radiological sciences and principal investigator of the study. Wen proposes to combine multiple sources of biomedical data­—including genetics, proteomics, and imaging—to shed new insights on the disease. His approach also incorporates data relating to organs other than the brain, such as the eye and heart, which have been shown to have a multidirectional connection with Alzheimer’s disease.

Diagnosing Alzheimer’s disease at an early stage is critical to slowing its progress through medications and lifestyle changes. However, current diagnostic techniques are limited, and robust biomarkers are still needed for both early diagnosis and prognosis.

Wen, who is also the director of imaging genetics research for Columbia's new Center for Innovation in Imaging Biomarkers and Integrated Diagnostics (CIMBID), aims to create a tool for predicting personalized susceptibility to Alzheimer’s disease. The derived individualized susceptibility scores can improve precision in diagnosis and prognosis compared to previous single-organ and single-omics approaches, ultimately improving healthcare outcomes.

The grant, titled “Multi-Organ Chart of Personalized Susceptability to Alzheimer’s Disease and Aging,” was provided by the National Institute on Aging, part of the NIH. (1RF1AG092412-01)