Associate Professor Joyita Dutta of the Biomedical Engineering (BME) Department and her research colleagues have come up with a groundbreaking approach for subtyping Alzheimer’s disease that promises “broad diagnostic utility,” according to research presented during the 2023 annual meeting of the Society of Nuclear Medicine and Molecular Imaging (SNMMI). This new computational technique for subtyping the genetically complex Alzheimer’s disease combines genomic and tau PET imaging data and is based on a novel clustering framework using “Sparse Canonical Correlation Analysis” (SCCA).

According to an article published by the National Institutes of Health (NIH), SCCA is a statistical tool that examines relationships between two types of variables and provides “sparse solutions” that include more-streamlined small subsets of variables of each type. Another review published by the NIH calls imaging genomics “an emerging field that is rapidly identifying genes that influence the brain, cognition, and risk for disease.”

The authors of the study presented at the SNMMI meeting – titled “An SCCA-clustering framework for Alzheimer’s disease subtyping using tau PET and genomics” – were Dutta and BME Post Doctoral Research Associate Fan Yang of UMass Amherst and Matthew Maher and Richa Saxena of the Massachusetts General Hospital in Boston. Yang, the first author of the imaging genomics abstract, also got a 2023 ERF-SNMMI Annual Meeting Travel Award to attend the conference.

“By identifying different subtypes of Alzheimer’s disease using both imaging and genomic information, researchers could gain potential new insights into the underlying biology of the disease and its progression,” said Dutta in a story published by Imaging Technology News. “Understanding the specific genetic associations for each subtype could also lead to the development of personalized treatment approaches in the future.”

As Dutta explained the ultimate significance of her new study, “Genomics- and imaging-guided individualized subtyping is vital for Alzheimer’s disease because different subtypes may also have distinct rates and profiles of cognitive decline, potentially affecting clinical-trial outcomes and treatment response. By combining molecular-imaging information with genomics, we have created a diagnostic technique that could be truly personalized for each patient. This has potential for broad diagnostic utility across many disease types, not only Alzheimer’s disease.”

Dutta and her collaborators have identified four subtypes of Alzheimer’s disease in their analysis: medial temporal lobe (MTL)-dominant; posterior; MTL-sparing; and lateral-temporal. The four researchers also identified top genes that were associated with each subtype.

In addition to Dutta’s innovative study presented at the SNMMI meeting, her team recently received a $250,000 grant for a pilot project from the Massachusetts Artificial Intelligence (AI) and Technology Center for Connected Care in Aging and Alzheimer's Disease.

Dutta, the Sleep Monitoring Lab Core Facilities at UMass Amherst, and Dreem (a developer of EEG headsets) launched the research project, which uses the Dreem headband and smartwatch to study AI-based sleep staging. (August 2023)

Article posted in Research