AI Used to Determine Cause of Alzheimer’s and Related Disorders

Brain Cancer, Artwork
Photo credit: Science Photo Library – SCIEPRO / Brand X Pictures / Getty Images

Novel artificial intelligence methods have uncovered unexpected microscopic abnormalities that can predict cognitive impairment, according to a study led by researchers at Mount Sinai. These results were published in the journal Acta Neuropathologica Communications in this week.

“AI represents a whole new paradigm for studying dementia and will have a transformative impact on the study of complex brain diseases, particularly Alzheimer’s disease,” said co-author John Crary, MD, PhD, Professor of Pathology, Molecular and Cell-Based Medicine, Neuroscience and Artificial Intelligence and Human Health at the Icahn School of Medicine at Mount Sinai.

He added: “The deep learning approach has been applied to the prediction of cognitive impairment, a challenging problem for which there is currently no human-performed histopathological diagnostic tool.”

The Mount Sinai team identified and analyzed the underlying architecture and cellular features of two regions in the brain, the medial temporal lobe and the frontal cortex. In an effort to improve the standard of postmortem brain assessment to identify signs of disease, the researchers used a low-monitored deep learning algorithm to generate slide images of autopsy human brain tissues from a group of more than 700 elderly donors investigate to predict the presence or absence of cognitive impairment.

The weakly supervised deep learning approach, they report, is able to process noisy, finite, or imprecise sources to provide signals for labeling large amounts of training data in a supervised learning environment. This model was used to detect a reduction in Luxol Fast Blue staining, which is used to quantify the amount of myelin, the protective layer around brain nerves.

Researchers identified a cognitive impairment signal associated with decreasing amounts of myelin staining; scattered in an uneven pattern across the fabric; and focuses on the white matter, which affects learning and brain function. The two sets of models trained and used by the researchers were able to predict the presence of cognitive impairment with an accuracy better than random guessing.

The team believes the reduced staining intensity in specific areas of the brain identified by AI could serve as a scalable platform to assess the presence of brain damage in other associated diseases. The methodology forms the basis for future studies, which could include the deployment of artificial intelligence models at larger scale, as well as further decomposing the algorithms to increase their predictive accuracy and reliability. The team said the ultimate goal of this neuropathology research program is to develop better tools for diagnosing and treating people suffering from Alzheimer’s disease and related disorders.

“Leveraging AI allows us to study exponentially more disease-related traits, a powerful approach when applied to a complex system like the human brain,” said co-author Kurt W. Farrell, PhD, assistant professor of pathology, molecular and neuroscience cell research. Based on medicine, neuroscience and artificial intelligence and human health at Icahn Mount Sinai. “It is critical to conduct further interpretability research in the fields of neuropathology and artificial intelligence so that advances in deep learning can be translated to safely and effectively improve diagnostic and treatment approaches for Alzheimer’s disease and related disorders.”

Lead author Andrew McKenzie, MD, PhD, Co-Chief Resident for Research in the Department of Psychiatry at Icahn Mount Sinai, added: “The interpretive analysis was able to identify some, but not all, of the signals that the artificial intelligence used to make predictions about cognitive impairment. As a result, additional challenges remain for the deployment and interpretation of these powerful deep learning models in the field of neuropathology.”

Researchers from the University of Texas Health Science Center in San Antonio, Texas, Newcastle University in Tyne, UK, Boston University School of Medicine in Boston, and UT Southwestern Medical Center in Dallas also contributed to this research.