B1219
Title: Machine learning for Alzheimer's patients stratification and target identification
Authors: Aamna AlShehhi - Khalifa University (United Arab Emirates) [presenting]
Abstract: Alzheimer's disease (A.D.) is an insidious, progressive, and degenerative neurodegenerative disease that destroys normal brain functionality. According to the World Health Organization (WHO), Alzheimer's disease is the most common form of dementia and contributes to approximately 70\% of all dementia cases. In 2018 Alzheimer's Association reported that an estimated 5.7 million Americans were diagnosed with A.D. The number of patients is expected to double by 2050. A.D. is known to be caused by the presence and aggregates of tau neurofibrillary tangles and amyloid-beta (A) plaques in the brain. The heterogeneity of A.D. patients and the complexity of the disease genomic mechanisms create challenges for disease diagnosis, addressing patients' needs, and understanding treatment response. That is why machine learning and deep learning models can play a vital role in addressing those challenges. Our study aims to cluster A.D. patients into clinically homogeneous groups by linking Electronic Health Records (HER) patients' data with genomic information using different machine learning models and discovering a pivotal biomarker related to each specific stratum. A different disease pathway detected for different cohorts is confirmed and discovered.
