학술논문

Stratification of patients with Alzheimer’s disease based on longitudinal neuropsychological tests
Document Type
Conference
Source
2020 IEEE International Conference on Healthcare Informatics (ICHI) Healthcare Informatics (ICHI), 2020 IEEE International Conference on. :1-7 Nov, 2020
Subject
Bioengineering
Computing and Processing
Signal Processing and Analysis
Databases
Medical services
Physiology
Data models
Alzheimer's disease
Informatics
Diseases
Alzheimer’s disease
ADNI
patient stratification
clustering
longitudinal data
Language
ISSN
2575-2634
Abstract
Alzheimer's disease represents a heterogeneous neurodegenerative disorder that affects millions of people worldwide. The heterogeneity of the disorder slows research on the physiological underpinnings of the disorder and impedes development of treatments. We modeled for 329 patients of the ADNI database the progression of their detailed cognitive abilities over a 2-year period and found three distinct sub groups. The groups could not be differentiated by demographic characteristics but only by their cognitive profiles. The two extreme groups show either only a mild impairment or a fast progression consistently across all cognitive sub domains. The third group, however, shows severe impairment in the memory domain but relatively spared loss of other cognitive function. The loss of hippocampal volume in this group underpins the severity of memory loss. Our study shows that models of longitudinal data with detailed features will help increase our understanding of the pathophysiology underlying Alzheimer's disease.