Comprehensive Summary
Alzheimer’s disease is a neurodegenerative disorder that affects a large population worldwide, so interventions and prevention techniques have been researched to help prevent an individual from losing their memory and sense of self. The research was performed using electroencephalogram (EEG) data to identify Alzheimer’s disease and mild cognitive impairments. A computer program analyzes the data from the signals using deep ensemble feature learning and a deep support vector machine (DEF-DVSM) architecture. This model breaks down the signals into subbands required for detecting Alzheimer’s and mild cognitive impairment. The DEF-DVSM model significantly advanced the detection of early-onset Alzheimer’s and proved effective, achieving 98.17% accuracy. This surpassed previously used models by classifying subbands of the EEG data, specifically the alpha, delta, and theta bands. The model was found to be robust and precise, as well as being able to diagnose mild cognitive impairments and early Alzheimer’s in patients.
Outcomes and Implications
This computer model is important for the medical community as the data can be generalized to other neurodegenerative disorders. Using early detection models will help to prevent the onset of not only Alzheimer’s, but a multitude of disorders that can affect an individual's well-being. The findings showing the high accuracy rate of this model show that it is crucial to develop similar models that can apply to a wide variety of neurodegenerative and neurological disorders.