Related Experiment Videos
A Deep Learning Approach to Neuroanatomical Characterisation of Alzheimer's Disease
Abhinit Kumar Ambastha1, Tze-Yun Leong1,
1Medical Computing Laboratory, School of Computing, National University of Singapore, Singapore.
Studies in Health Technology and Informatics
|January 4, 2018
Summary
This study uses computational methods to understand Alzheimer's disease (AD) progression by identifying brain region correlations. The approach achieved 81.79% accuracy in classifying AD patients from healthy individuals.
Area of Science:
- Neuroscience
- Computational Biology
- Medical Imaging Analysis
Background:
- Alzheimer's disease (AD) is a progressive neurological disorder causing mental deterioration.
- Understanding the patterns of brain degeneration in AD is crucial for early diagnosis and treatment.
Purpose of the Study:
- To introduce a novel computational approach for analyzing Alzheimer's disease progression.
- To identify structural correlations among brain regions affected by AD.
Main Methods:
- Utilized ensemble learning methods and deep neural networks.
- Analyzed structural correlations within brain regions exhibiting simultaneous degeneration in AD.
- Employed a single modality imaging dataset for classification.
Main Results:
- The computational technique successfully identified salient structural correlations indicative of AD progression.
- Achieved a classification accuracy of 81.79% for distinguishing Alzheimer's disease from healthy subjects.
Conclusions:
- The proposed computational approach enhances the understanding of how Alzheimer's disease progresses in the brain.
- This method shows potential for accurate AD diagnosis using neuroimaging data.