Investigating conversion from mild cognitive impairment to alzheimer's disease using latent space manipulation

Download

HIBIT22_paper_97.pdf

Date

2022-10

Author

Ayvaz, Deniz Sezin
Baytaş, İnci M.

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

222
views

79
downloads

Alzheimer’s disease, a progressive neurologic disorder, is the most common cause of dementia, affecting millions worldwide. Mild Cognitive Impairment (MCI) is considered an intermediate stage before Alzheimer's. Early prediction of the conversion from MCI to Alzheimer's is crucial to take necessary precautions for decelerating the disease progression and developing suitable treatments. This study proposes a deep learning framework to identify patients whose diagnoses might change from MCI to Alzheimer’s in the following stages. In particular, latent space manipulation techniques are applied to the latent space of a variational autoencoder trained with MCI and Alzheimer’s patients. The manipulation step aims to reveal significant attributes triggering the conversion. Secondly, a correlation between the manipulation’s magnitude and the conversion time is investigated to introduce a predictive perspective. Experimental results show promising quantitative and qualitative results on one of the literature's most extensive and commonly used Alzheimer’s disease neuroimaging datasets.

URI

https://hibit2022.ims.metu.edu.tr/
https://hdl.handle.net/11511/101308

Conference Name

The International Symposium on Health Informatics and Bioinformatics

Collections

Graduate School of Informatics, Conference / Seminar

Suggestions

OpenMETU
Core

COMPARATIVE ANALYSIS OF BRAIN CELL CULTURES AND TISSUES IN ALZHEIMER’S DISEASE BASED ON DIFFERENTIAL EXPRESSION AND GENE SET ENRICHMENT Burduroğlu, Hüseyin Cahit; Aydın Son, Yeşim; Department of Bioinformatics (2023-1-25) Alzheimer’s disease is currently the most common cause of dementia in the world. It is a neurodegenerative disease that is diagnosed neuropathologically by observing B-amyloid plaques and neurofibrillary tangles in the brain. Transcriptional differentiations, protein regulations, and the interactions in between have been investigated by recent studies to understand from which brain cell type the disease stems, such as microglia, astrocytes, and neurons. These studies are mostly performed on brain tiss...
Fluorescent labeling of SPOCK1 and localization studies in live neural cell lines Karaboğa, Zeynep Eda; Son, Çağdaş Devrim; Aydın Son, Yeşim; Department of Biology (2022-11-30) Alzheimer's disease is the most common form of dementia and is incurable. The late onset Alzheimer’s disease (LOAD) is associated with complex genetic transition, and the molecular background has not been enlightened yet. Our previous research on the meta-analysis of genome-wide association studies (GWAS) showed a statistical correlation between SPOCK1 variants and LOAD. Also, it is co-regulated with apolipoprotein (APP), one of the proteins clinically linked with LOAD. Furthermore, the SPOCK1 protein ...
Gene-level pathogenicity scores for alzheimer’s disease using genomic variants from rna-seq data Bozkurt, Fatma Betül; İlgün, Atılay; Uzuner, Dilara; Çakır, Tunahan (Orta Doğu Teknik Üniversitesi Enformatik Enstitüsü; 2022-10) Alzheimer’s disease (AD) is a complex neurodegenerative disorder affecting millions of people worldwide. Next-generation sequencing technologies such as whole-exome/genome sequencing have been widely used for detecting the variants in the genome to understand the disease etiology and unravel underlying molecular mechanisms. Alternatively, RNA-Seq data can also be used to detect variants. Since AD is a complex disease, several variants are involved in the disease pathogenesis. By using scoring algorithms, it...
Understanding the link between alzheimer’s disease and type 2 diabetesin terms of metabolic alterations Lüleci, Hatice Büşra; Çakır, Tunahan (Orta Doğu Teknik Üniversitesi Enformatik Enstitüsü; 2022-10) UNDERSTANDING THE LINK BETWEEN ALZHEIMER’S DISEASE AND TYPE 2 DIABETES IN TERMS OF METABOLIC ALTERATIONS Hatice Büşra Lüleci, Tunahan Çakır Department of Bioengineering, Gebze Technical University, Kocaeli, Turkey ABSTRACT Alzheimer’s disease (AD) is a type of dementia that causes impairment in memory, reasoning, and thinking. Type 2 diabetes (T2D) is common in the general elderly population and is significantly associated with a higher risk of dementia. However, metabolic alterations responsible for this a...
A human carboxypeptidase E/NF-alpha 1 gene mutation in an Alzheimer's disease patient leads to dementia and depression in mice Cheng, Y.; Cawley, N. X.; Yanık, Tülin; Murthy, S. R. K.; Liu, C.; Kasikci, F.; Abebe, D.; Loh, Y. P. (2016-12-06) Patients with Alzheimer's disease (AD), a common dementia among the aging population, often also suffer from depression. This comorbidity is poorly understood. Although most forms of AD are not genetically inherited, we have identified a new human mutation in the carboxypeptidase E (CPE)/neurotrophic factor-alpha 1 (NF-alpha 1) gene from an AD patient that caused memory deficit and depressive-like behavior in transgenic mice. This mutation consists of three adenosine inserts, introducing nine amino acids, i...

Citation Formats

D. S. Ayvaz and İ. M. Baytaş, “Investigating conversion from mild cognitive impairment to alzheimer’s disease using latent space manipulation,” Erdemli, Mersin, TÜRKİYE, 2022, p. 2097, Accessed: 00, 2023. [Online]. Available: https://hibit2022.ims.metu.edu.tr/.