Explainable concept mappings underlying deep learning brain disease classification

Motivation: While recent studies show high accuracy in the classification of Alzheimer’s disease using deep neural networks, the underlying learned concepts have not been investigated. Goal(s): To systematically identify the concepts learned by the deep neural network for model validation. Approach: Using R2* maps we separated Alzheimer's patients (n=117) from healthy controls (n=219) by using a deep neural network and systematically investigated the learned concepts using Concept Relevance Propagation (CRP). Results: In line with established histological findings, highly relevant concepts were primarily found in and adjacent to the basal ganglia. Impact: The identification of concepts learned by deep neural networks for disease classification enables validation of the models and improves reliability.

Authors

Publication Details

Journal
Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition
Published
2024-11-26
DOI
https://doi.org/10.58530/2024/0507
Primary Topic
Machine Learning in Healthcare
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Explainable concept mappings underlying deep learning brain disease classification

Sebastian Lapuschkin, Reduan Achtibat, Wojciech Samek, Christian Langkammer et al.
Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition
Machine Learning in Healthcare
article

Explainable concept mappings underlying deep learning brain disease classification

Sebastian Lapuschkin, Reduan Achtibat, Wojciech Samek, Christian Langkammer, Anna Damulina, Reinhold Schmidt, Frederik Pahde, Stefan Ropele, Christian Tinauer, Maximilian Sackl, Maximilian Dreyer
article en

Abstract

Motivation: While recent studies show high accuracy in the classification of Alzheimer’s disease using deep neural networks, the underlying learned concepts have not been investigated. Goal(s): To systematically identify the concepts learned by the deep neural network for model validation. Approach: Using R2* maps we separated Alzheimer's patients (n=117) from healthy controls (n=219) by using a deep neural network and systematically investigated the learned concepts using Concept Relevance Propagation (CRP). Results: In line with established histological findings, highly relevant concepts were primarily found in and adjacent to the basal ganglia. Impact: The identification of concepts learned by deep neural networks for disease classification enables validation of the models and improves reliability.

Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition
Openalex Percentile: Top 81%
Machine Learning in Healthcare
DOISourceOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Explainable concept mappings underlying deep learning brain disease classification — Sebastian Lapuschkin, Reduan Achtibat, et al. · Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition (2024) | TGRS Research Map | TGRS