Differentiating atypical parkinsonian syndromes with hyperbolic few-shot contrastive learning.

Autor: Choi WJ; Department of Information Convergence Engineering, Pusan National University, Busan 46241, South Korea., HwangBo J; Department of Neurology, Pusan National University Yangsan Hospital, Pusan National University School of Medicine, Yangsan 50612, South Korea., Duong QA; Department of Information Convergence Engineering, Pusan National University, Busan 46241, South Korea., Lee JH; Department of Neurology, Research Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan 50612, South Korea; Medical Research Institute, Pusan National University School of Medicine, Yangsan 50612, South Korea. Electronic address: jhlee.neuro@pusan.ac.kr., Gahm JK; School of Computer Science and Engineering, Pusan National University, Busan 46241, South Korea; Center for Artificial Intelligence Research, Pusan National University, Busan 46241, South Korea. Electronic address: gahmj@pusan.ac.kr.
Jazyk: angličtina
Zdroj: NeuroImage [Neuroimage] 2024 Dec 15; Vol. 304, pp. 120940. Date of Electronic Publication: 2024 Nov 24.
DOI: 10.1016/j.neuroimage.2024.120940
Abstrakt: Differences in iron accumulation patterns have been observed in susceptibility-weighted images across different classes of atypical parkinsonian syndromes (APS). Deep learning methods have shown great potential in automatically detecting these differences. However, the models typically require extensively labeled training datasets, which are costly and pose patient privacy risks. To address the issue of limited training datasets, we propose a novel few-shot learning framework for classifying multiple system atrophy parkinsonian (MSA-P) and progressive supranuclear palsy (PSP) within the APS category using fewer data items. Our method identifies feature areas where iron accumulation patterns occur in classes other than the target classification (MSA-P vs. PSP) and enhances stability by leveraging a superior hyperbolic space embedding technique. Experimental results demonstrate significantly improved performance over conventional methods, as validated by ablation studies and visualizations.
Competing Interests: Declaration of competing interest The authors declare no conflict of interest concerning their authorship or the publication of this article.
(Copyright © 2024 The Authors. Published by Elsevier Inc. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: