NeuroMoE: A Transformer-Based Mixture-of-Experts Framework for Multi-Modal Neurological Disorder Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Raza, Wajih Hassan, Shah, Aamir Bader, Wen, Yu, Shen, Yidan, Lemus, Juan Diego Martinez, Schiess, Mya Caryn, Ellmore, Timothy Michael, Hu, Renjie, Fu, Xin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908411450359808
author Raza, Wajih Hassan
Shah, Aamir Bader
Wen, Yu
Shen, Yidan
Lemus, Juan Diego Martinez
Schiess, Mya Caryn
Ellmore, Timothy Michael
Hu, Renjie
Fu, Xin
author_facet Raza, Wajih Hassan
Shah, Aamir Bader
Wen, Yu
Shen, Yidan
Lemus, Juan Diego Martinez
Schiess, Mya Caryn
Ellmore, Timothy Michael
Hu, Renjie
Fu, Xin
contents The integration of multi-modal Magnetic Resonance Imaging (MRI) and clinical data holds great promise for enhancing the diagnosis of neurological disorders (NDs) in real-world clinical settings. Deep Learning (DL) has recently emerged as a powerful tool for extracting meaningful patterns from medical data to aid in diagnosis. However, existing DL approaches struggle to effectively leverage multi-modal MRI and clinical data, leading to suboptimal performance. To address this challenge, we utilize a unique, proprietary multi-modal clinical dataset curated for ND research. Based on this dataset, we propose a novel transformer-based Mixture-of-Experts (MoE) framework for ND classification, leveraging multiple MRI modalities-anatomical (aMRI), Diffusion Tensor Imaging (DTI), and functional (fMRI)-alongside clinical assessments. Our framework employs transformer encoders to capture spatial relationships within volumetric MRI data while utilizing modality-specific experts for targeted feature extraction. A gating mechanism with adaptive fusion dynamically integrates expert outputs, ensuring optimal predictive performance. Comprehensive experiments and comparisons with multiple baselines demonstrate that our multi-modal approach significantly enhances diagnostic accuracy, particularly in distinguishing overlapping disease states. Our framework achieves a validation accuracy of 82.47\%, outperforming baseline methods by over 10\%, highlighting its potential to improve ND diagnosis by applying multi-modal learning to real-world clinical data.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14970
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle NeuroMoE: A Transformer-Based Mixture-of-Experts Framework for Multi-Modal Neurological Disorder Classification
Raza, Wajih Hassan
Shah, Aamir Bader
Wen, Yu
Shen, Yidan
Lemus, Juan Diego Martinez
Schiess, Mya Caryn
Ellmore, Timothy Michael
Hu, Renjie
Fu, Xin
Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
The integration of multi-modal Magnetic Resonance Imaging (MRI) and clinical data holds great promise for enhancing the diagnosis of neurological disorders (NDs) in real-world clinical settings. Deep Learning (DL) has recently emerged as a powerful tool for extracting meaningful patterns from medical data to aid in diagnosis. However, existing DL approaches struggle to effectively leverage multi-modal MRI and clinical data, leading to suboptimal performance. To address this challenge, we utilize a unique, proprietary multi-modal clinical dataset curated for ND research. Based on this dataset, we propose a novel transformer-based Mixture-of-Experts (MoE) framework for ND classification, leveraging multiple MRI modalities-anatomical (aMRI), Diffusion Tensor Imaging (DTI), and functional (fMRI)-alongside clinical assessments. Our framework employs transformer encoders to capture spatial relationships within volumetric MRI data while utilizing modality-specific experts for targeted feature extraction. A gating mechanism with adaptive fusion dynamically integrates expert outputs, ensuring optimal predictive performance. Comprehensive experiments and comparisons with multiple baselines demonstrate that our multi-modal approach significantly enhances diagnostic accuracy, particularly in distinguishing overlapping disease states. Our framework achieves a validation accuracy of 82.47\%, outperforming baseline methods by over 10\%, highlighting its potential to improve ND diagnosis by applying multi-modal learning to real-world clinical data.
title NeuroMoE: A Transformer-Based Mixture-of-Experts Framework for Multi-Modal Neurological Disorder Classification
topic Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2506.14970