CNN-ViT-Based Model Incorporating a Voting-Based Ensemble Method with Future Fusion for Multiple Sclerosis Classification

Background/Objectives: Multiple Sclerosis (MS) is a chronic neurodegenerative disease affecting the central nervous system and significantly reducing the quality of life. Early detection of the disease is critical for slowing its progression and developing appropriate treatment plans. In recent years, artificial intelligence and deep learning methods have begun to play a significant role in clinical decision support systems, particularly due to their high accuracy rates in the analysis of Magnetic Resonance (MR) images. Methods: In this study, a hybrid model was developed for the automatic classification of MS. Two different architectures were used for feature extraction in the developed model. These architectures were selected from among the most successful architectures out of 10 different Convolutional Neural Network (CNN) and Vision Transformer (ViT) architectures. Six different classifiers were used to classify the feature maps obtained in the proposed model and pre-trained models. The application of the VOTE operation in the final stage of the proposed model improved its performance. Results: The proposed model yielded competitive results compared to similar studies in the literature and the CNN and ViT architectures used in this study. The accuracy rate achieved with the proposed hybrid model was 97.28%. Conclusions: These results indicate that the proposed model has the potential to support computer-aided MS detection and may provide a useful framework for clinical decision support systems.

Authors

Institutions

Publication Details

Journal
Diagnostics
Published
2026-11-18
DOI
https://doi.org/10.3390/diagnostics16223867
Primary Topic
Multiple Sclerosis Research Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

CNN-ViT-Based Model Incorporating a Voting-Based Ensemble Method with Future Fusion for Multiple Sclerosis Classification

Harun Bingöl, Mücahit Karaduman, Muhammed A. Yıldırım, Sibel ÇIPLAK
4 links
Diagnostics
Multiple Sclerosis Research Studies
article

CNN-ViT-Based Model Incorporating a Voting-Based Ensemble Method with Future Fusion for Multiple Sclerosis Classification

Harun Bingöl, Mücahit Karaduman, Muhammed A. Yıldırım, Sibel ÇIPLAK
article en

Abstract

Background/Objectives: Multiple Sclerosis (MS) is a chronic neurodegenerative disease affecting the central nervous system and significantly reducing the quality of life. Early detection of the disease is critical for slowing its progression and developing appropriate treatment plans. In recent years, artificial intelligence and deep learning methods have begun to play a significant role in clinical decision support systems, particularly due to their high accuracy rates in the analysis of Magnetic Resonance (MR) images. Methods: In this study, a hybrid model was developed for the automatic classification of MS. Two different architectures were used for feature extraction in the developed model. These architectures were selected from among the most successful architectures out of 10 different Convolutional Neural Network (CNN) and Vision Transformer (ViT) architectures. Six different classifiers were used to classify the feature maps obtained in the proposed model and pre-trained models. The application of the VOTE operation in the final stage of the proposed model improved its performance. Results: The proposed model yielded competitive results compared to similar studies in the literature and the CNN and ViT architectures used in this study. The accuracy rate achieved with the proposed hybrid model was 97.28%. Conclusions: These results indicate that the proposed model has the potential to support computer-aided MS detection and may provide a useful framework for clinical decision support systems.

DiagnosticsVol. 16(22)
Fırat University (TR), Malatya Turgut Özal Üniversitesi (TR), Turgut Özal University (TR)
Good health and well-being
Multiple Sclerosis Research Studies
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.

Laika access coming soon