Capsule-ConvKAN: A Hybrid Neural Approach to Medical Image Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pituková, Laura, Sinčák, Peter, Kovács, László József, Wang, Peng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915433057091584
author Pituková, Laura
Sinčák, Peter
Kovács, László József
Wang, Peng
author_facet Pituková, Laura
Sinčák, Peter
Kovács, László József
Wang, Peng
contents This study conducts a comprehensive comparison of four neural network architectures: Convolutional Neural Network, Capsule Network, Convolutional Kolmogorov-Arnold Network, and the newly proposed Capsule-Convolutional Kolmogorov-Arnold Network. The proposed Capsule-ConvKAN architecture combines the dynamic routing and spatial hierarchy capabilities of Capsule Network with the flexible and interpretable function approximation of Convolutional Kolmogorov-Arnold Networks. This novel hybrid model was developed to improve feature representation and classification accuracy, particularly in challenging real-world biomedical image data. The architectures were evaluated on a histopathological image dataset, where Capsule-ConvKAN achieved the highest classification performance with an accuracy of 91.21%. The results demonstrate the potential of the newly introduced Capsule-ConvKAN in capturing spatial patterns, managing complex features, and addressing the limitations of traditional convolutional models in medical image classification.
format Preprint
id arxiv_https___arxiv_org_abs_2507_06417
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Capsule-ConvKAN: A Hybrid Neural Approach to Medical Image Classification
Pituková, Laura
Sinčák, Peter
Kovács, László József
Wang, Peng
Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
This study conducts a comprehensive comparison of four neural network architectures: Convolutional Neural Network, Capsule Network, Convolutional Kolmogorov-Arnold Network, and the newly proposed Capsule-Convolutional Kolmogorov-Arnold Network. The proposed Capsule-ConvKAN architecture combines the dynamic routing and spatial hierarchy capabilities of Capsule Network with the flexible and interpretable function approximation of Convolutional Kolmogorov-Arnold Networks. This novel hybrid model was developed to improve feature representation and classification accuracy, particularly in challenging real-world biomedical image data. The architectures were evaluated on a histopathological image dataset, where Capsule-ConvKAN achieved the highest classification performance with an accuracy of 91.21%. The results demonstrate the potential of the newly introduced Capsule-ConvKAN in capturing spatial patterns, managing complex features, and addressing the limitations of traditional convolutional models in medical image classification.
title Capsule-ConvKAN: A Hybrid Neural Approach to Medical Image Classification
topic Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2507.06417