An Efficient Medical Image Classification Method Based on a Lightweight Improved ConvNeXt-Tiny Architecture

TL;DR

This paper introduces a lightweight, optimized ConvNeXt-Tiny based method for efficient and accurate medical image classification suitable for resource-constrained environments, achieving high accuracy with reduced computational load.

Contribution

It proposes structural enhancements, a dual pooling feature fusion, a novel lightweight attention module, and a custom loss function to improve classification performance and efficiency.

Findings

Achieves 89.10% accuracy within 10 epochs on CPU-only setup.

Enhances feature extraction and classification with reduced computational complexity.

Demonstrates stable convergence and improved performance in resource-limited settings.

Abstract

Intelligent analysis of medical imaging plays a crucial role in assisting clinical diagnosis. However, achieving efficient and high-accuracy image classification in resource-constrained computational environments remains challenging. This study proposes a medical image classification method based on an improved ConvNeXt-Tiny architecture. Through structural optimization and loss function design, the proposed method enhances feature extraction capability and classification performance while reducing computational complexity. Specifically, the method introduces a dual global pooling (Global Average Pooling and Global Max Pooling) feature fusion strategy into the ConvNeXt-Tiny backbone to simultaneously preserve global statistical features and salient response information. A lightweight channel attention module, termed Squeeze-and-Excitation Vector (SEVector), is designed to improve the adaptive allocation of channel weights while minimizing parameter overhead. Additionally, a Feature Smoothing Loss is incorporated into the loss function to enhance intra-class feature consistency and suppress intra-class variance. Under CPU-only conditions (8 threads), the method achieves a maximum classification accuracy of 89.10% on the test set within 10 training epochs, exhibiting a stable convergence trend in loss values. Experimental results demonstrate that the proposed method effectively improves medical image classification performance in resource-limited settings, providing a feasible and efficient solution for the deployment and promotion of medical imaging analysis models.