Hybrid Deep Learning Framework for Classification of Kidney CT Images: Diagnosis of Stones, Cysts, and Tumors

2025-02-05

AI Summary

This study addresses the fine-grained pathological classification of kidney CT images into four categories: normal, renal calculi, cysts, and tumors. We propose a hybrid deep learning framework integrating a pretrained ResNet-101 backbone with a lightweight, custom-designed CNN. Our key innovation is a multi-level feature fusion architecture that jointly models high-level semantic features and low-level textural details, thereby significantly improving discriminative capability for small lesions and enhancing model generalizability. The framework is trained end-to-end under supervised learning on a clinical dataset comprising 12,446 CT images. It achieves 100% test accuracy while simultaneously improving precision and recall, and reduces per-image inference time to the millisecond level—enabling real-time clinical decision support. This work establishes a transferable technical paradigm for small-object, multi-class classification in medical imaging.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Object Detection & Categorization
• Search and Optimization: Learning to Search

Application Category

• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Medical image classification is a vital research area that utilizes advanced computational techniques to improve disease diagnosis and treatment planning. Deep learning models, especially Convolutional Neural Networks (CNNs), have transformed this field by providing automated and precise analysis of complex medical images. This study introduces a hybrid deep learning model that integrates a pre-trained ResNet101 with a custom CNN to classify kidney CT images into four categories: normal, stone, cyst, and tumor. The proposed model leverages feature fusion to enhance classification accuracy, achieving 99.73% training accuracy and 100% testing accuracy. Using a dataset of 12,446 CT images and advanced feature mapping techniques, the hybrid CNN model outperforms standalone ResNet101. This architecture delivers a robust and efficient solution for automated kidney disease diagnosis, providing improved precision, recall, and reduced testing time, making it highly suitable for clinical applications.

Problem

• Classify kidney CT images
• Diagnose stones, cysts, tumors
• Enhance classification accuracy

Innovation

• Hybrid deep learning model
• Integrates ResNet101 with CNN
• Feature fusion enhances accuracy