CLOG-CD: Curriculum Learning Based on Oscillating Granularity of Class Decomposed Medical Image Classification

2025-05-03

IEEE Transactions on Emerging Topics in Computing

AI Summary

To address misclassification in medical image classification caused by inter-class confusion and class imbalance, this paper proposes a CNN training framework integrating curriculum learning (CL) with class decomposition. We introduce an oscillating-granularity class decomposition CL paradigm: dynamically varying class granularity (coarse → fine → coarse), coupled with a reverse curriculum schedule (descending → ascending) and a class decomposition weight transfer mechanism to enhance discriminative robustness for minority classes; additionally, an adaptive pace function and acceleration factor are incorporated to regulate training progression. Evaluated on four benchmark medical datasets—Chest X-ray (CXR), brain tumor MRI, knee X-ray, and colorectal cancer histopathology—our method achieves up to 99.45% accuracy using ResNet-50 and DenseNet-121 backbones, significantly outperforming standard training and state-of-the-art curriculum learning approaches.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Medical and Biological Imaging
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Curriculum learning strategies have been proven to be effective in various applications and have gained significant interest in the field of machine learning. It has the ability to improve the final model's performance and accelerate the training process. However, in the medical imaging domain, data irregularities can make the recognition task more challenging and usually result in misclassification between the different classes in the dataset. Class-decomposition approaches have shown promising results in solving such a problem by learning the boundaries within the classes of the data set. In this paper, we present a novel convolutional neural network (CNN) training method based on the curriculum learning strategy and the class decomposition approach, which we call CLOG-CD, to improve the performance of medical image classification. We evaluated our method on four different imbalanced medical image datasets, such as Chest X-ray (CXR), brain tumour, digital knee X-ray, and histopathology colorectal cancer (CRC). CLOG-CD utilises the learnt weights from the decomposition granularity of the classes, and the training is accomplished from descending to ascending order (i.e., anti-curriculum technique). We also investigated the classification performance of our proposed method based on different acceleration factors and pace function curricula. We used two pre-trained networks, ResNet-50 and DenseNet-121, as the backbone for CLOG-CD. The results with ResNet-50 show that CLOG-CD has the ability to improve classification performance with an accuracy of 96.08% for the CXR dataset, 96.91% for the brain tumour dataset, 79.76% for the digital knee X-ray, and 99.17% for the CRC dataset, compared to other training strategies. In addition, with DenseNet-121, CLOG-CD has achieved 94.86%, 94.63%, 76.19%, and 99.45% for CXR, brain tumour, digital knee X-ray, and CRC datasets, respectively

Problem

• Improving medical image classification accuracy using curriculum learning
• Addressing data irregularities in medical imaging via class decomposition
• Enhancing CNN training with oscillating granularity for imbalanced datasets

Innovation

• Curriculum learning with oscillating granularity
• Class decomposition for medical image classification
• Anti-curriculum technique for training optimization