Chest X-ray Classification using Deep Convolution Models on Low-resolution images with Uncertain Labels

2025-04-12

AI Summary

To address pathological classification of low-resolution, label-uncertain chest X-ray (CXR) images in telemedicine, this paper proposes a robust multi-view deep learning framework. First, a random label flipping strategy is introduced to mitigate annotation noise. Second, a dual-view (anteroposterior and lateral) multi-label ensemble model is developed, enhanced with L2 regularization and comprehensive data augmentation to improve generalizability. Third, we conduct the first systematic evaluation of resolution impact on classification performance: on a CheXpert subset (5-pathology), 256×256 resolution proves clinically viable—achieving 3% higher accuracy than high-resolution baselines for Cardiomegaly, Consolidation, and Edema. Class activation maps (CAMs) confirm that the model attends to anatomically relevant regions. This work establishes a new paradigm for trustworthy AI-assisted diagnosis in resource-constrained settings.

Technology Category

Machine Learning: Multi-instance/Multi-view LearningComputer Vision: Medical and Biological ImagingIntelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

Responsible Web: Machine-in-the-loop, human agency and autonomySearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingWeb Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Deep Convolutional Neural Networks have consistently proven to achieve state-of-the-art results on a lot of imaging tasks over the past years' majority of which comprise of high-quality data. However, it is important to work on low-resolution images since it could be a cheaper alternative for remote healthcare access where the primary need of automated pathology identification models occurs. Medical diagnosis using low-resolution images is challenging since critical details may not be easily identifiable. In this paper, we report classification results by experimenting on different input image sizes of Chest X-rays to deep CNN models and discuss the feasibility of classification on varying image sizes. We also leverage the noisy labels in the dataset by proposing a Randomized Flipping of labels techniques. We use an ensemble of multi-label classification models on frontal and lateral studies. Our models are trained on 5 out of the 14 chest pathologies of the publicly available CheXpert dataset. We incorporate techniques such as augmentation, regularization for model improvement and use class activation maps to visualize the neural network's decision making. Comparison with classification results on data from 200 subjects, obtained on the corresponding high-resolution images, reported in the original CheXpert paper, has been presented. For pathologies Cardiomegaly, Consolidation and Edema, we obtain 3% higher accuracy with our model architecture.

Problem

• Classifying chest X-rays using deep learning on low-resolution images
• Improving accuracy with noisy labels via randomized flipping techniques
• Evaluating model performance across varying image sizes and pathologies

Innovation

• Deep CNN for low-resolution X-ray classification
• Randomized Flipping for noisy label handling
• Ensemble multi-label models with activation maps