Self-Guided Multiple Instance Learning for Weakly Supervised Disease Classification and Localization in Chest Radiographs

Abstract

The lack of fine-grained annotations hinders the deployment of automated diagnosis systems, which require human-interpretable justification for their decision process. In this paper, we address the problem of weakly supervised identification and localization of abnormalities in chest radiographs. To that end, we introduce a novel loss function for training convolutional neural networks increasing the \emph{localization confidence} and assisting the overall \emph{disease identification}. The loss leverages both image- and patch-level predictions to generate auxiliary supervision. Rather than forming strictly binary from the predictions as done in previous loss formulations, we create targets in a more customized manner, which allows the loss to account for possible misclassification. We show that the supervision provided within the proposed learning scheme leads to better performance and more precise predictions on prevalent datasets for multiple-instance learning as well as on the NIH~ChestX-Ray14 benchmark for disease recognition than previously used losses.

References in corpus (15)

• YOLOv3: An Incremental Improvement
• ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases
• CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning
• PadChest: A large chest x-ray image dataset with multi-label annotated reports
• Attention-based Deep Multiple Instance Learning
• Evaluate the Malignancy of Pulmonary Nodules Using the 3D Deep Leaky Noisy-or Network
• Learning to diagnose from scratch by exploiting dependencies among labels
• Clinically Accurate Chest X-Ray Report Generation
• Weakly Supervised Deep Learning for Thoracic Disease Classification and Localization on Chest X-rays
• ChestNet: A Deep Neural Network for Classification of Thoracic Diseases on Chest Radiography
• Weakly Supervised Medical Diagnosis and Localization from Multiple Resolutions
• Sound Event Detection and Time-Frequency Segmentation from Weakly Labelled Data
• Data-Efficient Weakly Supervised Learning for Low-Resource Audio Event Detection Using Deep Learning
• Multi-task Learning for Chest X-ray Abnormality Classification on Noisy Labels
• Vispi: Automatic Visual Perception and Interpretation of Chest X-rays