Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Multi-Label Classification of Thoracic Diseases using Dense Convolutional Network on Chest Radiographs

Dipkamal Bhusal, Sanjeeb Prasad Panday

TL;DR

This work tackles multi-label thoracic disease classification from chest X-rays, addressing limitations of single-label approaches. Using a DenseNet-121 backbone with weighted loss to handle 14 co-occurring pathologies and Grad-CAM based explanations, the model achieves strong per-condition performance on ChestX-ray8. Training on about 99,000 images with a patient-level data split and 300-epoch schedules yields robust AUC and reliable uncertainty estimates via confidence intervals. The results support potential clinical utility for automated triage and decision support while acknowledging the need for additional modalities, medical history, and clinician validation.

Abstract

Traditional methods of identifying pathologies in X-ray images rely heavily on skilled human interpretation and are often time-consuming. The advent of deep learning techniques has enabled the development of automated disease diagnosis systems. Still, the performance of such systems is opaque to end-users and limited to detecting a single pathology. In this paper, we propose a multi-label disease prediction model that allows the detection of more than one pathology at a given test time. We use a dense convolutional neural network (DenseNet) for disease diagnosis. Our proposed model achieved the highest AUC score of 0.896 for the condition Cardiomegaly with an accuracy of 0.826, while the lowest AUC score was obtained for Nodule, at 0.655 with an accuracy of 0.66. To build trust in decision-making, we generated heatmaps on X-rays to visualize the regions where the model paid attention to make certain predictions. Our proposed automated disease prediction model obtained highly confident high-performance metrics in multi-label disease prediction tasks.

Multi-Label Classification of Thoracic Diseases using Dense Convolutional Network on Chest Radiographs

TL;DR

This work tackles multi-label thoracic disease classification from chest X-rays, addressing limitations of single-label approaches. Using a DenseNet-121 backbone with weighted loss to handle 14 co-occurring pathologies and Grad-CAM based explanations, the model achieves strong per-condition performance on ChestX-ray8. Training on about 99,000 images with a patient-level data split and 300-epoch schedules yields robust AUC and reliable uncertainty estimates via confidence intervals. The results support potential clinical utility for automated triage and decision support while acknowledging the need for additional modalities, medical history, and clinician validation.

Abstract

Traditional methods of identifying pathologies in X-ray images rely heavily on skilled human interpretation and are often time-consuming. The advent of deep learning techniques has enabled the development of automated disease diagnosis systems. Still, the performance of such systems is opaque to end-users and limited to detecting a single pathology. In this paper, we propose a multi-label disease prediction model that allows the detection of more than one pathology at a given test time. We use a dense convolutional neural network (DenseNet) for disease diagnosis. Our proposed model achieved the highest AUC score of 0.896 for the condition Cardiomegaly with an accuracy of 0.826, while the lowest AUC score was obtained for Nodule, at 0.655 with an accuracy of 0.66. To build trust in decision-making, we generated heatmaps on X-rays to visualize the regions where the model paid attention to make certain predictions. Our proposed automated disease prediction model obtained highly confident high-performance metrics in multi-label disease prediction tasks.
Paper Structure (20 sections, 8 equations, 9 figures, 6 tables)

This paper contains 20 sections, 8 equations, 9 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Convolutional Neural Network
  4. Deep learning for thoracic disease predictions
  5. Deep learning interpretability
  6. Methodology
  7. Data preparation
  8. Addressing Data Leakage
  9. Addressing Class Imbalance
  10. Pre-processing
  11. Network Architecture
  12. Training
  13. Experiment and Results
  14. Data
  15. Evaluation Metrics
  16. ...and 5 more sections

Figures (9)

  • Figure 1: A sample of fundus photo (eye)
  • Figure 2: A sample of skin image (skin)
  • Figure 3: A sample of chest x-ray (ChestX-Ray8)
  • Figure 4: Pixel attributions or saliency maps for an image-classifier test case using Grad-CAM molnar2022
  • Figure 5: Solving class-imbalance problem
  • ...and 4 more figures