Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Few-shot learning for COVID-19 Chest X-Ray Classification with Imbalanced Data: An Inter vs. Intra Domain Study

Alejandro Galán-Cuenca, Antonio Javier Gallego, Marcelo Saval-Calvo, Antonio Pertusa

TL;DR

This work proposes a methodology based on Siamese neural networks in which a series of techniques are integrated to mitigate the effects of data scarcity and distribution imbalance, and concludes that the introduced techniques offer promising improvements over the baseline in almost all cases.

Abstract

Medical image datasets are essential for training models used in computer-aided diagnosis, treatment planning, and medical research. However, some challenges are associated with these datasets, including variability in data distribution, data scarcity, and transfer learning issues when using models pre-trained from generic images. This work studies the effect of these challenges at the intra- and inter-domain level in few-shot learning scenarios with severe data imbalance. For this, we propose a methodology based on Siamese neural networks in which a series of techniques are integrated to mitigate the effects of data scarcity and distribution imbalance. Specifically, different initialization and data augmentation methods are analyzed, and four adaptations to Siamese networks of solutions to deal with imbalanced data are introduced, including data balancing and weighted loss, both separately and combined, and with a different balance of pairing ratios. Moreover, we also assess the inference process considering four classifiers, namely Histogram, $k$NN, SVM, and Random Forest. Evaluation is performed on three chest X-ray datasets with annotated cases of both positive and negative COVID-19 diagnoses. The accuracy of each technique proposed for the Siamese architecture is analyzed separately and their results are compared to those obtained using equivalent methods on a state-of-the-art CNN. We conclude that the introduced techniques offer promising improvements over the baseline in almost all cases, and that the selection of the technique may vary depending on the amount of data available and the level of imbalance.

Few-shot learning for COVID-19 Chest X-Ray Classification with Imbalanced Data: An Inter vs. Intra Domain Study

TL;DR

This work proposes a methodology based on Siamese neural networks in which a series of techniques are integrated to mitigate the effects of data scarcity and distribution imbalance, and concludes that the introduced techniques offer promising improvements over the baseline in almost all cases.

Abstract

Medical image datasets are essential for training models used in computer-aided diagnosis, treatment planning, and medical research. However, some challenges are associated with these datasets, including variability in data distribution, data scarcity, and transfer learning issues when using models pre-trained from generic images. This work studies the effect of these challenges at the intra- and inter-domain level in few-shot learning scenarios with severe data imbalance. For this, we propose a methodology based on Siamese neural networks in which a series of techniques are integrated to mitigate the effects of data scarcity and distribution imbalance. Specifically, different initialization and data augmentation methods are analyzed, and four adaptations to Siamese networks of solutions to deal with imbalanced data are introduced, including data balancing and weighted loss, both separately and combined, and with a different balance of pairing ratios. Moreover, we also assess the inference process considering four classifiers, namely Histogram, NN, SVM, and Random Forest. Evaluation is performed on three chest X-ray datasets with annotated cases of both positive and negative COVID-19 diagnoses. The accuracy of each technique proposed for the Siamese architecture is analyzed separately and their results are compared to those obtained using equivalent methods on a state-of-the-art CNN. We conclude that the introduced techniques offer promising improvements over the baseline in almost all cases, and that the selection of the technique may vary depending on the amount of data available and the level of imbalance.
Paper Structure (19 sections, 5 equations, 5 figures, 7 tables)

This paper contains 19 sections, 5 equations, 5 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Siamese architecture
  4. Siamese Initialization
  5. Data augmentation
  6. Imbalanced data
  7. Inference stage
  8. Experimental setup
  9. Datasets
  10. Network architecture
  11. Metrics
  12. Results
  13. Initialization
  14. Data augmentation
  15. Dealing with Imbalanced Data
  16. ...and 4 more sections

Figures (5)

  • Figure 1: Diagram with the pipeline of the process. The proposed techniques to be studied are highlighted in yellow.
  • Figure 2: Illustrative samples from the evaluated datasets.
  • Figure 3: Graph of data augmentation. Five levels of augmented percentage are shown, from $0\%$ to $15\%$, for the four different levels of data imbalance, High to None.
  • Figure 4: Graph of pairing experimentation. Five different ratios of positive/negative pairs, and High to None data distribution cases.
  • Figure 5: Graph comparison of Siamese and CNN network architectures. The evaluation is carried out for three sizes of training sets, 100, 200, and 300 samples for the majority class, and High to None imbalance data distributions.