Generalization of Deep Convolutional Neural Networks -- A Case-study on Open-source Chest Radiographs

TL;DR

This study investigates the generalization capabilities of deep convolutional neural networks in medical imaging, showing that training on diverse datasets enhances external performance in chest radiograph analysis.

Contribution

Demonstrates that increasing training data heterogeneity improves DCNN generalization for chest pathology detection across different datasets.

Findings

Internal performance exceeds external performance.

Training on mixed datasets improves external generalization.

Models perform better on internal than external data.

Abstract

Deep Convolutional Neural Networks (DCNNs) have attracted extensive attention and been applied in many areas, including medical image analysis and clinical diagnosis. One major challenge is to conceive a DCNN model with remarkable performance on both internal and external data. We demonstrate that DCNNs may not generalize to new data, but increasing the quality and heterogeneity of the training data helps to improve the generalizibility factor. We use InceptionResNetV2 and DenseNet121 architectures to predict the risk of 5 common chest pathologies. The experiments were conducted on three publicly available databases: CheXpert, ChestX-ray14, and MIMIC Chest Xray JPG. The results show the internal performance of each of the 5 pathologies outperformed external performance on both of the models. Moreover, our strategy of exposing the models to a mix of different datasets during the training phase helps to improve model performance on the external dataset.