Deep Transfer Learning for the Multilabel Classification of Chest X-ray Images

Abstract

Chest X-ray (CXR) is widely used to diagnose conditions affecting the chest, its contents, and its nearby structures. In this study, we used a private data set containing 1630 CXR images with disease labels; most of the images were disease-free, but the others contained multiple sites of abnormalities. Here, we used deep convolutional neural network (CNN) models to extract feature representations and to identify possible diseases in these images. We also used transfer learning combined with large open-source image data sets to resolve the problems of insufficient training data and optimize the classification model. The effects of different approaches of reusing pretrained weights (model finetuning and layer transfer), source data sets of different sizes and similarity levels to the target data (ImageNet, ChestX-ray, and CheXpert), methods integrating source data sets into transfer learning (initiating, concatenating, and co-training), and backbone CNN models (ResNet50 and DenseNet121) on transfer learning were also assessed. The results demonstrated that transfer learning applied with the model finetuning approach typically afforded better prediction models. When only one source data set was adopted, ChestX-ray performed better than CheXpert; however, after ImageNet initials were attached, CheXpert performed better. ResNet50 performed better in initiating transfer learning, whereas DenseNet121 performed better in concatenating and co-training transfer learning. Transfer learning with multiple source data sets was preferable to that with a source data set. Overall, transfer learning can further enhance prediction capabilities and reduce computing costs for CXR images.

A real-time automated bone age assessment system based on the RUS-CHN method

“…Huang GH et al. discovered that using migration learning in chest X-rays can enhance prediction capabilities and reduce computing costs ( 27 ).…”

Predicting intensive care need for COVID-19 patients using deep learning on chest radiography

“…33 In these situations, deep learning models pretrained on nonmedical image datasets or medical image datasets from either a different imaging modality or same imaging modality but for different clinical tasks are fine-tuned with a relatively small medical imaging dataset for clinical decision-making tasks. [33][34][35][36][37][38][39] For example, Antropova et al 34 applied transfer learning on three different imaging modalities to extract deep features and fused them with human engineered radiomic features for the diagnostic classification of breast tumors, with results demonstrating statistically significant improved classification performance as compared to previous developed computer-aided diagnosis methods. Huang et al 35 applied deep transfer learning to identify possible disease on CXR images for multilabel classification task with improved prediction capacities.…”

“…[33][34][35][36][37][38][39] For example, Antropova et al 34 applied transfer learning on three different imaging modalities to extract deep features and fused them with human engineered radiomic features for the diagnostic classification of breast tumors, with results demonstrating statistically significant improved classification performance as compared to previous developed computer-aided diagnosis methods. Huang et al 35 applied deep transfer learning to identify possible disease on CXR images for multilabel classification task with improved prediction capacities. Samala et al 36 performed a multi-stage transfer learning for the classification of malignant and benign masses in digital breast tomosynthesis images and reported improved classification performance.…”

“…applied transfer learning on three different imaging modalities to extract deep features and fused them with human engineered radiomic features for the diagnostic classification of breast tumors, with results demonstrating statistically significant improved classification performance as compared to previous developed computer-aided diagnosis methods. Huang et al 35 . applied deep transfer learning to identify possible disease on CXR images for multilabel classification task with improved prediction capacities.…”

Editorial on Special Issue “Artificial Intelligence in Image-Based Screening, Diagnostics, and Clinical Care of Cardiopulmonary Diseases”

“…In another study, Huang et al [ 23 ] evaluated the gains achieved through transfer learning in a multi-label CXR classification task. They used a private CXR collection containing multiple abnormalities including aortic sclerosis/calcification, arterial curvature, consolidations, pulmonary fibrosis, enlarged hilar shadows, scoliosis, cardiomegaly, and intercostal pleural thickening, etc.…”