Learning Generic Lung Ultrasound Biomarkers for Decoupling Feature Extraction from Downstream Tasks
This work addresses the problem of reducing annotation and training costs for medical imaging tasks, particularly in lung ultrasound, though it is incremental as it builds on existing feature learning methods.
The paper tackles the high cost of training end-to-end neural networks for lung ultrasound tasks by proposing a decoupled approach that learns generic features from biomarker classification, which can then be used for various downstream clinical tasks with comparable accuracy to end-to-end models.
Contemporary artificial neural networks (ANN) are trained end-to-end, jointly learning both features and classifiers for the task of interest. Though enormously effective, this paradigm imposes significant costs in assembling annotated task-specific datasets and training large-scale networks. We propose to decouple feature learning from downstream lung ultrasound tasks by introducing an auxiliary pre-task of visual biomarker classification. We demonstrate that one can learn an informative, concise, and interpretable feature space from ultrasound videos by training models for predicting biomarker labels. Notably, biomarker feature extractors can be trained from data annotated with weak video-scale supervision. These features can be used by a variety of downstream Expert models targeted for diverse clinical tasks (Diagnosis, lung severity, S/F ratio). Crucially, task-specific expert models are comparable in accuracy to end-to-end models directly trained for such target tasks, while being significantly lower cost to train.