Surgical Aggregation: A Collaborative Learning Framework for Harmonizing Distributed Medical Imaging Datasets with Diverse Tasks

Large-scale chest x-ray datasets have been curated for the detection of abnormalities using deep learning, with the potential to provide substantial benefits across many clinical applications. However, each dataset focuses only on a subset of findings that can be simultaneously present in a patient, making it challenging to train models that aggregate multiple datasets together. Therefore, data harmonization is crucial to leverage these datasets in aggregate to train clinically useful models with a complete representation of abnormalities that may occur within the thorax. To that end, we propose surgical aggregation, a collaborative learning framework for harmonizing and aggregating knowledge from distributed heterogeneous datasets with partial annotations. We evaluate surgical aggregation across synthetic and real-world heterogeneous datasets with partial annotations. Our results indicate that surgical aggregation outperforms current strategies, generalizes better, and has the potential to facilitate the development of clinically useful models even when using datasets with heterogeneous disease labels.