Multi-Task Lung Nodule Detection in Chest Radiographs with a Dual Head Network
This work addresses a critical challenge for thoracic radiologists in improving lung cancer screening, but it appears incremental as it builds on existing multi-task detection approaches with specific enhancements.
The paper tackles the problem of missed lung nodule detection in chest radiographs by proposing a multi-task algorithm that predicts both global nodule presence and local locations using a Dual Head Network, achieving favorable detection performance compared to conventional methods and further enhanced by a novel Dual Head Augmentation strategy.
Lung nodules can be an alarming precursor to potential lung cancer. Missed nodule detections during chest radiograph analysis remains a common challenge among thoracic radiologists. In this work, we present a multi-task lung nodule detection algorithm for chest radiograph analysis. Unlike past approaches, our algorithm predicts a global-level label indicating nodule presence along with local-level labels predicting nodule locations using a Dual Head Network (DHN). We demonstrate the favorable nodule detection performance that our multi-task formulation yields in comparison to conventional methods. In addition, we introduce a novel Dual Head Augmentation (DHA) strategy tailored for DHN, and we demonstrate its significance in further enhancing global and local nodule predictions.