3D Anchor-Free Lesion Detector on Computed Tomography Scans
This work addresses the time-consuming task of lesion detection for physicians and radiologists, presenting an incremental improvement over existing CNN-based detectors.
The paper tackles the problem of detecting lesions in 3D CT scans by proposing an anchor-free detector that formalizes lesions as single keypoints, resulting in considerable gains in accuracy and inference speed compared to anchor-based methods.
Lesions are injuries and abnormal tissues in the human body. Detecting lesions in 3D Computed Tomography (CT) scans can be time-consuming even for very experienced physicians and radiologists. In recent years, CNN based lesion detectors have demonstrated huge potentials. Most of current state-of-the-art lesion detectors employ anchors to enumerate all possible bounding boxes with respect to the dataset in process. This anchor mechanism greatly improves the detection performance while also constraining the generalization ability of detectors. In this paper, we propose an anchor-free lesion detector. The anchor mechanism is removed and lesions are formalized as single keypoints. By doing so, we witness a considerable performance gain in terms of both accuracy and inference speed compared with the anchor-based baseline