Attentive CT Lesion Detection Using Deep Pyramid Inference with Multi-Scale Booster
This work addresses a bottleneck in medical diagnosis by improving lesion detection accuracy for CT scans, though it appears incremental as it builds upon existing CNN and FPN frameworks.
The paper tackled the problem of detecting lesions at vastly different scales in CT scans by proposing a Multi-Scale Booster with attention modules integrated into a Feature Pyramid Network, achieving superior performance against state-of-the-art methods on the DeepLesion benchmark dataset.
Accurate lesion detection in computer tomography (CT) slices benefits pathologic organ analysis in the medical diagnosis process. More recently, it has been tackled as an object detection problem using the Convolutional Neural Networks (CNNs). Despite the achievements from off-the-shelf CNN models, the current detection accuracy is limited by the inability of CNNs on lesions at vastly different scales. In this paper, we propose a Multi-Scale Booster (MSB) with channel and spatial attention integrated into the backbone Feature Pyramid Network (FPN). In each pyramid level, the proposed MSB captures fine-grained scale variations by using Hierarchically Dilated Convolutions (HDC). Meanwhile, the proposed channel and spatial attention modules increase the network's capability of selecting relevant features response for lesion detection. Extensive experiments on the DeepLesion benchmark dataset demonstrate that the proposed method performs superiorly against state-of-the-art approaches.