🤖 AI Summary
To address the need for real-time multi-organ localization in CT imaging, this paper proposes a sparse spatial classification method that bypasses full-image segmentation. The method employs large-receptive-field modeling and a gridded sliding-query mechanism to directly predict organ class labels at arbitrary 3D coordinates, while optionally generating quasi-segmentation maps. Its core innovation lies in the first application of a lightweight deep convolutional classifier for high-accuracy organ localization—achieving millisecond-level per-point inference speed (substantially faster than state-of-the-art segmentation models) while retaining scalable segmentation capability. By eliminating the computational redundancy inherent in conventional segmentation pipelines, the approach maintains clinically acceptable accuracy (Dice > 0.85) and enables real-time, interactive diagnostic applications. This work establishes an efficient new paradigm for automated medical image analysis.
📝 Abstract
Organ segmentation is a fundamental task in medical imaging since it is useful for many clinical automation pipelines. However, some tasks do not require full segmentation. Instead, a classifier can identify the selected organ without segmenting the entire volume. In this study, we demonstrate a classifier based method to obtain organ labels in real time by using a large context size with a sparse data sampling strategy. Although our method operates as an independent classifier at query locations, it can generate full segmentations by querying grid locations at any resolution, offering faster performance than segmentation algorithms. We compared our method with existing segmentation techniques, demonstrating its superior runtime potential for practical applications in medical imaging.