MedYOLO: A Medical Image Object Detection Framework
This work addresses the need for general-purpose object detection tools in 3-D medical imaging to reduce annotation time for tasks where voxel-level precision is not required, though it is incremental as it adapts existing YOLO methods to this domain.
The authors tackled the problem of reducing annotation effort in medical image analysis by developing MedYOLO, a 3-D object detection framework, which achieved high performance on medium and large-sized structures like the heart, liver, and pancreas across four datasets, but struggled with very small or rare structures.
Artificial intelligence-enhanced identification of organs, lesions, and other structures in medical imaging is typically done using convolutional neural networks (CNNs) designed to make voxel-accurate segmentations of the region of interest. However, the labels required to train these CNNs are time-consuming to generate and require attention from subject matter experts to ensure quality. For tasks where voxel-level precision is not required, object detection models offer a viable alternative that can reduce annotation effort. Despite this potential application, there are few options for general purpose object detection frameworks available for 3-D medical imaging. We report on MedYOLO, a 3-D object detection framework using the one-shot detection method of the YOLO family of models and designed for use with medical imaging. We tested this model on four different datasets: BRaTS, LIDC, an abdominal organ Computed Tomography (CT) dataset, and an ECG-gated heart CT dataset. We found our models achieve high performance on commonly present medium and large-sized structures such as the heart, liver, and pancreas even without hyperparameter tuning. However, the models struggle with very small or rarely present structures.