Abdominal Multi-Organ Segmentation Based on nnUNet

In recent years, medical image segmentation has been a critical task in the field of medical image analysis. Accurate segmentation of medical images is essential for many clinical applications such as diagnosis, treatment planning, and monitoring. In this study, we use the nnUNet model, a state-of-the-art deep learning architecture, to achieve very good results on abdominal multi-organ segmentation. Our dataset is selected from AMOS, which is a large-scale and diverse benchmark collected from abdominal multi-organ segmentation in real clinical scenarios, containing a total of 160 training data and 40 testing data. The nnUNet model demonstrated high accuracy, achieving excellent segmentation results on the dataset, covering 15 abdominal organs segmentation tasks including the spleen, right kidney, left kidney, gallbladder, etc.