BROWSE
Detail
Conference Paper
SSNet: Synergistic Segmentation of Brain MRI Scans using nnUNetv2 and SAM-track
Cited 0 time in 
Share 
Share
- Authors
- Yong Eun Jang, Kwang-Ju Kim, In-su Jang, Gwang Lee
- Issue Date
- 2024-02
- Citation
- International Conference on Artificial Intelligence in Information and Communication (ICAIIC) 2024, pp.614-616
- Language
- English
- Type
- Conference Paper
- Abstract
- Segmenting lesions using brain MRI (Magnetic Resonance Imaging) images is a method that can help doctors determine the NIHSS (National Institutes of Health Stroke Scale) score. Therefore, we proposed the SSNet model to segment brain MRI images using a foundation model. Our SSNet model was constructed using nnUNetv2, a model with good performance in medical image segmentation, and SAM-Tracker, a foundation model. Our model's IoU (Intersection over Union) is 0.463, and Dice is 0.611, and when only nnUNetv2 is used, the IoU is 0.569, and Dice is 0.702. Although the performance is currently low, the foundation model is evolving, so it is expected that the model we proposed will be able to utilize it in brain MRI images end-to-end.
- KSP Keywords
- Brain MRI images, End to End(E2E), Magnetic Resonance imaging(MRI), Magnetic resonance(MR), medical image segmentation