Computational Imaging 2

[IP6-03] Assessing U-Net and Attention-UW Net in the Segmentation of Brain Tumor Images from Magnetic Resonance Imaging

*Clara Lavita Angelina^1,2, Sunil Vyas³, Hsuan Ting Chang^1,2, Yuan Luo^3,4,5, Fu Ren Xiao³ (1. Department of Electrical Engineering, National Yunlin University of Science and Technology, 2. Graduate School of Engineering Science and Technology, National Yunlin University of Science and Technology, 3. Institution of Medical Device and Imaging, National Taiwan University, 4. YougLin Institute of Health, National Taiwan University, 5. Program for Precision Health and Intelligent Medicine, National Taiwan University)

Image segmentation in MRI is essential for diagnosis. Our study applied Attention-UW Net for brain lesion segmentation in MRI, showing its superiority over U-Net. Attention-UW Net achieved better performance (IoU: 0.80 vs. 0.75, precision: 0.86 vs. 0.85, recall: 0.83 vs. 0.70, F1 score: 0.80 vs. 0.81).