بخش بندي تصاوير MRI تومورهاي مغزي با استفاده از معماري تغيير يافته U-Net

Detecting brain tumor and their boundaries is crucial for timely diagnosis and reducing patient mortality. Specialists determine these borders, but tumor swelling, tissue differences, and variability in shape and size across patients often result in irregular and indeterminate tumor borders. In recent years, the encoder-decoder framework, including well-known architectures like U-Net, has shown promising results in medical image segmentation tasks to overcome these obstacles However, enhancing the quality of tumor segmentation and delineating precise tumor boundaries are immense challenges for effective diagnosis and treatment planning. In this research, a new Semi-Wavelet Attention (SWATT) module is utilized within the U-Net model to enhance the quality of tumor segmentation, resulting in the modified model referred to as SWATT U-Net. SWATT incorporates a semi-wavelet layer and its inverse, enabling multi-resolution analysis of input images by scaling the original image to focus on specific parts. The SWATT U-Net model is trained on the BraTS2020 dataset. The Mean Intersection over unio‎n (IoU) criterion is used for eva‎luation. The results demonstrate an improvement, with the Mean-IoU increasing from 82.82% to 87.16%. Moreover, SWATT is applied to PSPNET, ResNet, SegNet, and DeepLabV3 models. The modified models showcase significant Mean-IoU enhancements ranging from 14.53% to 20.88%, compared to the models without SWATT. Notably, the DeepLabV3 network exhibits the highest improvement of 20.88%. Finally, the statistical analysis using t-tests confirms the superiority of the modified models incorporating the SWATT attention mechanism, leading to improved segmentation results.

6