Brain Glioma Segmentation Model based on Multi-scale Information Fusion

Predicting histological grading and genetic biomarkers of gliomas using medical imaging and deep learning is a challenging task that is important for personalised treatment and survival prediction. Previous studies require physicians to manually outline the lesion area, which consumes a lot of human resources. Therefore, in this paper, we propose a brain glioma segmentation model based on multi-scale information fusion. Our model is based on the U-Net architecture, which incorporates the use of a multi-scale information fusion module. Compared to the original network model, our approach shows a higher lesion focus and shows strong potential for practical applications.