Automated Kidney MRI Segmentation using connected U-Net Architecture

Polycystic kidney syndrome, is a genetic disease in which the renal tubules become structurally abnormal, resulting in the outgrowth of multiple cysts and a decline in renal function. Thus, a crucial step in enhancing the quality of life for cancer patients is treatment planning. Although Magnetic Resonance Imaging (MRI) is a popular imaging method for evaluating these tumours, the volume of data it generates makes it difficult to manually segment the images in a reasonable length of time, which restricts the use of precise quantitative assessments in clinical practise. The enormous spatial and structural heterogeneity among brain tumours makes automatic segmentation a difficult task, hence effective and automatic segmentation methods are needed. UNet and its variants are one of the most advanced models for medical image segmentation, and they performed well on MRI images. So, in this paper we designed an automatic Kidney segmentation approach using Connected-UNets, which connects two UNets using additional modified skip connections. To highlight the contextual information within the encoder-decoder network design, we integrate Atrous Spatial Pyramid Pooling (ASPP) in the two conventional UNets. We also apply the proposed architecture on the Attention UNet (AUNet) and the Residual UNet (ResUNet). To evaluate the proposed model the dataset obtained from Myo clinic was considered. The experimental results give Dice Coefficient for the three architectures Connected UNets, Connected AUNets, and Connected ResUNets as 93.36%, 93.52 %, and 94.13%, and IoU score as 85.75%, 86.01%, and 87.63% respectively.