| Summary: | Simple Summary: Bladder cancer segmentation on MRI images is critical to determine if the cancer spread to the nearby muscles. In this study, we aimed to assess the performance of three deep learning models in outlining bladder tumors from MRI images. Using the MRI data of 53 patients, we trained Unet, MAnet, and PSPnet models to segment tumors using different loss functions and evaluated their performances. The results showed MAnet and PSPnet models performed better overall in segmenting bladder tumors, especially when they used a hybrid loss function (CE+DSC). Our findings could improve the way bladder cancer is segmented on MRI images, potentially leading to a better choice of deep learning algorithms and loss functions for future research. Background: Bladder cancer (BC) segmentation on MRI images is the first step to determining the presence of muscular invasion. This study aimed to assess the tumor segmentation performance of three deep learning (DL) models on multi-parametric MRI (mp-MRI) images. Methods: We studied 53 patients with bladder cancer. Bladder tumors were segmented on each slice of T2-weighted (T2WI), diffusion-weighted imaging/apparent diffusion coefficient (DWI/ADC), and T1-weighted contrast-enhanced (T1WI) images acquired at a 3Tesla MRI scanner. We trained Unet, MAnet, and PSPnet using three loss functions: cross-entropy (CE), dice similarity coefficient loss (DSC), and focal loss (FL). We evaluated the model performances using DSC, Hausdorff distance (HD), and expected calibration error (ECE). Results: The MAnet algorithm with the CE+DSC loss function gave the highest DSC values on the ADC, T2WI, and T1WI images. PSPnet with CE+DSC obtained the smallest HDs on the ADC, T2WI, and T1WI images. The segmentation accuracy overall was better on the ADC and T1WI than on the T2WI. The ECEs were the smallest for PSPnet with FL on the ADC images, while they were the smallest for MAnet with CE+DSC on the T2WI and T1WI. Conclusions: Compared to Unet, MAnet and PSPnet with a hybrid CE+DSC loss function displayed better performances in BC segmentation depending on the choice of the evaluation metric.
|
|---|