Development of Conditional Random Field Insert for UNet-based Zonal Prostate Segmentation on T2-Weighted MRI

TL;DR

This study introduces a novel conditional random field insert (CRFI) combined with SegNet to enhance the accuracy and boundary clarity of prostate segmentation in T2-weighted MRI, outperforming traditional UNet models.

Contribution

The paper proposes a new CRFI methodology integrated with SegNet, improving boundary delineation and robustness in prostate MRI segmentation over existing UNet-based approaches.

Findings

SegNet+CRFI achieved high Dice coefficients (up to 0.89) for prostate zones.

CRFI corrected segmentation errors, producing smoother, more consistent boundaries.

The proposed method outperformed baseline UNet and SegNet models in accuracy and robustness.

Abstract

Purpose: A conventional 2D UNet convolutional neural network (CNN) architecture may result in ill-defined boundaries in segmentation output. Several studies imposed stronger constraints on each level of UNet to improve the performance of 2D UNet, such as SegNet. In this study, we investigated 2D SegNet and a proposed conditional random field insert (CRFI) for zonal prostate segmentation from clinical T2-weighted MRI data. Methods: We introduced a new methodology that combines SegNet and CRFI to improve the accuracy and robustness of the segmentation. CRFI has feedback connections that encourage the data consistency at multiple levels of the feature pyramid. On the encoder side of the SegNet, the CRFI combines the input feature maps and convolution block output based on their spatial local similarity, like a trainable bilateral filter. For all networks, 725 2D images (i.e., 29 MRI cases) were used in training; while, 174 2D images (i.e., 6 cases) were used in testing. Results: The SegNet with CRFI achieved the relatively high Dice coefficients (0.76, 0.84, and 0.89) for the peripheral zone, central zone, and whole gland, respectively. Compared with UNet, the SegNet+CRFIs segmentation has generally higher Dice score and showed the robustness in determining the boundaries of anatomical structures compared with the SegNet or UNet segmentation. The SegNet with a CRFI at the end showed the CRFI can correct the segmentation errors from SegNet output, generating smooth and consistent segmentation for the prostate. Conclusion: UNet based deep neural networks demonstrated in this study can perform zonal prostate segmentation, achieving high Dice coefficients compared with those in the literature. The proposed CRFI method can reduce the fuzzy boundaries that affected the segmentation performance of baseline UNet and SegNet models.