MRI-based prostate and dominant lesion segmentation using cascaded scoring convolutional neural network

Abstract

Purpose: Dose escalation to dominant intraprostatic lesions (DILs) is a novel treatment strategy to improve the treatment outcome of prostate radiation therapy. Treatment planning requires accurate and fast delineation of the prostate and DILs. In this study, a 3D cascaded scoring convolutional neural network is proposed to automatically segment the prostate and DILs from MRI.

Methods and materials: The proposed cascaded scoring convolutional neural network performs end-to-end segmentation by locating a region-of-interest (ROI), identifying the object within the ROI, and defining the target. A scoring strategy, which is learned to judge the segmentation quality of DIL, is integrated into cascaded convolutional neural network to solve the challenge of segmenting the irregular shapes of the DIL. To evaluate the proposed method, 77 patients who underwent MRI and PET/CT were retrospectively investigated. The prostate and DIL ground truth contours were delineated by experienced radiologists. The proposed method was evaluated with fivefold cross-validation and holdout testing.

Results: The average centroid distance, volume difference, and Dice similarity coefficient (DSC) value for prostate/DIL are 4.3 ± 7.5/3.73 ± 3.78 mm, 4.5 ± 7.9/0.41 ± 0.59 cc, and 89.6 ± 8.9/84.3 ± 11.9%, respectively. Comparable results were obtained in the holdout test. Similar or superior segmentation outcomes were seen when compared the results of the proposed method to those of competing segmentation approaches.

Conclusions: The proposed automatic segmentation method can accurately and simultaneously segment both the prostate and DILs. The intended future use for this algorithm is focal boost prostate radiation therapy.

Keywords: MRI; deep learning; prostate and dominant lesion segmentation.

Figure 1.

Schematic flow chart of the proposed algorithm. The black arrows denote the feed forward paths during both training and inference stages. The blue arrows denote the feed forward paths during inference stage.

Figure 2.

Manual (red) vs segmented (green dashed) prostate and DIL of axial MRI. From left to right are the prostate manual and segmented contours overlaid on MRI, and two DIL manual and segmented contours overlaid on MRI. The upper and bottom rows are from two representative patients.

Figure 3.

Sensitivity threshold vs reported metric values for prostate and DIL hold-out tests

Figure 4.

Combination violin and box and whisker plots for prostate and DIL cross validation

Figure 5.

Linear regression and Bland-Altman plots of the prostate and DIL for hold-out test