Rotationally resliced 3D prostate TRUS segmentation using convex optimization with shape priors

Qiu, Wu; Yuan, Jing; Ukwatta, Eranga; Fenster, Aaron

doi:10.1118/1.4906129

Cited by 14 publications

(12 citation statements)

References 62 publications

(108 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Predicted 3D prostate segmentations were obtained by segmenting multiple 2D radial frames generated by rotation around a central axis, followed by reconstruction to a 3D surface following a reconstruction method similar to Qiu et al 11 Previous observations have noted that segmenting the prostate on slices near the apex and base of the prostate can be challenging due to boundary incompleteness, 15 so we chose to radially slice the 3D prostate image as opposed to transverse slicing in an attempt to improve segmentations at all boundaries. This choice was motivated by the experience of segmenting the prostate when the center of the gland is inplane, which typically presents as an easier image to accurately define and segment the boundaries on the left and right sides of the 2D image.…”

Section: B3 3d Reconstructionmentioning

confidence: 99%

“…In addition, we did not directly assess intra-observer variability over several time points. Inter-and intra-observer variability in endfire 3D TRUS images were previously assessed by our group, 11 and are summarized here. To assess intra-observer variability, one observer segmented 15 three-dimensional images five times each, resulting in a 3D DSC of 93.0 AE 2.1%.…”

Section: D Limitations and Future Workmentioning

confidence: 99%

“…7 Minimizing procedure time through fully or semiautomatic 3D TRUS prostate segmentation has been previously investigated. [8][9][10][11] Many methods have been shown to be promising, but have lacked clinical translation due to computational complexity, computation time, and robustness to diverse clinical datasets. Convolutional neural networks (CNNs) have received widespread attention in many image processing applications with much work investigating their accuracy and speed in medical imaging tasks.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic prostate segmentation using deep learning on clinically diverse 3D transrectal ultrasound images

et al. 2020

Self Cite

View full text Add to dashboard Cite

Purpose: Needle-based procedures for diagnosing and treating prostate cancer, such as biopsy and brachytherapy, have incorporated three-dimensional (3D) transrectal ultrasound (TRUS) imaging to improve needle guidance. Using these images effectively typically requires the physician to manually segment the prostate to define the margins used for accurate registration, targeting, and other guidance techniques. However, manual prostate segmentation is a time-consuming and difficult intraoperative process, often occurring while the patient is under sedation (biopsy) or anesthetic (brachytherapy). Minimizing procedure time with a 3D TRUS prostate segmentation method could provide physicians with a quick and accurate prostate segmentation, and allow for an efficient workflow with improved patient throughput to enable faster patient access to care. The purpose of this study was to develop a supervised deep learning-based method to segment the prostate in 3D TRUS images from different facilities, generated using multiple acquisition methods and commercial ultrasound machine models to create a generalizable algorithm for needle-based prostate cancer procedures. Methods: Our proposed method for 3D segmentation involved prediction on two-dimensional (2D) slices sampled radially around the approximate central axis of the prostate, followed by reconstruction into a 3D surface. A 2D U-Net was modified, trained, and validated using images from 84 endfire and 122 side-fire 3D TRUS images acquired during clinical biopsies and brachytherapy procedures. Modifications to the expansion section of the standard U-Net included the addition of 50% dropouts and the use of transpose convolutions instead of standard upsampling followed by convolution to reduce overfitting and improve performance, respectively. Manual contours provided the annotations needed for the training, validation, and testing datasets, with the testing dataset consisting of 20 end-fire and 20 side-fire unseen 3D TRUS images. Since predicting with 2D images has the potential to lose spatial and structural information, comparisons to 3D reconstruction and optimized 3D networks including 3D V-Net, Dense V-Net, and High-resolution 3D-Net were performed following an investigation into different loss functions. An extended selection of absolute and signed error metrics were computed, including pixel map comparisons [dice similarity coefficient (DSC), recall, and precision], volume percent differences (VPD), mean surface distance (MSD), and Hausdorff distance (HD), to assess 3D segmentation accuracy. Results: Overall, our proposed reconstructed modified U-Net performed with a median [first quartile, third quartile] absolute DSC, recall, precision, VPD, MSD, and HD of 94.1 [92.6, 94.9]%, 96.0 [93.1, 98.5]%, 93.2 [88.8, 95.4]%, 5.78 [2.49, 11.50]%, 0.89 [0.73, 1.09] mm, and 2.89 [2.37, 4.35] mm, respectively. When compared to the best-performing optimized 3D network (i.e., 3D V-Net with a Dice plus cross-entropy loss function), our proposed method performed with a significant ...

show abstract

Section: B3 3d Reconstructionmentioning

confidence: 99%

Section: D Limitations and Future Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic prostate segmentation using deep learning on clinically diverse 3D transrectal ultrasound images

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…These methods can be classified into two types, either using unsupervised or supervised models. The unsupervised methods perform tissue classification based on TRUS contour information, and shape priority . The supervised methods train a classifier using a set of training data with their associated labels (prostate or non‐prostate) and then the well‐trained classifier performs the segmentation for a newly acquired ultrasound image .…”

Section: Introductionmentioning

confidence: 99%

Ultrasound prostate segmentation based on multidirectional deeply supervised V‐Net

Lei

Tian²,

He³

et al. 2019

Medical Physics

107

View full text Add to dashboard Cite

Purpose: Transrectal ultrasound (TRUS) is a versatile and real-time imaging modality that is commonly used in image-guided prostate cancer interventions (e.g., biopsy and brachytherapy). Accurate segmentation of the prostate is key to biopsy needle placement, brachytherapy treatment planning, and motion management. Manual segmentation during these interventions is time-consuming and subject to inter-and intraobserver variation. To address these drawbacks, we aimed to develop a deep learning-based method which integrates deep supervision into a three-dimensional (3D) patch-based V-Net for prostate segmentation. Methods and materials: We developed a multidirectional deep-learning-based method to automatically segment the prostate for ultrasound-guided radiation therapy. A 3D supervision mechanism is integrated into the V-Net stages to deal with the optimization difficulties when training a deep network with limited training data. We combine a binary cross-entropy (BCE) loss and a batch-based Dice loss into the stage-wise hybrid loss function for a deep supervision training. During the segmentation stage, the patches are extracted from the newly acquired ultrasound image as the input of the well-trained network and the well-trained network adaptively labels the prostate tissue. The final segmented prostate volume is reconstructed using patch fusion and further refined through a contour refinement processing. Results: Forty-four patients' TRUS images were used to test our segmentation method. Our segmentation results were compared with the manually segmented contours (ground truth). The mean prostate volume Dice similarity coefficient (DSC), Hausdorff distance (HD), mean surface distance (MSD), and residual mean surface distance (RMSD) were 0.92 AE 0.03, 3.94 AE 1.55, 0.60 AE 0.23, and 0.90 AE 0.38 mm, respectively. Conclusion: We developed a novel deeply supervised deep learning-based approach with reliable contour refinement to automatically segment the TRUS prostate, demonstrated its clinical feasibility, and validated its accuracy compared to manual segmentation. The proposed technique could be a useful tool for diagnostic and therapeutic applications in prostate cancer.

show abstract

“…There have been efforts to use the prostate shape, obtained from MRI, to constrain the prostate segmentation in TRUS [10,24,11]. For segmentation of 3D TRUS images a common approach is to start with the segmentation from the mid-gland and propagate the contour to the neighbouring slices using edge-based or region-based methods [13,23,19]. However, these methods are typically much less accurate and less robust than their counterparts for prostate segmentation in MRI [14,9,8].…”

Section: Introductionmentioning

confidence: 99%

Automatic Segmentation of the Prostate on 3D Trans-rectal Ultrasound Images using Statistical Shape Models and Convolutional Neural Networks

Samei,

Karimi,

Kesch

et al. 2021

Preprint

View full text Add to dashboard Cite

In this work we propose to segment the prostate on a challenging dataset of trans-rectal ultrasound (TRUS) images using convolutional neural networks (CNNs) and statistical shape models (SSMs). TRUS is commonly used for a number of image-guided interventions on the prostate. Fast and accurate segmentation on the organ in these images is crucial to planning and fusion with other modalities such as magnetic resonance images (MRIs) . However, TRUS has limited soft tissue contrast and signal to noise ratio which makes the task of segmenting the prostate challenging and subject to inter-observer and intra-observer variability. This is especially problematic at the base and apex where the gland boundary is hard to define. In this paper, we aim to tackle this problem by taking advantage of shape priors learnt on an MR dataset which has higher soft tissue contrast allowing the prostate to be contoured more accurately. We use this shape prior in combination with a prostate tissue probability map computed by a CNN for segmentation.

show abstract

Rotationally resliced 3D prostate TRUS segmentation using convex optimization with shape priors

Cited by 14 publications

References 62 publications

Automatic prostate segmentation using deep learning on clinically diverse 3D transrectal ultrasound images

Automatic prostate segmentation using deep learning on clinically diverse 3D transrectal ultrasound images

Ultrasound prostate segmentation based on multidirectional deeply supervised V‐Net

Automatic Segmentation of the Prostate on 3D Trans-rectal Ultrasound Images using Statistical Shape Models and Convolutional Neural Networks

Contact Info

Product

Resources

About