Mm For Bladder Research Articles

Deep-Learning-Based Multi-Organ Auto-Segmentation on 3D Transrectal Ultrasound for Ultrasound-Guided Prostate Brachytherapy

Ultrasound (US) imaging has been widely used in image-guided prostate brachytherapy. Current prostate brachytherapy uses transrectal US (TRUS) images for implant guidance, where contours of prostate and organs-at risk are necessary for treatment planning and dose evaluation. This work aims to develop a deep learning-based method for pelvic multi-organ TRUS segmentation to improve TRUS-guided prostate brachytherapy.We developed an anchor-free mask convolutional neural network (CNN) that consist of three subnetworks, i.e., a backbone, a fully convolutional one-state object detector (FCOS) head, and a mask head. The backbone extracts multi-level and multi-scale features from US image. The FOCS head utilizes these features to detect the location of volume-of-interest (VOI) of each organ. In contrast to previous mask regional CNN (Mask R-CNN) method, which perform detection and segmentation within several pre-defined potential VOIs (called as anchors), FCOS head is anchor-free, which can capture spatial correlation of multi-organs due to whole image input. Mask head performs segmentation on each detected VOI, where a spatial attention strategy is integrated into mask head that helps to focus on informative features that can well-represent the organ boundary and suppress noise introduced by previous subnetwork and by arrival TRUS image itself. For evaluation, we retrospectively investigated 80 prostate cancer patients by a five-fold cross-validation. Prostate, bladder, rectum and urethra were segmented and compared with manual contours using Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), and mean surface distance (MSD).For all patients, the DSC, HD95 and MSD were 0.93 ± 0.03, 2.28 ± 0.64 mm and 0.57 ± 0.20 mm for prostate; 0.75 ± 0.12, 2.58 ± 0.7 mm and 1.26 ± 0.23 mm for bladder; 0.90 ± 0.07, 1.65 ± 0.52 mm and 0.34 ± 0.16 mm for rectum; and 0.86 ± 0.07, 1.85 ± 1.71 mm and 0.44 ± 0.32mm for urethra. The proposed method outperforms two state-of-art methods (U-Net and Mask R-CNN) with better agreement with physicians' manual contours and less mis-identified speckles.We proposed an automatic deep learning-based pelvic multi-organ segmentation on 3D TRUS images for TRUS-guided prostate brachytherapy. The proposed method could provide fast and accurate multi-organ segmentations for prostate, bladder, rectum and urethra, and outperformed two competing methods. This deep-learning-based multi-organ auto-segmentation may play an important role in future clinical practice to improve auto-planning and auto-evaluation during TRUS-guided prostate brachytherapy.

Male pelvic CT multi-organ segmentation using synthetic MRI-aided dual pyramid networks

The delineation of the prostate and organs-at-risk (OARs) is fundamental to prostate radiation treatment planning, but is currently labor-intensive and observer-dependent. We aimed to develop an automated computed tomography (CT)-based multi-organ (bladder, prostate, rectum, left and right femoral heads (RFHs)) segmentation method for prostate radiation therapy treatment planning. The proposed method uses synthetic MRIs (sMRIs) to offer superior soft-tissue information for male pelvic CT images. Cycle-consistent adversarial networks (CycleGAN) were used to generate CT-based sMRIs. Dual pyramid networks (DPNs) extracted features from both CTs and sMRIs. A deep attention strategy was integrated into the DPNs to select the most relevant features from both CTs and sMRIs to identify organ boundaries. The CT-based sMRI generated from our previously trained CycleGAN and its corresponding CT images were inputted to the proposed DPNs to provide complementary information for pelvic multi-organ segmentation. The proposed method was trained and evaluated using datasets from 140 patients with prostate cancer, and were then compared against state-of-art methods. The Dice similarity coefficients and mean surface distances between our results and ground truth were 0.95 ± 0.05, 1.16 ± 0.70 mm; 0.88 ± 0.08, 1.64 ± 1.26 mm; 0.90 ± 0.04, 1.27 ± 0.48 mm; 0.95 ± 0.04, 1.08 ± 1.29 mm; and 0.95 ± 0.04, 1.11 ± 1.49 mm for bladder, prostate, rectum, left and RFHs, respectively. Mean center of mass distances was within 3 mm for all organs. Our results performed significantly better than those of competing methods in most evaluation metrics. We demonstrated the feasibility of sMRI-aided DPNs for multi-organ segmentation on pelvic CT images, and its superiority over other networks. The proposed method could be used in routine prostate cancer radiotherapy treatment planning to rapidly segment the prostate and standard OARs.

Development and validation of a deep learning algorithm for auto-delineation of clinical target volume and organs at risk in cervical cancer radiotherapy

PurposeThe delineation of the clinical target volume (CTV) is a crucial, laborious and subjective step in cervical cancer radiotherapy. The aim of this study was to propose and evaluate a novel end-to-end convolutional neural network (CNN) for fully automatic and accurate CTV in cervical cancer. MethodsA total of 237 computed tomography (CT) scans of patients with locally advanced cervical cancer were collected and evaluated. A novel 2.5D CNN network, called DpnUNet, was developed for CTV delineation and further applied for CTV and organ-at-risk (OAR) delineation simultaneously. Comprehensive comparisons and experiments were performed. The mean Dice similarity coefficient (DSC), 95th percentile Hausdorff distance (95HD) and subjective evaluation were used to assess the performance of this method. ResultsThe mean DSC and 95HD values were 0.86 and 5.34 mm for the delineated CTVs. The clinical experts’ subjective assessments showed that 90% of the predicted contours were acceptable for clinical usage. The mean DSC and 95HD values were 0.91 and 4.05 mm for bladder, 0.85 and 2.16 mm for bone marrow, 0.90 and 1.27 mm for left femoral head, 0.90 and 1.51 mm for right femoral head, 0.82 and 4.29 mm for rectum, 0.85 and 4.35 mm for bowel bag, 0.82 and 4.96 mm for spinal cord respectively. The average delineation time for one patient’s CT images was within 15 seconds. ConclusionThe experimental results demonstrate that the CTV and OARs delineated for cervical cancer by DpnUNet was in close agreement with the ground truth. DpnUNet could significantly reduce the radiation oncologists’ contouring time.

Pelvic multi-organ segmentation on cone-beam CT for prostate adaptive radiotherapy.

The purpose of this study is to develop a deep learning-based approach to simultaneously segment five pelvic organs including prostate, bladder, rectum, left and right femoral heads on cone-beam CT (CBCT), as required elements for prostate adaptive radiotherapy planning. We propose to utilize both CBCT and CBCT-based synthetic MRI (sMRI) for the segmentation of soft tissue and bony structures, as they provide complementary information for pelvic organ segmentation. CBCT images have superior bony structure contrast and sMRIs have superior soft tissue contrast. Prior to segmentation, sMRI was generated using a cycle-consistent adversarial networks (CycleGAN), which was trained using paired CBCT-MR images. To combine the advantages of both CBCT and sMRI, we developed a cross-modality attention pyramid network with late feature fusion. Our method processes CBCT and sMRI inputs separately to extract CBCT-specific and sMRI-specific features prior to combining them in a late-fusion network for final segmentation. The network was trained and tested using 100 patients' datasets, with each dataset including the CBCT and manual physician contours. For comparison, we trained another two networks with different network inputs and architectures. The segmentation results were compared to manual contours for evaluations. For the proposed method, dice similarity coefficients and mean surface distances between the segmentation results and the ground truth were 0.96±0.03, 0.65±0.67mm; 0.91±0.08, 0.93±0.96mm; 0.93±0.04, 0.72±0.61mm; 0.95±0.05, 1.05±1.40mm; and 0.95±0.05, 1.08±1.48mm for bladder, prostate, rectum, left and right femoral heads, respectively. As compared to the other two competing methods, our method has shown superior performance in terms of the segmentation accuracy. We developed a deep learning-based segmentation method to rapidly and accurately segment five pelvic organs simultaneously from daily CBCTs. The proposed method could be used in the clinic to support rapid target and organs-at-risk contouring for prostate adaptive radiation therapy.

TU‐AB‐202‐08: Generating Organ Surfaces to Overcome Random Contouring Errors and Slice Thickness Variations On Multimodality Images

Purpose:To propose a new method that generates smooth organ surface meshes that overcomes random contouring errors and volume averaging artifacts for hollow organs.Methods:The proposed contour‐to‐smooth mesh generation method (CSMG) consists of two parts; 1) reconstructing the inner and outer surfaces of the hollow organ with triangular meshes based on parametric active surface (PAS) and 2) modifying the reconstructed inner surface in order to satisfy minimum wall thickness constraints and matching the patient's known organ wall volume. In both steps, PAS evolved gradually to physician contours by minimizing defined energy functions. CSMG surface meshes were evaluated using two digital phantoms and 72 contours from 3 cervix cancer patients (4 images per patient, 3 organs per image, 2 inner‐and‐outer‐wall contours per organ). The CSMG and treatment planning system (TPS) surface meshes were compared using Gaussian curvature, Dice conformity index (DCI) and signed minimum distance (SMD). DCI and SMD were calculated between the physician contours and the contour extracted by averaging the surface over the slice thickness.Results:CSMG meshes were smoother and more continuous than TPS surface meshes. The DCI differences between CSMG and TPS were 0.2%, 0.9% and 1.5% on average for bladder, rectum and vagina, respectively. TPS meshes had larger average SMDs (0.44±0.64, 0.43±0.62 and 0.42±0.61 mm for the three organs) compared to CSMG meshes (0.01±0.84, −0.05±0.93 and 0.13±0.91 mm). Larger grand mean and smaller pooled standard deviation of TPS SMDs implied that TPS overfitted the noise on contours but PM smoothed out the noise. The wall thicknesses were 2.7±1.3, 2.6±1.1 and 1.6±0.5 mm for bladder, rectum and vagina, respectively. The wall volume differed from reference volume less than ±2.5%.Conclusion:The proposed method reconstructs the smooth and continuous organ surface from sparsely sampled data points without introducing large systematic errors.This work was supported in part by Varian Research agreement. The authors have the following disclosures: NIH (Williamson, Hugo, Christensen, and Weiss), Philips research agreement (Hugo and Weiss), Varian license agreement (Hugo and Weiss), Roger Koch fund (Christensen), UpToDate royalties (Weiss). There is no COI.

SU‐G‐JeP3‐05: Geometry Based Transperineal Ultrasound Probe Positioning for Image Guided Radiotherapy

Purpose:The use of ultrasound (US) imaging in radiotherapy is not widespread, primarily due to the need for skilled operators performing the scans. Automation of probe positioning has the potential to remove this need and minimize operator dependence. We introduce an algorithm for obtaining a US probe position that allows good anatomical structure visualization based on clinical requirements. The first application is on 4D transperineal US images of prostate cancer patients.Methods:The algorithm calculates the probe position and orientation using anatomical information provided by a reference CT scan, always available in radiotherapy workflows. As initial test, we apply the algorithm on a CIRS pelvic US phantom to obtain a set of possible probe positions. Subsequently, five of these positions are randomly chosen and used to acquire actual US volumes of the phantom. Visual inspection of these volumes reveal if the whole prostate, and adjacent edges of bladder and rectum are fully visualized, as clinically required. In addition, structure positions on the acquired US volumes are compared to predictions of the algorithm.Results:All acquired volumes fulfill the clinical requirements as specified in the previous section. Preliminary quantitative evaluation was performed on thirty consecutive slices of two volumes, on which the structures are easily recognizable. The mean absolute distances (MAD) between actual anatomical structure positions and positions predicted by the algorithm were calculated. This resulted in MAD of 2.4±0.4 mm for prostate, 3.2±0.9 mm for bladder and 3.3±1.3 mm for rectum.Conclusion:Visual inspection and quantitative evaluation show that the algorithm is able to propose probe positions that fulfill all clinical requirements. The obtained MAD is on average 2.9 mm. However, during evaluation we assumed no errors in structure segmentation and probe positioning. In future steps, accurate estimation of these errors will allow for better evaluation of the achieved accuracy.

Mo1707 The Effect of Pentadecapeptide BPC 157 on Colovesical Fistula Healing

Introduction. A colovesical fistulas present an important clinical problems. They usually occur as a complication of diverticulosis, IBD or malignant process. The pentadecapeptide BPC 157 , in clinical trials for IBD therapy, has already shown the effectiveness in healing colocutaneous fistulas ( J. Pharmacol Sci 2008 ) , gastrocutaneous fistulas ( Dig Dis Sci 2009 ). Therefore we suggest it as a possible therapy of colovesical fistulas. Matherials and methods. Wistar Albino rats were randomly assigned, at 5 cm from anocutanous borderline, with the diameter of 3 mm. The surgical procedure was performed according to rules brought by the Local Ethical Committee. Medication of BPC 157 (10ug/kg, 10ng/kg) was per-orally, in drinkingwater and intraperitoneally, once daily, the first dose immediately after the operation, the last dose at 24 h before sacrifice, at 7, 15, 30 post-operative day, with the biomechanical, functional, macroscopic and microscopic assessment. Results. Biomechanical analysis :we measured pressure of the fistulas rupture for all groups of animals with the system for central venous pressure assesment .In the control group pressures for fistulas rupture were similar for every experimental period : ≤1mL H2O, 1mL±0,3 mL, 2 mL±0,5 ml H2O. In the second group( BPC per-orally in drinking water 10ug/kg and 10ng/kg ) water pressures were 1mL±0,3; 2mL±0,3 and 5mL±0,5 but without fistulas rupture at the and of 30 days experimental period.In the third group( BPC intraperitoneally 10ug/kg and 10ng/kg ) water pressure was increased from 2mL±0,5mL until 5 mL±0,5 at the and of 15 days experimental period. In most animals fistulas were healed at the end of 30 days experimental period. Macroscopic analysis: We measured diameters of fistulas on the colonic and the bladder side.In the control group diameters of fistulas were similar for both sides for all experimental periods : 3mm ±0.5mm on the colon and bladder side,3mm±1mm and 2,5±0,5mm. In the second group fistulas diameter was 3±0,2 mm for both sides ( 7 days ),2mm±0,5 mm for colon and 2,5±0,5 mm for bladder ( 15 days ), 1,5mm±0,5 for colon, 2mm±0,5 mm for bladder side ( 30 days ). In the third group fistulas were reduced : 2,5mm±0,2mm colon, 3mm±0,5 bladder ( 7 days ), 1,5mm±0,5mm colon; 2mm±0,5mm bladder ( 15 days ) and at the end oft he 30 days experimental period in most animals fistulas were healed er reduced diameter at 1mm±0,3mm for both sides.. Conclusion. According to these results the pentadecapeptide BPC 157 could present a new possible pathway in therapy of colovesical fistulas.

Mo1708 Serum-Derived Bovine Immunoglobulin/Protein Isolate Attenuated DSSInduced Colitis in a Defined Floral Model

Introduction. A colovesical fistulas present an important clinical problems. They usually occur as a complication of diverticulosis, IBD or malignant process. The pentadecapeptide BPC 157 , in clinical trials for IBD therapy, has already shown the effectiveness in healing colocutaneous fistulas ( J. Pharmacol Sci 2008 ) , gastrocutaneous fistulas ( Dig Dis Sci 2009 ). Therefore we suggest it as a possible therapy of colovesical fistulas. Matherials and methods. Wistar Albino rats were randomly assigned, at 5 cm from anocutanous borderline, with the diameter of 3 mm. The surgical procedure was performed according to rules brought by the Local Ethical Committee. Medication of BPC 157 (10ug/kg, 10ng/kg) was per-orally, in drinkingwater and intraperitoneally, once daily, the first dose immediately after the operation, the last dose at 24 h before sacrifice, at 7, 15, 30 post-operative day, with the biomechanical, functional, macroscopic and microscopic assessment. Results. Biomechanical analysis :we measured pressure of the fistulas rupture for all groups of animals with the system for central venous pressure assesment .In the control group pressures for fistulas rupture were similar for every experimental period : ≤1mL H2O, 1mL±0,3 mL, 2 mL±0,5 ml H2O. In the second group( BPC per-orally in drinking water 10ug/kg and 10ng/kg ) water pressures were 1mL±0,3; 2mL±0,3 and 5mL±0,5 but without fistulas rupture at the and of 30 days experimental period.In the third group( BPC intraperitoneally 10ug/kg and 10ng/kg ) water pressure was increased from 2mL±0,5mL until 5 mL±0,5 at the and of 15 days experimental period. In most animals fistulas were healed at the end of 30 days experimental period. Macroscopic analysis: We measured diameters of fistulas on the colonic and the bladder side.In the control group diameters of fistulas were similar for both sides for all experimental periods : 3mm ±0.5mm on the colon and bladder side,3mm±1mm and 2,5±0,5mm. In the second group fistulas diameter was 3±0,2 mm for both sides ( 7 days ),2mm±0,5 mm for colon and 2,5±0,5 mm for bladder ( 15 days ), 1,5mm±0,5 for colon, 2mm±0,5 mm for bladder side ( 30 days ). In the third group fistulas were reduced : 2,5mm±0,2mm colon, 3mm±0,5 bladder ( 7 days ), 1,5mm±0,5mm colon; 2mm±0,5mm bladder ( 15 days ) and at the end oft he 30 days experimental period in most animals fistulas were healed er reduced diameter at 1mm±0,3mm for both sides.. Conclusion. According to these results the pentadecapeptide BPC 157 could present a new possible pathway in therapy of colovesical fistulas.

SU‐E‐T‐288: A Finite‐Element Model to Predict Prostate Displacement and Deformation Due to Bladder Filling, Rectal Distension and Patient Posture during Prostate Brachytherapy

Purpose: Noninvasive determination of the location and deformation of pelvic organs in different stages of organs fullness and patient‐related situations encountered during the post‐implant prostate irradiation period. To help perform a more realistic treatment planning, a biomechanical finite‐element (FE) model is presented which provides prospective information on individual patients based on their initial MRI. Methods: Three patients with different prostate sizes (mean 54.6 g, range 32–71 g) were studied. Sequential MR images at different stages of bladder and rectal filling and patient positions (supine and left decubitus) were acquired. The geometry of pelvic organs was 3D reconstructed using MR images of supine patients and then transferred to the ABAQUS software for FE modeling. To tune the FE model, the tissue mechanical properties, distension forces and boundary conditions were adjusted to find the optimum combinations of these parameters, to individually produce the best match with images of each patient in all positions and stages. Finally, the model was tested in a predictive mode, in which the averaged coefficients were used to predict prostate displacement and deformation following bladder and rectal filling for supine and left‐decubitus positions. Results: The tuned model showed prostate displacements/deformations that matched the measurements to within 1 mm. In the supine position, modeling the same bladder and rectal volumes as their image‐based measurements, mean prediction discrepancy for bladder‐filling prostate displacement was 1.5 mm, and 2.9 mm for bladder filling plus rectal distension. The corresponding left‐decubitus discrepancies were 4.3 mm and 4.1 mm. Mean prostate deformation discrepancies due to large rectal distension were 1.0 mm in both supine and left‐decubitus positions. Conclusions: The model predicts prostate displacement and deformation with reasonable accuracy and offers the potential to provide more representative dose distributions. This model can potentially be used to improve organs dosimetry in brachytherapy.

Clinical Evaluation of Soft Tissue Organ Boundary Visualization on Cone-Beam Computed Tomographic Imaging

Cone-beam computed tomographic images (CBCTs) are increasingly used for setup correction, soft tissue targeting, and image-guided adaptive radiotherapy. However, CBCT image quality is limited by low contrast and imaging artifacts. This analysis investigates the detectability of soft tissue boundaries in CBCT by performing a multiple-observer segmentation study. In four prostate cancer patients prostate, bladder and rectum were repeatedly delineated by five observers on CBCTs and fan-beam CTs (FBCTs). A volumetric analysis of contouring variations was performed by calculating coefficients of variation (COV: standard deviation/average volume). The topographical distribution of contouring variations was analyzed using an average surface mesh-based method. Observer- and patient-averaged COVs for FBCT/CBCT were 0.09/0.19 for prostate, 0.05/0.08 for bladder, and 0.09/0.08 for rectum. Contouring variations on FBCT were significantly smaller than on CBCT for prostate (p< 0.03) and bladder (p < 0.04), but not for rectum (p < 0.37; intermodality differences). Intraobserver variations from repeated contouring of the same image set were not significant for either FBCT or CBCT (p < 0.05). Average standard deviations of individual observers' contour differences from average surface meshes on FBCT vs. CBCT were 1.5 vs. 2.1 mm for prostate, 0.7 vs. 1.4 mm for bladder, and 1.3 vs. 1.5 mm for rectum. The topographical distribution of contouring variations was similar for FBCT and CBCT. Contouring variations were larger on CBCT than FBCT, except for rectum. Given the well-documented uncertainty in soft tissue contouring in the pelvis, improvement of CBCT image quality and establishment of well-defined soft tissue identification rules are desirable for image-guided radiotherapy.

TH‐C‐304A‐04: Simultaneous Delineation of Bladder and Prostate From 3D CT: An Optimal Graph‐Theoretic Approach

Introduction Precise target delineation is critical for a successful 3‐D radiotherapy treatment planning for prostate cancer treatment. Automatic segmentation techniques are urgently needed due to large amounts of 3‐D image data that require increased time for manually contouring. We report a novel segmentation method for simultaneous 3‐D surface detection of bladder and prostate, incorporating both edge and shape information. Our method is the first attempt to employ an arc‐weighted graph‐based representation for simultaneous surface detection, which delivers a globally optimal solution in low‐order polynomial time. Method and Material The core of our approach is to represent segmentation problem using an arc‐weighted graph, which permits a wider spectrum of constraints into 3‐D graph‐theoretic approach. Hard surface interacting constraints are enforced in interacting regions to preserve geometric relationships between partially interacting boundary surfaces of prostate and bladder. Soft smoothness shape compliance is further employed to incorporate shape information. Two globally mutually interacting optimal surfaces are then computed by solving a maximum flow problem. Testing the approach was accomplished on 13 CT data sets from different patients. Computed results were compared with expert‐traced standard. Results Our method yielded high‐quality segmentation, despite of the low saliency of CT images. The average probability of correct detection was 0.958±0.024 for bladder and 0.852±0.046 for prostate. The average probability of false detection was 0.115±0.062 for bladder and 0.136±0.045 for prostate. The unsigned surface distance error was 1.04±1.00mm for bladder and 1.38±1.08mm for prostate. The percentage of expert‐defined surface points with less than 5mm from our computed surface was 99% for bladder and 98.6% for prostate. Conclusion We develop a novel segmentation method for simultaneous detection of bladder and prostate. A sub‐voxel accuracy was achieved, which demonstrates the promise of our algorithm.