Inter- and Intra-Observer Variability in Prostate Definition With Tissue Harmonic and Brightness Mode Imaging

Contrast-Specific Spherical Lesion Phantoms and Ancillary Analysis Software for the Objective Evaluation of Transrectal Ultrasound System Contrast Detectability

Brachytherapy is an efficacious treatment option because of its benefits for patient recovery, dose localization and conformity, but these favorable outcomes can be ensured only if the transrectal ultrasound (TRUS) system is optimized for the specific application of ultrasound-guided prostate brachytherapy. The ability to delineate the prostate from surrounding tissue during TRUS-guided prostate brachytherapy is vital for treatment planning, and consequently, so is the contrast resolution. This study describes the development of task-specific contrast-detail phantoms with clinically relevant contrast and spherical target sizes for contrast-detail performance evaluation of TRUS systems used in the brachytherapy procedure. The procedure for objective assessment of the contrast detectability of the TRUS systems is also described; a program was developed in MATLAB (R2017a, The MathWorks, Natick, MA, USA) to quantitatively analyze image quality in terms of the lesion signal-to-noise ratio (LSNR) and validated with representative control test images. The LSNR of the Hitachi EUB-7500A (2013, Hitachi, Ltd, Tokyo, Japan) TRUS system was measured on sagittal and transverse TRUS images of the contrast-detail phantoms described in this work. Results revealed the efficacy of the device as an image quality evaluation tool and the impact of the size, depth and relative contrast of the targets to the surrounding tissue on the contrast detectability of a TRUS system for both transducer arrays. The MATLAB program objectively measured the contrast detectability of the TRUS system and has the potential to determine optimized imaging parameters that could be designed as part of standardization of the imaging protocol used in TRUS-guided prostate brachytherapy for prostate cancer.

Deep Learning for Real-time, Automatic, and Scanner-adapted Prostate (Zone) Segmentation of Transrectal Ultrasound, for Example, Magnetic Resonance Imaging-transrectal Ultrasound Fusion Prostate Biopsy

BACKGROUND

Although recent advances in multiparametric magnetic resonance imaging (MRI) led to an increase in MRI-transrectal ultrasound (TRUS) fusion prostate biopsies, these are time consuming, laborious, and costly. Introduction of deep-learning approach would improve prostate segmentation.

OBJECTIVE

To exploit deep learning to perform automatic, real-time prostate (zone) segmentation on TRUS images from different scanners.

DESIGN, SETTING, AND PARTICIPANTS

Three datasets with TRUS images were collected at different institutions, using an iU22 (Philips Healthcare, Bothell, WA, USA), a Pro Focus 2202a (BK Medical), and an Aixplorer (SuperSonic Imagine, Aix-en-Provence, France) ultrasound scanner. The datasets contained 436 images from 181 men.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Manual delineations from an expert panel were used as ground truth. The (zonal) segmentation performance was evaluated in terms of the pixel-wise accuracy, Jaccard index, and Hausdorff distance.

RESULTS AND LIMITATIONS

The developed deep-learning approach was demonstrated to significantly improve prostate segmentation compared with a conventional automated technique, reaching median accuracy of 98% (95% confidence interval 95-99%), a Jaccard index of 0.93 (0.80-0.96), and a Hausdorff distance of 3.0 (1.3-8.7) mm. Zonal segmentation yielded pixel-wise accuracy of 97% (95-99%) and 98% (96-99%) for the peripheral and transition zones, respectively. Supervised domain adaptation resulted in retainment of high performance when applied to images from different ultrasound scanners (p > 0.05). Moreover, the algorithm's assessment of its own segmentation performance showed a strong correlation with the actual segmentation performance (Pearson's correlation 0.72, p < 0.001), indicating that possible incorrect segmentations can be identified swiftly.

CONCLUSIONS

Fusion-guided prostate biopsies, targeting suspicious lesions on MRI using TRUS are increasingly performed. The requirement for (semi)manual prostate delineation places a substantial burden on clinicians. Deep learning provides a means for fast and accurate (zonal) prostate segmentation of TRUS images that translates to different scanners.

PATIENT SUMMARY

Artificial intelligence for automatic delineation of the prostate on ultrasound was shown to be reliable and applicable to different scanners. This method can, for example, be applied to speed up, and possibly improve, guided prostate biopsies using magnetic resonance imaging-transrectal ultrasound fusion.

A review of the recommendations governing quality assurance of ultrasound systems used for guidance in prostate brachytherapy

Ultrasound guided brachytherapy for the treatment of prostate cancer has become a routine treatment option, due to many benefits including patient recovery and dose localisation [1]; however it is not clear whether the standards which govern the image quality for these systems are adequate. Upon review of the recommended standards for ultrasound systems used in prostate brachytherapy procedures, the recommended tests do not appear to be specific to the clinical application of ultrasound guided prostate brachytherapy. Rather they are generic and similar to those recommended for other clinical applications such as general abdominal scanning [2]. Furthermore, there is growing evidence that these tests should be specific to the clinical application [3,4] in order to gain meaningful data about the performance of the system for the application, and also to detect clinically relevant changes in quality control results. An additional problem is that there are no clinically relevant test phantom recommended for the quality assurance of ultrasound systems used in prostate brachytherapy. The image quality for this application of ultrasound needs to be monitored to ensure consistent levels of confidence in the procedure. This paper reviews the currently recommended test guidelines and test phantoms for ultrasound systems used in prostate brachytherapy from the different standard bodies and professional organisations. A critical analysis of those tests which are most reflective of the imaging and guidance tasks undertaken in an ultrasound guided prostate brachytherapy procedure will also be presented to inform the design of a TRUS quality assurance protocol.

Development and initial evaluation of a novel 3D volumetric outlining system

Aim

The novel three-dimensional (3D) radiotherapy interactive outlining tool allows volumes to be created from a handful of points within axial, sagittal and coronal planes. 3D volumetric visualisation allows users to directly manipulate the resulting volume using innovative-sculpting tools. This paper discusses the development and initial evaluation of the software ahead of formal clinical testing.

Materials and methods

User feedback was collated as part of the software development phase to ensure clinical suitability, define user training strategies and identify best practice. A loosely structured format was adopted with leading descriptive questions aiming to generate suggestions for improvements and initiate further discussion.

Results

The four participants reported great satisfaction and value in being able to use all three planes for outlining, although orientation in 3D was evidently a problem. All participants felt that the software was capable of producing acceptable outlines rapidly and that the multi-planar capability allowed for improved outlining of the prostate apex.

Findings

Mesh generation from a small number of points placed on a range of planes is a rapid and effective means of target delineation. Multi-slice volume sculpting and 3D orientation is challenging and may indicate a need for a paradigm shift in anatomy and computed tomography training.