Improving 3D ultrasound prostate localisation in radiotherapy through increased automation of interfraction matching

BACKGROUND AND PURPOSE

Daily image guidance is standard care for prostate radiotherapy. Innovations which improve the accuracy and efficiency of ultrasound guidance are needed, particularly with respect to reducing interobserver variation. This study explores automation tools for this purpose, demonstrated on the Elekta Clarity Autoscan®. The study was conducted as part of the Clarity-Pro trial (NCT02388308).

MATERIALS AND METHODS

Ultrasound scan volumes were collected from 32 patients. Prostate matches were performed using two proposed workflows and the results compared with Clarity's proprietary software. Gold standard matches derived from manually localised landmarks provided a reference. The two workflows incorporated a custom 3D image registration algorithm, which was benchmarked against a third-party application (Elastix).

RESULTS

Significant reductions in match errors were reported from both workflows compared to standard protocol. Median (IQR) absolute errors in the left-right, anteroposterior and craniocaudal axes were lowest for the Manually Initiated workflow: 0.7(1.0) mm, 0.7(0.9) mm, 0.6(0.9) mm compared to 1.0(1.7) mm, 0.9(1.4) mm, 0.9(1.2) mm for Clarity. Median interobserver variation was ≪0.01 mm in all axes for both workflows compared to 2.2 mm, 1.7 mm, 1.5 mm for Clarity in left-right, anteroposterior and craniocaudal axes. Mean matching times was also reduced to 43 s from 152 s for Clarity. Inexperienced users of the proposed workflows attained better match precision than experienced users on Clarity.

CONCLUSION

Automated image registration with effective input and verification steps should increase the efficacy of interfraction ultrasound guidance compared to the current commercially available tools.

Deformable registration of 3D ultrasound volumes using automatic landmark generation

US image registration is an important task e.g. in Computer Aided Surgery. Due to tissue deformation occurring between pre-operative and interventional images often deformable registration is necessary. We present a registration method focused on surface structures (i.e. saliencies) of soft tissues like organ capsules or vessels. The main concept follows the idea of representative landmarks (so called leading points). These landmarks represent saliencies in each image in a certain region of interest. The determination of deformation was based on a geometric model assuming that saliencies could locally be described by planes. These planes were calculated from the landmarks using two dimensional linear regression. Once corresponding regions in both images were found, a displacement vector field representing the local deformation was computed. Finally, the deformed image was warped to match the pre-operative image. For error calculation we used a phantom representing the urinary bladder and the prostate. The phantom could be deformed to mimic tissue deformation. Error calculation was done using corresponding landmarks in both images. The resulting target registration error of this procedure amounted to 1.63 mm. With respect to patient data a full deformable registration was performed on two 3D-US images of the abdomen. The resulting mean distance error was 2.10 ± 0.66 mm compared to an error of 2.75 ± 0.57 mm from a simple rigid registration. A two-sided paired t-test showed a p-value < 0.001. We conclude that the method improves the results of the rigid registration considerably. Provided an appropriate choice of the filter there are many possible fields of applications.

Automatic and efficient MRI-US segmentations for improving intraoperative image fusion in image-guided neurosurgery

Knowledge of the exact tumor location and structures at risk in its vicinity are crucial for neurosurgical interventions. Neuronavigation systems support navigation within the patient's brain, based on preoperative MRI (preMRI). However, increasing tissue deformation during the course of tumor resection reduces navigation accuracy based on preMRI. Intraoperative ultrasound (iUS) is therefore used as real-time intraoperative imaging. Registration of preMRI and iUS remains a challenge due to different or varying contrasts in iUS and preMRI. Here, we present an automatic and efficient segmentation of B-mode US images to support the registration process. The falx cerebri and the tentorium cerebelli were identified as examples for central cerebral structures and their segmentations can serve as guiding frame for multi-modal image registration. Segmentations of the falx and tentorium were performed with an average Dice coefficient of 0.74 and an average Hausdorff distance of 12.2 mm. The subsequent registration incorporates these segmentations and increases accuracy, robustness and speed of the overall registration process compared to purely intensity-based registration. For validation an expert manually located corresponding landmarks. Our approach reduces the initial mean Target Registration Error from 16.9 mm to 3.8 mm using our intensity-based registration and to 2.2 mm with our combined segmentation and registration approach. The intensity-based registration reduced the maximum initial TRE from 19.4 mm to 5.6 mm, with the approach incorporating segmentations this is reduced to 3.0 mm. Mean volumetric intensity-based registration of preMRI and iUS took 40.5 s, including segmentations 12.0 s.

•

We demonstrate that our segmentation-based registration increases accuracy, robustness, and speed of multi-modal image registration of preoperative MRI and intraoperative ultrasound images for improving intraoperative image guided neurosurgery.

•

For this we provide a fast and efficient segmentation of central anatomical structures of the perifalcine region on ultrasound images.

•

We demonstrate the advantages of our method by comparing the results of our segmentation-based registration with the initial registration provided by the navigation system and with an intensity-based registration approach.

Robust deformable registration of pre- and post-resection ultrasound volumes for visualization of residual tumor in neurosurgery

The brain tissue deforms significantly during neurosurgery, which has led to the use of intra-operative ultrasound in many sites to provide updated ultrasound images of tumor and critical parts of the brain. Several factors degrade the quality of post-resection ultrasound images such as hemorrhage, air bubbles in tumor cavity and the application of blood-clotting agent around the edges of the resection. As a result, registration of post- and pre-resection ultrasound is of significant clinical importance. In this paper, we propose a nonrigid symmetric registration (NSR) framework for accurate alignment of pre- and post-resection volumetric ultrasound images in near real-time. We first formulate registration as the minimization of a regularized cost function, and analytically derive its derivative to efficiently optimize the cost function. We use Efficient Second-order Minimization (ESM) method for fast and robust optimization. Furthermore, we use inverse-consistent deformation method to generate realistic deformation fields. The results show that NSR significantly improves the quality of alignment between pre- and post-resection ultrasound images.

Evaluation of a new transperineal ultrasound probe for inter-fraction image-guidance for definitive and post-operative prostate cancer radiotherapy

PURPOSE

The aim of this study was to evaluate a new system based on transperineal ultrasound (TP-US) acquisitions for prostate and post-prostatectomy pre-treatment positioning by comparing this device to cone-beam computed tomography (CBCT).

METHODS

The differences between CBCT/CT and TP-US/TP-US registrations were analyzed on 427 and 453 sessions for 13 prostate and 14 post-prostatectomy patients, respectively. The inter-operator variability (IOV) of the registration process, and the impact and variability of the probe pressure were also evaluated.

RESULTS

CBCT and TP-US shift agreements at ± 5 mm were 76.6%, 95.1%, 96.3% and 90.3%, 85.0%, 97.6% in anterior-posterior, superior-inferior and left-right directions, for prostate and post-prostatectomy patients, respectively. IOV values were similar between the 2 modalities. Displacements above 5 mm due to strong pressures were observed on both localizations, but such pressures were rarely reproduced during treatment courses.

CONCLUSIONS

High concordance between CBCT/CT and TP-US/TP-US localization of prostates or prostatic beds was found in this study. TP-US based prepositioning is a feasible method to ensure accurate treatment delivery, and represents an attractive alternative to invasive and/or irradiating imaging modalities.

Ultrasound versus Cone-beam CT image-guided radiotherapy for prostate and post-prostatectomy pretreatment localization

PURPOSE

To evaluate the accuracy of an intra-modality trans-abdominal ultrasound (TA-US) device against soft-tissue based Cone-Beam Computed tomography (CBCT) registration for prostate and post-prostatectomy pre-treatment positioning.

METHODS

The differences between CBCT and US shifts were calculated on 25 prostate cancer patients (cohort A) and 11 post-prostatectomy patients (cohort B), resulting in 284 and 106 paired shifts for cohorts A and B, respectively. As a second step, a corrective method was applied to the US registration results to decrease the systematic shifts observed between TA-US and CBCT results. This method consisted of subtracting the mean difference obtained between US and CBCT registration results during the first 3 sessions from the US registration results of the subsequent sessions. Inter-operator registration variability (IOV) was also investigated for both modalities.

RESULTS

After initial review, about 20% of the US images were excluded because of insufficient quality. The average differences between US and CBCT were: 2.8 ± 4.1 mm, -0.9 ± 4.2 mm, 0.4 ± 3.4 mm for cohort A and 1.3 ± 5.0 mm, -2.3 ± 4.6 mm, 0.5 ± 2.9 mm for cohort B, in the anterior-posterior (AP), superior-inferior (SI) and lateral (LR) directions, respectively. After applying the corrective method, only the differences in the AP direction remained significant (p < 0.05). The IOV values were between 0.6-2.0 mm and 2.1-3.5 mm for the CBCT and TA-US modalities, respectively.

CONCLUSIONS

Based on the obtained results and on the image quality, the TA-US imaging modality is not safely interchangeable with CBCT for pre-treatment repositioning. Treatment margins adaptation based on the correction of the systematic shifts should be considered.

[Image-guided radiotherapy: Overview of devices and practice in France in 2015]

Image-guided radiation therapy consists in acquiring in-room images to improve patient and mainly tumour set up accuracy. Many devices based on ionising or non-ionising radiations were designed in recent years. The use of such devices is of major importance in the management of patient radiotherapy courses. Those imaging sessions require to clearly define procedures in each radiotherapy department (image modality, acquisition frequency, corrective action, staff training and tasks). A quick review of the different existing image-guided radiation therapy devices is presented. In addition, the results of a French national survey about image-guided radiation therapy are presented: the survey is about both equipment and procedures. A total of 57 radiotherapy departments have participated, representing more than 160 treatment devices. About three linear accelerators out of four are equipped with an image-guiding device. The most common equipment is the CBCT system. Most centres have set up training sessions for the technicians to allow them to analyse online daily images. The management of in-room imaging dose is still under investigation, but many centres use an accounting scheme. While the devices are used to adjust the positioning of patients, in more than half of the centres, the practice had an impact on the choice of clinical and planning target volume margins. This survey led to an inventory in 2015, and could be renewed in some years.