Dataset Information

Validation of automated magnetic resonance image segmentation for radiation therapy planning in prostate cancer.

ABSTRACT:

Background and purpose

Magnetic resonance imaging (MRI) is increasingly used in radiation therapy planning of prostate cancer (PC) to reduce target volume delineation uncertainty. This study aimed to assess and validate the performance of a fully automated segmentation tool (AST) in MRI based radiation therapy planning of PC.

Material and methods

Pelvic structures of 65 PC patients delineated in an MRI-only workflow according to established guidelines were included in the analysis. Automatic vs manual segmentation by an experienced oncologist was compared with geometrical parameters, such as the dice similarity coefficient (DSC). Fifteen patients had a second MRI within 15 days to assess repeatability of the AST for prostate and seminal vesicles. Furthermore, we investigated whether hormonal therapy or body mass index (BMI) affected the AST results.

Results

The AST showed high agreement with manual segmentation expressed as DSC (mean, SD) for delineating prostate (0.84, 0.04), bladder (0.92, 0.04) and rectum (0.86, 0.04). For seminal vesicles (0.56, 0.17) and penile bulb (0.69, 0.12) the respective agreement was moderate. Performance of AST was not influenced by neoadjuvant hormonal therapy, although those on treatment had significantly smaller prostates than the hormone-naïve patients (p < 0.0001). In repeat assessment, consistency of prostate delineation resulted in mean DSC of 0.89, (SD 0.03) between the paired MRI scans for AST, while mean DSC of manual delineation was 0.82, (SD 0.05).

Conclusion

Fully automated MRI segmentation tool showed good agreement and repeatability compared with manual segmentation and was found clinically robust in patients with PC. However, manual review and adjustment of some structures in individual cases remain important in clinical use.

SUBMITTER: Kuisma A

PROVIDER: S-EPMC7807774 | biostudies-literature | 2020 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Validation of automated magnetic resonance image segmentation for radiation therapy planning in prostate cancer.

Kuisma Anna A Ranta Iiro I Keyriläinen Jani J Suilamo Sami S Wright Pauliina P Pesola Marko M Warner Lizette L Löyttyniemi Eliisa E Minn Heikki H

Physics and imaging in radiation oncology 20200101

<h4>Background and purpose</h4>Magnetic resonance imaging (MRI) is increasingly used in radiation therapy planning of prostate cancer (PC) to reduce target volume delineation uncertainty. This study aimed to assess and validate the performance of a fully automated segmentation tool (AST) in MRI based radiation therapy planning of PC.<h4>Material and methods</h4>Pelvic structures of 65 PC patients delineated in an MRI-only workflow according to established guidelines were included in the analysis ...[more]

PMID: 33458302

Similar Datasets

Project description:PURPOSE:To develop and validate a technique for radiation therapy gating using slow (?1 frame per second) magnetic resonance imaging (MRI) and a motion model. Proposed uses of the technique include radiation therapy gating using T2-weighted images and conducting additional imaging studies during gated treatments. METHODS AND MATERIALS:The technique uses a physiologically guided breathing motion model to interpolate deformed target position between 2-dimensional (2D) MRI images acquired every 1 to 3 seconds. The model is parameterized by a 1-dimensional respiratory bellows surrogate and is continuously updated with the most recently acquired 2D images. A phantom and 8 volunteers were imaged with a 0.35T MRI-guided radiation therapy system. A balanced steady-state free precession sequence with a 2D frame rate of 3 frames per second was used to evaluate the technique. The accuracy and beam-on positive predictive value (PPV) of the model-based gating decisions were evaluated using the gating decisions derived from imaging as a ground truth. A T2-weighted gating offline proof-of-concept study using a half-Fourier, single-shot, turbo-spin echo sequence is reported. RESULTS:Model-interpolated gating accuracy, beam-on PPV, and median absolute distances between model and image-tracked target centroids were, on average, 98.3%, 98.4%, and 0.33 mm, respectively, in the balanced steady-state free precession phantom studies and 93.7%, 92.1%, and 0.86 mm, respectively, in the volunteer studies. T2 model-interpolated gating in 6 volunteers yielded an average accuracy and PPV of 94.3% and 92.5%, respectively, and the mean absolute median distance between modeled and imaged target centroids was 0.86 mm. CONCLUSIONS:This work demonstrates the concept of model-interpolated gating for MRI-guided radiation therapy. The technique was found to be potentially sufficiently accurate for clinical use. Further development is needed to accommodate out-of-plane motion and the use of an internal MR-based respiratory surrogate.

Project description:To investigate the undesired irradiations outside of the treatment field by electron streams in air (air-electron-stream) during magnetic-resonance image-guided radiation therapy (MR-IGRT). A custom-made support phantom adjusting angles between the beam central axis (CAX) and the phantom surface (termed phantom-angles), were used. Using the ViewRay system, a rectangular parallelepiped phantom placed on the support phantom, was irradiated with field sizes of 6.3 cm × 6.3 cm (FS6.3) and 12.6 cm × 12.6 cm (FS12.6) at gantry angles of 0°, 30°, and 330°, and phantom-angles of 10°, 20°, and 30°. For each beam delivery, the isocenter was located at the center of mass of the phantom and 3 Gy was delivered to the isocenter (prescription dose = 3 Gy). The doses given by the air-electron-streams were measured using the EBT3 films on the panels placed orthogonal to the direction of the magnetic field at distances of 10 and 17 cm from CAX. Two dose distributions per irradiation were measured on the panel facing the phantom surface of the incident beam (front panel) and on the panel facing the phantom surface of the beam exit (end panel). We investigated the doses by the air-electron-streams by calculating the average doses inside the circles drawn around a point of the maximum dose with radii of x cm (DRx) from the dose distributions on the panels (x = 1-5 cm). The largest value of DRx was DR1 (1.64 Gy, 55% of the prescription dose) at 10 cm distance from CAX, with FS12.6, at 30° phantom-angle and 330° gantry angle. The average difference of the DR1 at the end panels (FS12.6) between the calculations and measurements was 1.36 Gy. The average global gamma passing rate with 3%/3 mm on the dose distributions at the end panels (FS12.6) was 40.3%. The calculated dose distributions on both panels were not coincident with the measured dose distributions. The Spearman's rank correlation coefficients between the projected areas and the DRx values were always higher than 0.75 (all with p < 0.001). The doses by the air-electron-streams increased with the projected areas of the cross-sections of the treatment beams on the panels.

Project description:IntroductionAs the life expectancy of children with congenital heart disease (CHD) is rapidly increasing and the adult population with CHD is growing, there is an unmet need to improve clinical workflow and efficiency of analysis. Cardiovascular magnetic resonance (CMR) is a noninvasive imaging modality for monitoring patients with CHD. CMR exam is based on multiple breath-hold 2-dimensional (2D) cine acquisitions that should be precisely prescribed and is expert and institution dependent. Moreover, 2D cine images have relatively thick slices, which does not allow for isotropic delineation of ventricular structures. Thus, development of an isotropic 3D cine acquisition and automatic segmentation method is worthwhile to make CMR workflow straightforward and efficient, as the present work aims to establish.MethodsNinety-nine patients with many types of CHD were imaged using a non-angulated 3D cine CMR sequence covering the whole-heart and great vessels. Automatic supervised and semi-supervised deep-learning-based methods were developed for whole-heart segmentation of 3D cine images to separately delineate the cardiac structures, including both atria, both ventricles, aorta, pulmonary arteries, and superior and inferior vena cavae. The segmentation results derived from the two methods were compared with the manual segmentation in terms of Dice score, a degree of overlap agreement, and atrial and ventricular volume measurements.ResultsThe semi-supervised method resulted in a better overlap agreement with the manual segmentation than the supervised method for all 8 structures (Dice score 83.23 ± 16.76% vs. 77.98 ± 19.64%; P-value ≤0.001). The mean difference error in atrial and ventricular volumetric measurements between manual segmentation and semi-supervised method was lower (bias ≤ 5.2 ml) than the supervised method (bias ≤ 10.1 ml).DiscussionThe proposed semi-supervised method is capable of cardiac segmentation and chamber volume quantification in a CHD population with wide anatomical variability. It accurately delineates the heart chambers and great vessels and can be used to accurately calculate ventricular and atrial volumes throughout the cardiac cycle. Such a segmentation method can reduce inter- and intra- observer variability and make CMR exams more standardized and efficient.

Project description:PurposeAdaptive stereotactic body radiation therapy (SBRT) for prostate cancer (PC) by the 1.5 T MR-linac currently requires online planning by an expert user. A fully automated and user-independent solution to adaptive planning (mCycle) of PC-SBRT was compared with user's plans for the 1.5 T MR-linac.Methods and materialsFifty adapted plans on daily magnetic resonance imaging scans for 10 patients with PC treated by 35 Gy (prescription dose [Dp]) in 5 fractions were reoptimized offline from scratch, both by an expert planner (manual) and by mCycle. Manual plans consisted of multicriterial optimization (MCO) of the fluence map plus manual tweaking in segmentation, whereas in mCycle plans, the objectives were sequentially optimized by MCO according to an a-priori assigned priority list. The main criteria for planning approval were a dose ≥95% of the Dp to at least 95% of the planning target volume (PTV), V33.2 (PTV) ≥ 95%, a dose less than the Dp to the hottest cubic centimeter (V35 ≤ 1 cm3) of rectum, bladder, penile bulb, and urethral planning risk volume (ie, urethra plus 3 mm isotropically), and V32 ≤ 5%, V28 ≤ 10%, and V18 ≤ 35% to the rectum. Such dose-volume metrics, plus some efficiency and deliverability metrics, were used for the comparison of mCycle versus manual plans.ResultsmCycle plans improved target dose coverage, with V33.2 (PTV) passing on average (±1 SD) from 95.7% (±1.0%) for manual plans to 97.5% (±1.3%) for mCycle plans (P < .001), and rectal dose sparing, with significantly reduced V32, V28, and V18 (P ≤ .004). Although at an equivalent number of segments, mCycle plans consumed moderately more monitor units (+17%) and delivery time (+9%) (P < .001), whereas they were generally faster (-19%) in terms of optimization times (P < .019). No significant differences were found for the passing rates of locally normalized γ (3 mm, 3%) (P = .059) and γ (2 mm, 2%) (P = .432) deliverability metrics.ConclusionsIn the offline setting, mCycle proved to be a trustable solution for automated planning of PC-SBRT on the 1.5 T MR-linac. mCycle integration in the online workflow will free the user from the challenging online-optimization task.

Dataset Information

Validation of automated magnetic resonance image segmentation for radiation therapy planning in prostate cancer.

Background and purpose

Material and methods

Results

Conclusion

Publications

Validation of automated magnetic resonance image segmentation for radiation therapy planning in prostate cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets