A Novel Positive-Contrast Magnetic Resonance Imaging Line Marker for High-Dose-Rate (HDR) MRI-Assisted Radiosurgery (MARS).

Magnetic resonance imaging (MRI) can facilitate accurate organ delineation and optimal dose distributions in high-dose-rate (HDR) MRI-Assisted Radiosurgery (MARS). Its use for this purpose has been limited by the lack of positive-contrast MRI markers that can clearly delineate the lumen of the HDR applicator and precisely show the path of the HDR source on T1- and T2-weighted MRI sequences. We investigated a novel MRI positive-contrast HDR brachytherapy or interventional radiotherapy line marker, C4:S, consisting of C4 (visible on T1-weighted images) complexed with saline. Longitudinal relaxation time (T1) and transverse relaxation time (T2) for C4:S were measured on a 1.5 T MRI scanner. High-density polyethylene (HDPE) tubing filled with C4:S as an HDR brachytherapy line marker was tested for visibility on T1- and T2-weighted MRI sequences in a tissue-equivalent female ultrasound training pelvis phantom. Relaxivity measurements indicated that C4:S solution had good T1-weighted contrast (relative to oil [fat] signal intensity) and good T2-weighted contrast (relative to water signal intensity) at both room temperature (relaxivity ratio > 1; r2/r1 = 1.43) and body temperature (relaxivity ratio > 1; r2/r1 = 1.38). These measurements were verified by the positive visualization of the C4:S (C4/saline 50:50) HDPE tube HDR brachytherapy line marker on both T1- and T2-weighted MRI sequences. Orientation did not affect the relaxivity of the C4:S contrast solution. C4:S encapsulated in HDPE tubing can be visualized as a positive line marker on both T1- and T2-weighted MRI sequences. MRI-guided HDR planning may be possible with these novel line markers for HDR MARS for several types of cancer.

Uncertainty in magnetic resonance imaging-based prostate postimplant dosimetry: Results of a 10-person human observer study, and comparisons with automatic postimplant dosimetry.

PURPOSE: Uncertainties in postimplant quality assessment (QA) for low-dose-rate prostate brachytherapy (LDRPBT) are introduced at two steps: seed localization and contouring. We quantified how interobserver variability (IoV) introduced in both steps impacts dose-volume-histogram (DVH) parameters for MRI-based LDRPBT, and compared it with automatically derived DVH parameters. METHODS AND MATERIALS: Twenty-five patients received MRI-based LDRPBT. Seven clinical observers contoured the prostate and four organs at risk, and 4 dosimetrists performed seed localization, on each MRI. Twenty-eight unique manual postimplant QAs were created for each patient from unique observer pairs. Reference QA and automatic QA were also performed for each patient. IoV of prostate, rectum, and external urinary sphincter (EUS) DVH parameters owing to seed localization and contouring was quantified with coefficients of variation. Automatically derived DVH parameters were compared with those of the reference plans. RESULTS: Coefficients of variation (CoVs) owing to contouring variability (CoVcontours) were significantly higher than those due to seed localization variability (CoVseeds) (median CoVcontours vs. median CoVseeds: prostate D90-15.12% vs. 0.65%, p < 0.001; prostate V100-5.36% vs. 0.37%, p < 0.001; rectum V100-79.23% vs. 8.69%, p < 0.001; EUS V200-107.74% vs. 21.18%, p < 0.001). CoVcontours were lower when the contouring observers were restricted to the 3 radiation oncologists, but were still markedly higher than CoVseeds. Median differences in prostate D90, prostate V100, rectum V100, and EUS V200 between automatically computed and reference dosimetry parameters were 3.16%, 1.63%, -0.00 mL, and -0.00 mL, respectively. CONCLUSIONS: Seed localization introduces substantially less variability in postimplant QA than does contouring for MRI-based LDRPBT. While automatic seed localization may potentially help improve workflow efficiency, it has limited potential for improving the consistency and quality of postimplant dosimetry.

Proton therapy for the management of localized prostate cancer: Long-term clinical outcomes at a comprehensive cancer center.

BACKGROUND AND PURPOSE: Proton therapy (PT) has emerged as a standard-of-care treatment option for localized prostate cancer at our comprehensive cancer center. However, there are few large-scale analyses examining the long-term clinical outcomes. Therefore, this article aims to evaluate the long-term effectiveness and toxicity of PT in patients with localized prostate cancer. MATERIALS AND METHODS: Review of 2772 patients treated from May 2006 through January 2020. Disease risk was stratified according to National Comprehensive Cancer Network guidelines as low [LR, n = 640]; favorable-intermediate [F-IR, n = 850]; unfavorable-intermediate [U-IR, n = 851]; high [HR, n = 315]; or very high [VHR, n = 116]. Biochemical failure and toxicity were analyzed using Kaplan-Meier estimates and multivariate models. RESULTS: The median patient age was 66 years; the median follow-up time was 7.0 years. Pelvic lymph node irradiation was prescribed to 28 patients (1%) (2 [0.2%] U-IR, 11 [3.5%] HR, and 15 [12.9%] VHR). The median dose was 78 Gy in 1.8-2.0 Gy(RBE) fractions. Freedom from biochemical relapse (FFBR) rates at 5 years and 10 years were 98.2% and 96.8% for the LR group; 98.3% and 93.6%, F-IR; 94.2% and 90.2%, U-IR; 94.3% and 85.2%, HR; and 86.1% and 68.5%, VHR. Two patients died of prostate cancer. Overall rates of late grade ≥ 3 GU and GI toxicity were 0.87% and 1.01%. CONCLUSIONS: Proton therapy for localized prostate cancer demonstrated excellent clinical outcomes in this large cohort, even among higher-risk groups with historically poor outcomes despite aggressive therapy.

Prospective Evaluation of Prostate and Organs at Risk Segmentation Software for MRI-based Prostate Radiation Therapy.

The segmentation of the prostate and surrounding organs at risk (OARs) is a necessary workflow step for performing dose-volume histogram analyses of prostate radiation therapy procedures. Low-dose-rate prostate brachytherapy (LDRPBT) is a curative prostate radiation therapy treatment that delivers a single fraction of radiation over a period of days. Prior studies have demonstrated the feasibility of fully convolutional networks to segment the prostate and surrounding OARs for LDRPBT dose-volume histogram analyses. However, performance evaluations have been limited to measures of global similarity between algorithm predictions and a reference. To date, the clinical use of automatic segmentation algorithms for LDRPBT has not been evaluated, to the authors' knowledge. The purpose of this work was to assess the performance of fully convolutional networks for prostate and OAR delineation on a prospectively identified cohort of patients who underwent LDRPBT by using clinically relevant metrics. Thirty patients underwent LDRPBT and were imaged with fully balanced steady-state free precession MRI after implantation. Custom automatic segmentation software was used to segment the prostate and four OARs. Dose-volume histogram analyses were performed by using both the original automatically generated contours and the physician-refined contours. Dosimetry parameters of the prostate, external urinary sphincter, and rectum were compared without and with the physician refinements. This study observed that physician refinements to the automatic contours did not significantly affect dosimetry parameters. Keywords: MRI, Neural Networks, Radiation Therapy, Radiation Therapy/Oncology, Genital/Reproductive, Prostate, Segmentation, Dosimetry Supplemental material is available for this article. © RSNA, 2022.

Computer-aided segmentation on MRI for prostate radiotherapy, part II: Comparing human and computer observer populations and the influence of annotator variability on algorithm variability.

BACKGROUND AND PURPOSE: Comparing deep learning (DL) algorithms to human interobserver variability, one of the largest sources of noise in human-performed annotations, is necessary to inform the clinical application, use, and quality assurance of DL for prostate radiotherapy. MATERIALS AND METHODS: One hundred fourteen DL algorithms were developed on 295 prostate MRIs to segment the prostate, external urinary sphincter (EUS), seminal vesicles (SV), rectum, and bladder. Fifty prostate MRIs of 25 patients undergoing MRI-based low-dose-rate prostate brachytherapy were acquired as an independent test set. Groups of DL algorithms were created based on the loss functions used to train them, and the spatial entropy (SE) of their predictions on the 50 test MRIs was computed. Five human observers contoured the 50 test MRIs, and SE maps of their contours were compared with those of the groups of the DL algorithms. Additionally, similarity metrics were computed between DL algorithm predictions and consensus annotations of the 5 human observers' contours of the 50 test MRIs. RESULTS: A DL algorithm yielded statistically significantly higher similarity metrics for the prostate than did the human observers (H) (prostate Matthew's correlation coefficient, DL vs. H: planning-0.931 vs. 0.903, p < 0.001; postimplant-0.925 vs. 0.892, p < 0.001); the same was true for the 4 organs at risk. The SE maps revealed that the DL algorithms and human annotators were most variable in similar anatomical regions: the prostate-EUS, prostate-SV, prostate-rectum, and prostate-bladder junctions. CONCLUSIONS: Annotation quality is an important consideration when developing, evaluating, and using DL algorithms clinically.

Computer-aided segmentation on MRI for prostate radiotherapy, Part I: Quantifying human interobserver variability of the prostate and organs at risk and its impact on radiation dosimetry.

BACKGROUND AND PURPOSE: Quantifying the interobserver variability (IoV) of prostate and periprostatic anatomy delineation on prostate MRI is necessary to inform its use for treatment planning, treatment delivery, and treatment quality assessment. MATERIALS AND METHODS: Twenty five prostate cancer patients underwent MRI-based low-dose-rate prostate brachytherapy (LDRPBT). The patients were scanned with a 3D T2-weighted sequence for treatment planning and a 3D T2/T1-weighted sequence for quality assessment. Seven observers involved with the LDRPBT workflow delineated the prostate, external urinary sphincter (EUS), seminal vesicles, rectum, and bladder on all 50 MRIs. IoV was assessed by measuring contour similarity metrics, differences in organ volumes, and differences in dosimetry parameters between unique observer pairs. Measurements from a group of 3 radiation oncologists (G1) were compared against those from a group consisting of the other 4 clinical observers (G2). RESULTS: IoV of the prostate was lower for G1 than G2 (Matthew's correlation coefficient [MCC], G1 vs. G2: planning-0.906 vs. 0.870, p < 0.001; postimplant-0.899 vs. 0.861, p < 0.001). IoV of the EUS was highest of all the organs for both groups, but was lower for G1 (MCC, G1 vs. G2: planning-0.659 vs. 0.402, p < 0.001; postimplant-0.684 vs. 0.398, p < 0.001). Large differences in prostate dosimetry parameters were observed (G1 maximum absolute prostate ΔD90: planning-76.223 Gy, postimplant-36.545 Gy; G1 maximum absolute prostate ΔV100: planning-13.927%, postimplant-8.860%). CONCLUSIONS: While MRI is optimal in the management of prostate cancer with radiation therapy, significant interobserver variability of the prostate and external urinary sphincter still exist.

Synthetic generation of DSC-MRI-derived relative CBV maps from DCE MRI of brain tumors.

PURPOSE: Perfusion MRI with gadolinium-based contrast agents is useful for diagnosis and treatment response evaluation of brain tumors. Dynamic susceptibility contrast (DSC) MRI and dynamic contrast enhanced (DCE) MRI are two gadolinium-based contrast agent perfusion imaging techniques that provide complementary information about the tumor vasculature. However, each requires a separate administration of a gadolinium-based contrast agent. The purpose of this retrospective study was to determine the feasibility of synthesizing relative cerebral blood volume (rCBV) maps, as computed from DSC MRI, from DCE MRI of brain tumors. METHODS: One hundred nine brain-tumor patients underwent both DCE and DSC MRI. Relative CBV maps were computed from the DSC MRI, and blood plasma volume fraction maps were computed from the DCE MRIs. Conditional generative adversarial networks were developed to synthesize rCBV maps from the DCE MRIs. Tumor-to-white matter ratios were calculated from real rCBV, synthetic rCBV, and plasma volume fraction maps and compared using correlation analysis. Real and synthetic rCBV in white and gray matter regions were also compared. RESULTS: Pearson correlation analysis showed that both the tumor rCBV and tumor-to-white matter ratios in the synthetic and real rCBV maps were strongly correlated (ρ = 0.87, P < .05 and ρ = 0.86, P < .05, respectively). Tumor plasma volume fraction and real rCBV were not strongly correlated (ρ = 0.47). Bland-Altman analysis showed a mean difference between the synthetic and real rCBV tumor-to-white matter ratios of 0.20 with a 95% confidence interval of ±0.47. CONCLUSION: Realistic rCBV maps can be synthesized from DCE MRI and contain quantitative information, enabling robust brain-tumor perfusion imaging of DSC and DCE parameters with a single gadolinium-based contrast agent administration.

Fully Balanced SSFP Without an Endorectal Coil for Postimplant QA of MRI-Assisted Radiosurgery (MARS) of Prostate Cancer: A Prospective Study.

PURPOSE: To investigate fully balanced steady-state free precession (bSSFP) with optimized acquisition protocols for magnetic resonance imaging (MRI)-based postimplant quality assessment of low-dose-rate (LDR) prostate brachytherapy without an endorectal coil (ERC). METHODS AND MATERIALS: Seventeen patients at a major academic cancer center who underwent MRI-assisted radiosurgery (MARS) LDR prostate cancer brachytherapy were imaged with moderate, high, or very high spatial resolution fully bSSFP MRIs without using an ERC. Between 1 and 3 signal averages (NEX) were acquired with acceleration factors (R) between 1 and 2, with the goal of keeping scan times between 4 and 6 minutes. Acquisitions with R >1 were reconstructed with parallel imaging and compressed sensing (PICS) algorithms. Radioactive seeds were identified by 3 medical dosimetrists. Additionally, some of the MRI techniques were implemented and tested at a community hospital; 3 patients underwent MARS LDR prostate brachytherapy and were imaged without an ERC. RESULTS: Increasing the in-plane spatial resolution mitigated partial volume artifacts and improved overall seed and seed marker visualization at the expense of reduced signal-to-noise ratio (SNR). The reduced SNR as a result of imaging at higher spatial resolution and without an ERC was partially compensated for by the multi-NEX acquisitions enabled by PICS. Resultant image quality was superior to the current clinical standard. All 3 dosimetrists achieved near-perfect precision and recall for seed identification in the 17 patients. The 3 postimplant MRIs acquired at the community hospital were sufficient to identify 208 out of 211 seeds implanted without reference to computed tomography (CT). CONCLUSIONS: Acquiring postimplant prostate brachytherapy MRI without an ERC has several advantages including better patient tolerance, lower costs, higher clinical throughput, and widespread access to precision LDR prostate brachytherapy. This prospective study confirms that the use of an ERC can be circumvented with fully bSSFP and advanced MRI scan techniques in a major academic cancer center and community hospital, potentially enabling postimplant assessment of MARS LDR prostate brachytherapy without CT.

MetNet: Computer-aided segmentation of brain metastases in post-contrast T1-weighted magnetic resonance imaging.

PURPOSE: Brain metastases are manually contoured during stereotactic radiosurgery (SRS) treatment planning, which is time-consuming, potentially challenging, and laborious. The purpose of this study was to develop and investigate a 2-stage deep learning (DL) approach (MetNet) for brain metastasis segmentation in pre-treatment magnetic resonance imaging (MRI). MATERIALS AND METHODS: We retrospectively analyzed postcontrast 3D T1-weighted spoiled gradient echo MRIs from 934 patients who underwent SRS between August 2009 and August 2018. Neuroradiologists manually identified brain metastases in the MRIs. The treating radiation oncologist or physicist contoured the brain metastases. We constructed a 2-stage DL ensemble consisting of detection and segmentation models to segment the brain metastases on the MRIs. We evaluated the performance of MetNet by computing sensitivity, positive predictive value (PPV), and Dice similarity coefficient (DSC) with respect to metastasis size, as well as free-response receiver operating characteristics. RESULTS: The 934 patients (mean [±standard deviation] age 59 ± 13 years, 474 women) were randomly split into 80% training and 20% testing groups (748:186). For patients with metastases 1-52 mm (n = 766), 648 (85%) were detected and segmented with a mean segmentation DSC of 81% ± 15%. Patient-averaged sensitivity was 88% ± 19%, PPV was 58% ± 25%, and DSC was 85% ± 13% with 3 ± 3 false positives (FPs) per patient. When considering only metastases ≥6 mm, patient-averaged sensitivity was 99% ± 5%, PPV was 67% ± 28%, and DSC was 87% ± 13% with 1 ± 2 FPs per patient. CONCLUSION: MetNet can segment brain metastases across a broad range of metastasis sizes with high sensitivity, low FPs, and high segmentation accuracy in postcontrast T1-weighted MRI, potentially aiding treatment planning for SRS.

Machine Segmentation of Pelvic Anatomy in MRI-Assisted Radiosurgery (MARS) for Prostate Cancer Brachytherapy.

PURPOSE: To investigate machine segmentation of pelvic anatomy in magnetic resonance imaging (MRI)-assisted radiosurgery (MARS) for prostate cancer using prostate brachytherapy MRIs acquired with different pulse sequences and image contrasts. METHODS AND MATERIALS: Two hundred 3-dimensional (3D) preimplant and postimplant prostate brachytherapy MRI scans were acquired with a T2-weighted sequence, a T2/T1-weighted sequence, or a T1-weighted sequence. One hundred twenty deep machine learning models were trained to segment the prostate, seminal vesicles, external urinary sphincter, rectum, and bladder using the MRI scans acquired with T2-weighted and T2/T1-weighted image contrast. The deep machine learning models consisted of 18 fully convolutional networks (FCNs) with different convolutional encoders. Both 2-dimensional and 3D U-Net FCNs were constructed for comparison. Six objective functions were investigated: cross-entropy, Jaccard distance, focal loss, and 3 variations of Tversky distance. The performance of the models was compared using similarity metrics, including pixel accuracy, Jaccard index, Dice similarity coefficient (DSC), 95% Hausdorff distance, relative volume difference, Matthews correlation coefficient, precision, recall, and average symmetrical surface distance. We selected the highest-performing architecture and investigated how the amount of training data, use of skip connections, and data augmentation affected segmentation performance. In addition, we investigated whether segmentation on T1-weighted MRI was possible with FCNs trained on only T2-weighted and T2/T1-weighted image contrast. RESULTS: Overall, an FCN with a DenseNet201 encoder trained via cross-entropy minimization yielded the highest combined segmentation performance. For the 53 3D test MRI scans acquired with T2-weighted or T2/T1-weighted image contrast, the DSCs of the prostate, external urinary sphincter, seminal vesicles, rectum, and bladder were 0.90 ± 0.04, 0.70 ± 0.15, 0.80 ± 0.12, 0.91 ± 0.06, and 0.96 ± 0.04, respectively, after model fine-tuning. For the 5 T1-weighted images, the DSCs of these organs were 0.82 ± 0.07, 0.17 ± 0.15, 0.46 ± 0.21, 0.87 ± 0.06, and 0.88 ± 0.05, respectively. CONCLUSIONS: Machine segmentation of the prostate and surrounding anatomy on 3D MRIs acquired with different pulse sequences for MARS low-dose-rate prostate brachytherapy is possible with a single FCN.

Computer-aided Detection of Brain Metastases in T1-weighted MRI for Stereotactic Radiosurgery Using Deep Learning Single-Shot Detectors.

Background Brain metastases are manually identified during stereotactic radiosurgery (SRS) treatment planning, which is time consuming and potentially challenging. Purpose To develop and investigate deep learning (DL) methods for detecting brain metastasis with MRI to aid in treatment planning for SRS. Materials and Methods In this retrospective study, contrast material-enhanced three-dimensional T1-weighted gradient-echo MRI scans from patients who underwent gamma knife SRS from January 2011 to August 2018 were analyzed. Brain metastases were manually identified and contoured by neuroradiologists and treating radiation oncologists. DL single-shot detector (SSD) algorithms were constructed and trained to map axial MRI slices to a set of bounding box predictions encompassing metastases and associated detection confidences. Performances of different DL SSDs were compared for per-lesion metastasis-based detection sensitivity and positive predictive value (PPV) at a 50% confidence threshold. For the highest-performing model, detection performance was analyzed by using free-response receiver operating characteristic analysis. Results Two hundred sixty-six patients (mean age, 60 years ± 14 [standard deviation]; 148 women) were randomly split into 80% training and 20% testing groups (212 and 54 patients, respectively). For the testing group, sensitivity of the highest-performing (baseline) SSD was 81% (95% confidence interval [CI]: 80%, 82%; 190 of 234) and PPV was 36% (95% CI: 35%, 37%; 190 of 530). For metastases measuring at least 6 mm, sensitivity was 98% (95% CI: 97%, 99%; 130 of 132) and PPV was 36% (95% CI: 35%, 37%; 130 of 366). Other models (SSD with a ResNet50 backbone, SSD with focal loss, and RetinaNet) yielded lower sensitivities of 73% (95% CI: 72%, 74%; 171 of 234), 77% (95% CI: 76%, 78%; 180 of 234), and 79% (95% CI: 77%, 81%; 184 of 234), respectively, and lower PPVs of 29% (95% CI: 28%, 30%; 171 of 581), 26% (95% CI: 26%, 26%; 180 of 681), and 13% (95% CI: 12%, 14%; 184 of 1412). Conclusion Deep-learning single-shot detector models detected nearly all brain metastases that were 6 mm or larger with limited false-positive findings using postcontrast T1-weighted MRI. © RSNA, 2020 See also the editorial by Kikinis and Wells in this issue.

Development and clinical implementation of SeedNet: A sliding-window convolutional neural network for radioactive seed identification in MRI-assisted radiosurgery (MARS).

PURPOSE: To develop and evaluate a sliding-window convolutional neural network (CNN) for radioactive seed identification in MRI of the prostate after permanent implant brachytherapy. METHODS: Sixty-eight patients underwent prostate cancer low-dose-rate (LDR) brachytherapy using radioactive seeds stranded with positive contrast MR-signal seed markers and were scanned using a balanced steady-state free precession pulse sequence with and without an endorectal coil (ERC). A sliding-window CNN algorithm (SeedNet) was developed to scan the prostate images using 3D sub-windows and to identify the implanted radioactive seeds. The algorithm was trained on sub-windows extracted from 18 patient images. Seed detection performance was evaluated by computing precision, recall, F1 -score, false discovery rate, and false-negative rate. Seed localization performance was evaluated by computing the RMS error (RMSE) between the manually identified and algorithm-inferred seed locations. SeedNet was implemented into a clinical software package and evaluated on sub-windows extracted from 40 test patients. RESULTS: SeedNet achieved 97.6 ± 2.2% recall and 97.2 ± 1.9% precision for radioactive seed detection and 0.19 ± 0.04 mm RMSE for seed localization in the images acquired with an ERC. Without the ERC, the recall remained high, but the false-positive rate increased; the RMSE of the seed locations increased marginally. The clinical integration of SeedNet slightly increased the run-time, but the overall run-time was still low. CONCLUSION: SeedNet can be used to perform automated radioactive seed identification in prostate MRI after LDR brachytherapy. Image quality improvement through pulse sequence optimization is expected to improve SeedNet's performance when imaging without an ERC.

Parallel imaging compressed sensing for accelerated imaging and improved signal-to-noise ratio in MRI-based postimplant dosimetry of prostate brachytherapy.

PURPOSE: To investigate the feasibility of using parallel imaging compressed sensing (PICS) to reduce scan time and improve signal-to-noise ratio (SNR) in MRI-based postimplant dosimetry of prostate brachytherapy. METHODS AND MATERIALS: Ten patients underwent low-dose-rate prostate brachytherapy with radioactive seeds stranded with positive magnetic resonance-signal seed markers and were scanned on a Siemens 1.5T Aera. MRI comprised a fully balanced steady-state free precession sequence with two 18-channel external pelvic array coils with and without a rigid two-channel endorectal coil. The fully sampled data sets were retrospectively subsampled with increasing acceleration factors and reconstructed with parallel imaging and compressed sensing algorithms. The images were assessed in a blinded reader study by board-certified care providers. Rating scores were compared for statistically significant differences between reconstruction types. RESULTS: Images reconstructed from subsampling up to an acceleration factor of 4 with PICS demonstrated consistently sufficient quality for dosimetry with no apparent loss of SNR, anatomy depiction, or seed/marker conspicuity when compared to the fully sampled images. Images obtained with acceleration factors of 5 or 6 revealed reduced spatial resolution and seed marker contrast. Nevertheless, the reader study revealed that images obtained with an acceleration factor of up to 5 and reconstructed with PICS were adequate-to-good for postimplant dosimetry. CONCLUSIONS: Combined parallel imaging and compressed sensing can substantially reduce scan time in fully balanced steady-state free precession imaging of the prostate while maintaining adequate-to-good image quality for postimplant dosimetry. The saved scan time can be used for multiple signal averages and improved SNR, potentially obviating the need for an endorectal coil in MRI-based postimplant dosimetry.