Generating 3D images of VMAT plans for predictive models and activation maps associated with plan deliverability.

BACKGROUND: Intensity modulation with dynamic multi-leaf collimator (MLC) and monitor unit (MU) changes across control points (CPs) characterizes volumetric modulated arc therapy (VMAT). The increased uncertainty in plan deliverability required patient-specific quality assurance (PSQA), which remained inefficient upon Quality Assurance (QA) failure. To prevent waste before QA, plan complexity metrics (PCMs) and machine learning models with the metrics were generated, which were lack of providing CP-specific information upon QA failures. PURPOSE: By generating 3D images from digital imaging and comminications in medicine in radiation therapy (DICOM RT) plan, we proposed a predictive model that can estimate the deliverability of VMAT plans and visualize CP-specific regions associated with plan deliverability. METHODS: The patient cohort consisted of 259 and 190 cases for left- and right-breast VMAT treatments, which were split into 235 and 166 cases for training and 24 cases from each treatment for testing the networks. Three-channel 3D images generated from DICOM RT plans were fed into a DenseNet-based deep learning network. To reflect VMAT plan complexity as an image, the first two channels described MLC and MU variations between two consecutive CPs, while the last channel assigned the beam field size. The network output was defined as binary classified PSQA results, indicating deliverability. The predictive performance was assessed by accuracy, sensitivity, specificity, F1-score, and area under the curve (AUC). The gradient-weighted class activation map (Grad-CAM) highlighted the regions of CPs in VMAT plans associated with deliverability, compared against PCMs by Spearman correlation. RESULTS: The DenseNet-based predictive model yielded AUCs of 92.2% and 93.8%, F1-scores of 97.0% and 93.8% and accuracies of 95.8% and 91.7% for the left- and right-breast VMAT cases. Additionally, the specificity of 87.5% for both cases indicated that the predictive model accurately detected QA failing cases. The activation maps significantly differentiated QA failing-labeled from passing-labeled classes for the non-deliverable cases. The PCM with the highest correlation to the Grad-CAM varied from patient cases, implying that plan deliverability would be considered patient-specific. CONCLUSION: This work demonstrated that the deep learning-based network based on visualization of dynamic VMAT plan information successfully predicted plan deliverability, which also provided control-point specific planning parameter information associated with plan deliverability in a patient-specific manner.

Artificial Intelligence in Spinal Imaging and Patient Care: A Review of Recent Advances.

Artificial intelligence (AI) is transforming spinal imaging and patient care through automated analysis and enhanced decision-making. This review presents a clinical task-based evaluation, highlighting the specific impact of AI techniques on different aspects of spinal imaging and patient care. We first discuss how AI can potentially improve image quality through techniques like denoising or artifact reduction. We then explore how AI enables efficient quantification of anatomical measurements, spinal curvature parameters, vertebral segmentation, and disc grading. This facilitates objective, accurate interpretation and diagnosis. AI models now reliably detect key spinal pathologies, achieving expert-level performance in tasks like identifying fractures, stenosis, infections, and tumors. Beyond diagnosis, AI also assists surgical planning via synthetic computed tomography generation, augmented reality systems, and robotic guidance. Furthermore, AI image analysis combined with clinical data enables personalized predictions to guide treatment decisions, such as forecasting spine surgery outcomes. However, challenges still need to be addressed in implementing AI clinically, including model interpretability, generalizability, and data limitations. Multicenter collaboration using large, diverse datasets is critical to advance the field further. While adoption barriers persist, AI presents a transformative opportunity to revolutionize spinal imaging workflows, empowering clinicians to translate data into actionable insights for improved patient care.

HaN-Seg: The head and neck organ-at-risk CT and MR segmentation challenge.

BACKGROUND AND PURPOSE: To promote the development of auto-segmentation methods for head and neck (HaN) radiation treatment (RT) planning that exploit the information of computed tomography (CT) and magnetic resonance (MR) imaging modalities, we organized HaN-Seg: The Head and Neck Organ-at-Risk CT and MR Segmentation Challenge. MATERIALS AND METHODS: The challenge task was to automatically segment 30 organs-at-risk (OARs) of the HaN region in 14 withheld test cases given the availability of 42 publicly available training cases. Each case consisted of one contrast-enhanced CT and one T1-weighted MR image of the HaN region of the same patient, with up to 30 corresponding reference OAR delineation masks. The performance was evaluated in terms of the Dice similarity coefficient (DSC) and 95-percentile Hausdorff distance (HD95), and statistical ranking was applied for each metric by pairwise comparison of the submitted methods using the Wilcoxon signed-rank test. RESULTS: While 23 teams registered for the challenge, only seven submitted their methods for the final phase. The top-performing team achieved a DSC of 76.9 % and a HD95 of 3.5 mm. All participating teams utilized architectures based on U-Net, with the winning team leveraging rigid MR to CT registration combined with network entry-level concatenation of both modalities. CONCLUSION: This challenge simulated a real-world clinical scenario by providing non-registered MR and CT images with varying fields-of-view and voxel sizes. Remarkably, the top-performing teams achieved segmentation performance surpassing the inter-observer agreement on the same dataset. These results set a benchmark for future research on this publicly available dataset and on paired multi-modal image segmentation in general.

Assessment of traumatic mandibular nerve using MR neurography sequence: a preliminary study.

BACKGROUND: Iatrogenic mandibular nerve damage resulting from oral surgeries and dental procedures is painful and a formidable challenge for patients and oral surgeons alike, mainly because the absence of objective and quantitative methods for diagnosing nerve damage renders treatment and compensation ambiguous while often leading to medico-legal disputes. The aim of this study was to examine discriminating factors of traumatic mandibular nerve within a specific magnetic resonance imaging (MRI) protocol and to suggest tangible diagnostic criteria for peripheral trigeminal nerve injury. METHODS: Twenty-six patients with ipsilateral mandibular nerve trauma underwent T2 Flex water, 3D short tau inversion recovery (STIR), and diffusion-weighted imaging (DWI) acquired by periodically rotating overlapping parallel lines with enhanced reconstruction (PROPELLER) pulse sequences; 26 injured nerves were thus compared with contra-lateral healthy nerves at anatomically corresponding sites. T2 Flex apparent signal to noise ratio (FSNR), T2 Flex apparent nerve-muscle contrast to noise ratio (FNMCNR) 3D STIR apparent signal to noise ratio (SSNR), 3D STIR apparent nerve-muscle contrast to noise ratio (SNMCNR), apparent diffusion coefficient (ADC) and area of cross-sectional nerve (Area) were evaluated. RESULTS: Mixed model analysis revealed FSNR and FNMCNR to be the dual discriminators for traumatized mandibular nerve (p < 0.05). Diagnostic performance of both parameters was also determined with area under the receiver operating characteristic curve (AUC for FSNR = 0.712; 95% confidence interval [CI]: 0.5660, 0.8571 / AUC for FNMCNR = 0.7056; 95% confidence interval [CI]: 1.011, 1.112). CONCLUSIONS: An increase in FSNR and FNMCNR within our MRI sequence seems to be accurate indicators of the presence of traumatic nerve. This prospective study may serve as a foundation for sophisticated model diagnosing trigeminal nerve trauma within large patient cohorts.

Radiomics-based hybrid model for predicting radiation pneumonitis: A systematic review and meta-analysis.

PURPOSE: This study reviewed and meta-analyzed evidence on radiomics-based hybrid models for predicting radiation pneumonitis (RP). These models are crucial for improving thoracic radiotherapy plans and mitigating RP, a common complication of thoracic radiotherapy. We examined and compared the RP prediction models developed in these studies with the radiomics features employed in RP models. METHODS: We systematically searched Google Scholar, Embase, PubMed, and MEDLINE for studies published up to April 19, 2024. Sixteen studies met the inclusion criteria. We compared the RP prediction models developed in these studies and the radiomics features employed. RESULTS: Radiomics, as a single-factor evaluation, achieved an area under the receiver operating characteristic curve (AUROC) of 0.73, accuracy of 0.69, sensitivity of 0.64, and specificity of 0.74. Dosiomics achieved an AUROC of 0.70. Clinical and dosimetric factors showed lower performance, with AUROCs of 0.59 and 0.58. Combining clinical and radiomic factors yielded an AUROC of 0.78, while combining dosiomic and radiomics factors produced an AUROC of 0.81. Triple combinations, including clinical, dosimetric, and radiomics factors, achieved an AUROC of 0.81. The study identifies key radiomics features, such as the Gray Level Co-occurrence Matrix (GLCM) and Gray Level Size Zone Matrix (GLSZM), which enhance the predictive accuracy of RP models. CONCLUSIONS: Radiomics-based hybrid models are highly effective in predicting RP. These models, combining traditional predictive factors with radiomic features, particularly GLCM and GLSZM, offer a clinically feasible approach for identifying patients at higher RP risk. This approach enhances clinical outcomes and improves patient quality of life. PROTOCOL REGISTRATION: The protocol of this study was registered on PROSPERO (CRD42023426565).

Dosimetric analysis of six whole-breast irradiation techniques in supine and prone positions.

In breast cancer radiation therapy, minimizing radiation-related risks and toxicity is vital for improving life expectancy. Tailoring radiotherapy techniques and treatment positions can reduce radiation doses to normal organs and mitigate treatment-related toxicity. This study entailed a dosimetric comparison of six different external beam whole-breast irradiation techniques in both supine and prone positions. We selected fourteen breast cancer patients, generating six treatment plans in both positions per patient. We assessed target coverage and organs at risk (OAR) doses to evaluate the impact of treatment techniques and positions. Excess absolute risk was calculated to estimate potential secondary cancer risk in the contralateral breast, ipsilateral lung, and contralateral lung. Additionally, we analyzed the distance between the target volume and OARs (heart and ipsilateral lung) while considering the treatment position. The results indicate that prone positioning lowers lung exposure in X-ray radiotherapy. However, particle beam therapies (PBTs) significantly reduce the dose to the heart and ipsilateral lung regardless of the patient's position. Notably, negligible differences were observed between arc-delivery and static-delivery PBTs in terms of target conformity and OAR sparing. This study provides critical dosimetric evidence to facilitate informed decision-making regarding treatment techniques and positions.

Development of shielding evaluation and management program for O-ring type linear accelerators.

The shielding parameters can vary depending on the geometrical structure of the linear accelerators (LINAC), treatment techniques, and beam energies. Recently, the introduction of O-ring type linear accelerators is increasing. The objective of this study is to evaluate the shielding parameters of new type of linac using a dedicated program developed by us named ORSE (O-ring type Radiation therapy equipment Shielding Evaluation). The shielding evaluation was conducted for a total of four treatment rooms including Elekta Unity, Varian Halcyon, and Accuray Tomotherapy. The developed program possesses the capability to calculate transmitted dose, maximum treatable patient capacity, and shielding wall thickness based on patient data. The doses were measured for five days using glass dosimeters to compare with the results of program. The IMRT factors and use factors obtained from patient data showed differences of up to 65.0% and 33.8%, respectively, compared to safety management report. The shielding evaluation conducted in each treatment room showed that the transmitted dose at every location was below 1% of the dose limit. The results of program and measurements showed a maximum difference of 0.003 mSv/week in transmitted dose. The ORSE program allows for the shielding evaluation results to the clinical environment of each institution based on patient data.

Dose-toxicity surface histogram-based prediction of radiation dermatitis severity and shape.

Objective.This study aimed to develop a new approach to predict radiation dermatitis (RD) by using the skin dose distribution in the actual area of RD occurrence to determine the predictive dose by grade.Approach.Twenty-three patients with head and neck cancer treated with volumetric modulated arc therapy were prospectively and retrospectively enrolled. A framework was developed to segment the RD occurrence area in skin photography by matching the skin surface image obtained using a 3D camera with the skin dose distribution. RD predictive doses were generated using the dose-toxicity surface histogram (DTH) calculated from the skin dose distribution within the segmented RD regions classified by severity. We then evaluated whether the developed DTH-based framework could visually predict RD grades and their occurrence areas and shapes according to severity.Main results.The developed framework successfully generated the DTH for three different RD severities: faint erythema (grade 1), dry desquamation (grade 2), and moist desquamation (grade 3); 48 DTHs were obtained from 23 patients: 23, 22, and 3 DTHs for grades 1, 2, and 3, respectively. The RD predictive doses determined using DTHs were 28.9 Gy, 38.1 Gy, and 54.3 Gy for grades 1, 2, and 3, respectively. The estimated RD occurrence area visualized by the DTH-based RD predictive dose showed acceptable agreement for all grades compared with the actual RD region in the patient. The predicted RD grade was accurate, except in two patients.Significance. The developed DTH-based framework can classify and determine RD predictive doses according to severity and visually predict the occurrence area and shape of different RD severities. The proposed approach can be used to predict the severity and shape of potential RD in patients and thus aid physicians in decision making.

Clinical feasibility of deep learning-based synthetic CT images from T2-weighted MR images for cervical cancer patients compared to MRCAT.

This work aims to investigate the clinical feasibility of deep learning-based synthetic CT images for cervix cancer, comparing them to MR for calculating attenuation (MRCAT). Patient cohort with 50 pairs of T2-weighted MR and CT images from cervical cancer patients was split into 40 for training and 10 for testing phases. We conducted deformable image registration and Nyul intensity normalization for MR images to maximize the similarity between MR and CT images as a preprocessing step. The processed images were plugged into a deep learning model, generative adversarial network. To prove clinical feasibility, we assessed the accuracy of synthetic CT images in image similarity using structural similarity (SSIM) and mean-absolute-error (MAE) and dosimetry similarity using gamma passing rate (GPR). Dose calculation was performed on the true and synthetic CT images with a commercial Monte Carlo algorithm. Synthetic CT images generated by deep learning outperformed MRCAT images in image similarity by 1.5% in SSIM, and 18.5 HU in MAE. In dosimetry, the DL-based synthetic CT images achieved 98.71% and 96.39% in the GPR at 1% and 1 mm criterion with 10% and 60% cut-off values of the prescription dose, which were 0.9% and 5.1% greater GPRs over MRCAT images.

Navigation-guided endoscopic lumbar decompression on foramen and lateral recess in advanced scoliosis.

An 84-year-old woman presented with left leg radiating pain for 18 months and a numeric rating scale score of 8. From examination, motoric on left knee extension was grade 4, with dysesthesia and numbness along the left anterolateral thigh. Imaging showed left L3-4 foraminal and lateral recess stenosis with severe-degree scoliosis. The patient underwent navigation-guided endoscopic transforaminal foraminotomy and lateral recess decompression on the left L3-4 level with a good outcome. Three-years' follow-up showed a well-maintained clinical outcome and coronal sagittal balance. This video explores navigation-guided endoscopic lumbar decompression for neural compression in advanced scoliosis. Further research is encouraged to establish long-term efficacy. The video can be found here: https://stream.cadmore.media/r10.3171/2024.1.FOCVID23195.

Assessment of radio-activation using spectroscopy in medical linear accelerators.

In radiotherapy, when photon energy exceeding 8 MV is utilized, photoneutrons can activate the components within the gantry of the linear accelerator (linac). At the end of the linac's lifecycle, radiation workers are tasked with its dismantling and disposal, potentially exposing them to unintentional radiation. This study aims to identify and measure the radioisotopes generated by this activation through spectroscopy, and to evaluate the effective dose rate. We selected nine medical linacs, considering various factors such as manufacturer (Siemens, Varian, and Elekta), model, energy, period of operation, and workload. We identified the radionuclides in the linac head by employing an in situ high-purity germanium (HPGe) detector. Spectroscopy and dose-rate measurements were conducted post-shutdown. We also measured the dose rates at the beam-exit window following irradiation with 10 MV and 15 MV photon beams. As a result of the spectroscopy, we identified approximately 20 nuclides including those with half-lives of 100 days or longer, such as 54Mn, 60Co, 65Zn, 122Sb, and 198Au. The dose rate measurements after 10 MV irradiation decreased to the background level in 10 min. By contrast, on 15 MV irradiation, the dose rate was 628 nSv/h after 10 min and decreased to 268 nSv/h after 1.5 hours. It was confirmed that the difference in the level of radiation and the type of nuclide depends on the period of use, energy, and workload. However, the type of nuclide does not differ significantly between the linacs. It is necessary to propose appropriate guidelines for the safety of workers, and disposal/move-install should be planned while taking into consideration the equipment's energy usage rate.

Combined unilateral biportal endoscopy and video-assisted thoracoscopic surgery for complete excision of a T3-T4 right ganglioneuroma.

Ganglioneuroma (GN) is a rare solid neoplasm developing from neural crest cells of sympathetic ganglia or adrenal medulla. It usually presents as an asymptomatic mass in the retroperitoneal space and mediastinum. Resection through open surgery or minimal access is recommended. The video illustrates the case of a 23-year-old female with an incidental finding of thoracic GN. The authors performed a combined, staged approach to ensure complete resection, which involved unilateral T3-4 biportal endoscopy (UBE) for rhizotomy and nerve root decompression, followed by video-assisted thoracoscopic surgery (VATS) for complete excision. The procedure was uneventful, with full recovery and no postoperative complications. The video can be found here: https://stream.cadmore.media/r10.3171/2024.2.FOCVID23210.

Dosimetric Comparison of CPAP and DIBH for Left-sided Breast Cancer Radiation Therapy.

Purpose: Despite the increasing interest in using continuous positive airway pressure (CPAP) in radiation therapy (RT), direct comparisons with the more widely used deep inspiration breath-hold (DIBH) have been limited. This planning study aimed to offer comprehensive geometric and dosimetric evidence by comparing CPAP and DIBH-based RT plans. Materials and Methods: A retrospective data set of 35 patients with left-sided breast cancer with planning computed tomography scans under three breathing conditions (free breathing (FB), CPAP with 10 cmH2O pressure, and DIBH) was collected. Volumetric arc therapy plans aimed for 95% dose coverage to 95% of the planning target volume with a maximum dose below 107%. A comparative dosimetric analysis among the three plans was conducted. Additionally, geometric differences were assessed by calculating the minimum distance between the heart and the clinical target volume (CTV) in each planning computed tomography. Results: CPAP and DIBH plans demonstrated comparable mean heart doses (1.05 Gy), which were significantly lower than the FB plan (1.34 Gy). The maximum dose to the left anterior descending artery was smallest in the CPAP plan (4.44 Gy), followed by DIBH (4.73 Gy) and FB (7.33 Gy) plans. Other organ-at-risk doses for CPAP and DIBH were similar, with mean contralateral breast doses of 2.27 and 2.21 Gy, mean ipsilateral lung doses of 4.09 and 4.08 Gy, V20 at 6.11% and 6.31%, and mean contralateral lung doses of 0.94 and 0.92 Gy, respectively. No significant difference was found in the minimum heart-to-CTV distance between CPAP and DIBH. DIBH exhibited the greatest lung volume (3908 cc), followed by CPAP (3509 cc), and FB(2703 cc). Conclusions: The comparison between CPAP and DIBH shows their similarity in both geometric and dosimetric aspects, providing strong evidence for CPAP's effectiveness and feasibility in RT. This suggests its potential as an alternative to DIBH for patients with left-sided breast cancer.

Polytomous Rasch Analyses of Surgeons' Decision-Making on Choice of Procedure in Endoscopic Lumbar Spinal Stenosis Decompression Surgeries.

BACKGROUND: With the growing prevalence of lumbar spinal stenosis, endoscopic surgery, which incorporates techniques such as transforaminal, interlaminar, and unilateral biportal (UBE) endoscopy, is increasingly considered. However, the patient selection criteria are debated among spine surgeons. OBJECTIVE: This study used a polytomous Rasch analysis to evaluate the factors influencing surgeon decision-making in selecting patients for endoscopic surgical treatment of lumbar spinal stenosis. METHODS: A comprehensive survey was distributed to a representative sample of 296 spine surgeons. Questions encompassed various patient-related and clinical factors, and responses were captured on a logit scale graphically displaying person-item maps and category probability curves for each test item. Using a Rasch analysis, the data were subsequently analyzed to determine the latent traits influencing decision-making. RESULTS: The Rasch analysis revealed that surgeons' preferences for transforaminal, interlaminar, and UBE techniques were easily influenced by comfort level and experience with the endoscopic procedure and patient-related factors. Harder-to-agree items included technological aspects, favorable clinical outcomes, and postoperative functional recovery and rehabilitation. Descriptive statistics suggested interlaminar as the best endoscopic spinal stenosis decompression technique. However, logit person-item analysis integral to the Rasch methodology showed highest intensity for transforaminal followed by interlaminar endoscopic lumbar stenosis decompression. The UBE technique was the hardest to agree on with a disordered person-item analysis and thresholds in category probability curve plots. CONCLUSION: Surgeon decision-making in selecting patients for endoscopic surgery for lumbar spinal stenosis is multifaceted. While the framework of clinical guidelines remains paramount, on-the-ground experience-based factors significantly influence surgeons' selection of patients for endoscopic lumbar spinal stenosis surgeries. The Rasch methodology allows for a more granular psychometric evaluation of surgeon decision-making and accounts better for years-long experience that may be lost in standardized clinical guideline development. This new approach to assessing spine surgeons' thought processes may improve the implementation of evidence-based protocol change dictated by technological advances was endorsed by the Interamerican Society for Minimally Invasive Spine Surgery (SICCMI), the International Society for Minimal Intervention in Spinal Surgery (ISMISS), the Mexican Spine Society (AMCICO), the Brazilian Spine Society (SBC), the Society for Minimally Invasive Spine Surgery (SMISS), the Korean Minimally Invasive Spine Society (KOMISS), and the International Society for the Advancement of Spine Surgery (ISASS).

Comparative Analysis of Learning Curve, Complexity, Psychological Stress, and Work Relative Value Units for CPT 62380 Endoscopic Lumbar Spinal Decompression vs Traditional Lumbar Spine Surgeries: A Paired Rasch Survey Study.

BACKGROUND: Effective 1 January 2017, single-level endoscopic lumbar discectomy received a Category I Current Procedural Terminology (CPT) code 62380. However, no work relative value units (RVUs) are currently assigned to the procedure. An international team of endoscopic spine surgeons conducted a study, endorsed by several spine societies, analyzing the learning curve, difficulty, psychological intensity, and estimated work RVUs of endoscopic lumbar spinal decompression compared with other common lumbar spine surgeries. METHODS: A survey comparing CPT 62380 to 10 other comparator CPT codes reflective of common spine surgeries was developed to assess the work RVUs in terms of learning curve, difficulty, psychological intensity, and work effort using a paired Rasch method. RESULTS: The survey was sent to 542 spine specialists. Of 322 respondents, 150 completed the survey for a 43.1% completion rate. Rasch analysis of the submitted responses statistically corroborated common knowledge that the learning curve with lumbar endoscopic spinal surgery is steeper and more complex than with traditional translaminar lumbar decompression surgeries. It also showed that the psychological stress and mental and work effort with the lumbar endoscopic decompression surgery were perceived to be higher by responding spine surgeons compared with posterior comparator decompression and fusion surgeries and even posterior interbody and posterolateral fusion surgeries. The regression analysis of work effort vs procedural difficulty showed the real-world evaluation of the lumbar endoscopic decompression surgery described in CPT code 62380 with a calculated work RVU of 18.2464. CONCLUSION: The Rasch analysis suggested the valuation for the endoscopic lumbar decompression surgery should be higher than for standard lumbar surgeries: 111.1% of the laminectomy with exploration and/or decompression of spinal cord and/or cauda equina (CPT 63005), 118.71% of the laminectomy code (CPT 63047), which includes foraminotomy and facetectomy, 152.1% of the hemilaminectomy code (CPT 63030), and 259.55% of the interlaminar or interspinous process stabilization/distraction without decompression code (CPT 22869). This research methodology was endorsed by the Interamerican Society for Minimally Invasive Spine Surgery (SICCMI), the Mexican Society of Spinal Surgeons (AMCICO), the International Society For Minimally Invasive Spine Surgery (ISMISS), the Brazilian Spine Society (SBC), the Society for Minimally Invasive Spine Surgery (SMISS), the Korean Minimally Invasive Spine Surgery (KOMISS), and the International Society for the Advancement of Spine Surgery (ISASS). CLINICAL RELEVANCE: This study provides an updated reimbursement recommendation for endoscopic spine surgery. LEVEL OF EVIDENCE: Level 3.

Inter-scanner super-resolution of 3D cine MRI using a transfer-learning network for MRgRT.

Objective. Deep-learning networks for super-resolution (SR) reconstruction enhance the spatial-resolution of 3D magnetic resonance imaging (MRI) for MR-guided radiotherapy (MRgRT). However, variations between MRI scanners and patients impact the quality of SR for real-time 3D low-resolution (LR) cine MRI. In this study, we present a personalized super-resolution (psSR) network that incorporates transfer-learning to overcome the challenges in inter-scanner SR of 3D cine MRI.Approach: Development of the proposed psSR network comprises two-stages: (1) a cohort-specific SR (csSR) network using clinical patient datasets, and (2) a psSR network using transfer-learning to target datasets. The csSR network was developed by training on breath-hold and respiratory-gated high-resolution (HR) 3D MRIs and their k-space down-sampled LR MRIs from 53 thoracoabdominal patients scanned at 1.5 T. The psSR network was developed through transfer-learning to retrain the csSR network using a single breath-hold HR MRI and a corresponding 3D cine MRI from 5 healthy volunteers scanned at 0.55 T. Image quality was evaluated using the peak-signal-noise-ratio (PSNR) and the structure-similarity-index-measure (SSIM). The clinical feasibility was assessed by liver contouring on the psSR MRI using an auto-segmentation network and quantified using the dice-similarity-coefficient (DSC).Results. Mean PSNR and SSIM values of psSR MRIs were increased by 57.2% (13.8-21.7) and 94.7% (0.38-0.74) compared to cine MRIs, with the reference 0.55 T breath-hold HR MRI. In the contour evaluation, DSC was increased by 15% (0.79-0.91). Average time consumed for transfer-learning was 90 s, psSR was 4.51 ms per volume, and auto-segmentation was 210 ms, respectively.Significance. The proposed psSR reconstruction substantially increased image and segmentation quality of cine MRI in an average of 215 ms across the scanners and patients with less than 2 min of prerequisite transfer-learning. This approach would be effective in overcoming cohort- and scanner-dependency of deep-learning for MRgRT.

Development and Feasibility Evaluation of Smart Cancer Care 2.0 Based on Patient-Reported Outcomes for Post-Discharge Management of Patients with Cancer.

Purpose: A "Smart Cancer Care" platform that integrates patient-reported outcomes (PROs) with management has been established in Korea. This study focused on improving health behaviors and connecting patients to welfare services by introducing and assessing the feasibility of "Smart Cancer Care 2.0," an enhanced version designed for monitoring complications post-cancer treatment. Materials and Methods: Smart Cancer Care 2.0 was developed by conducting a literature review and consulting with expert panels to identify symptoms or variables requiring monitoring and management guidelines based on the treatment type. Qualitative and quantitative surveys were conducted to assess the feasibility of the app and web system based on the experiences of patients with cancer and healthcare workers. Results: A total of 81 symptoms or variables (chemotherapy-, surgery-, radiotherapy-, rehabilitation-, and health management-related) were selected for management in Smart Cancer Care 2.0. PROs for these symptoms were basically categorized into three severity grades: (1) preventive management, (2) self-treatment, and (3) consultation with a healthcare worker or visit to a healthcare institution. The overall mean scores in the feasibility evaluation by patients and healthcare workers were 3.83 and 3.90 points, respectively, indicating high usefulness. Conclusion: Smart Cancer Care 2.0 leverages the existing ICT-based platform, Smart Cancer Care, and further includes health behaviors and welfare services. Smart Cancer Care 2.0 may play a crucial role in establishing a comprehensive post-discharge management system for patients with cancer as it provides suitable interventions based on patients' responses and allows the regularly collected PROs to be easily viewed for streamlined care.

Experience of Implementing Deep Learning-Based Automatic Contouring in Breast Radiation Therapy Planning: Insights From Over 2000 Cases.

PURPOSE: This study evaluated the impact and clinical utility of an auto-contouring system for radiation therapy treatments. METHODS AND MATERIALS: The auto-contouring system was implemented in 2019. We evaluated data from 2428 patients who underwent adjuvant breast radiation therapy before and after the system's introduction. We collected the treatment's finalized contours, which were reviewed and revised by a multidisciplinary team. After implementation, the treatment contours underwent a finalization process that involved manual review and adjustment of the initial auto-contours. For the preimplementation group (n = 369), auto-contours were generated retrospectively. We compared the auto-contours and final contours using the Dice similarity coefficient (DSC) and the 95% Hausdorff distance (HD95). RESULTS: We analyzed 22,215 structures from final and corresponding auto-contours. The final contours were generally larger, encompassing more slices in the superior or inferior directions. Among organs at risk (OAR), the heart, esophagus, spinal cord, and contralateral breast demonstrated significantly increased DSC and decreased HD95 postimplementation (all P < .05), except for the lungs, which presented inaccurate segmentation. Among target volumes, CTVn_L2, L3, L4, and the internal mammary node showed increased DSC and decreased HD95 postimplementation (all P < .05), although the increase was less pronounced than the OAR outcomes. The analysis also covered factors contributing to significant differences, pattern identification, and outlier detection. CONCLUSIONS: In our study, the adoption of an auto-contouring system was associated with an increased reliance on automated settings, underscoring its utility and the potential risk of automation bias. Given these findings, we underscore the importance of considering the integration of stringent risk assessments and quality management strategies as a precautionary measure for the optimal use of such systems.

Evaluation of the deliverability of dynamic conformal arc therapy (DCAT) by gantry wobble and its influence on dose.

We aimed to investigate the deliverability of dynamic conformal arc therapy (DCAT) by gantry wobble owing to the intrinsic inter-segment break of the Elekta linear accelerator (LINAC) and its adverse influence on the dose to the patient. The deliverability of DCAT was evaluated according to the plan parameters, which affect the gantry rotation speed and resultant positional inaccuracies; the deliverability according to the number of control points and dose rates was investigated by using treatment machine log files and dosimetry devices, respectively. A non-negligible degradation in DCAT deliverability due to gantry wobble was observed in both the treatment machine log files and dosimetry devices. The resulting dose-delivery error occurred below a certain number of control points or above a certain dose rate. Dose simulations in the patient domain showed a similar impact on deteriorated deliverability. For targets located primarily in the isocenter, the dose differences were negligible, whereas for organs at risk located mainly off-isocenter, the dose differences were significant up to - 8.77%. To ensure safe and accurate radiotherapy, optimal plan parameters should be selected, and gantry angle-specific validations should be conducted before treatment.