Characterization of spatial integrity with active and passive implants in a low-field magnetic resonance linear accelerator scanner.

Background and Purpose: Standard imaging protocols can guarantee the spatial integrity of magnetic resonance (MR) images utilized in radiotherapy. However, the presence of metallic implants can significantly compromise this integrity. Our proposed method aims at characterizing the geometric distortions induced by both passive and active implants commonly encountered in planning images obtained from a low-field 0.35 T MR-linear accelerator (LINAC). Materials and Methods: We designed a spatial integrity phantom defining 1276 control points and covering a field of view of 20x20x20 cm3. This phantom was scanned in a water tank with and without different implants used in hip and shoulder arthroplasty procedures as well as with active cardiac stimulators. The images were acquired with the clinical planning sequence (balanced steady-state free-precession, resolution 1.5x1.5x1.5 mm3). Spatial integrity was assessed by the Euclidian distance between the control point detected on the image and their theoretical locations. A first plane free of artefact (FPFA) was defined to evaluate the spatial integrity beyond the larger banding artefact. Results: In the region extending up to 20 mm from the largest banding artefacts, the tested passive and active implants could cause distortions up to 2 mm and 3 mm, respectively. Beyond this region the spatial integrity was recovered and the image could be considered as unaffected by the implants. Conclusions: We characterized the impact of common implants on a low field MR-LINAC planning sequence. These measurements could support the creation of extra margin while contouring organs at risk and target volumes in the vicinity of implants.

Assessment of bias in scoring of AI-based radiotherapy segmentation and planning studies using modified TRIPOD and PROBAST guidelines as an example.

BACKGROUND AND PURPOSE: Studies investigating the application of Artificial Intelligence (AI) in the field of radiotherapy exhibit substantial variations in terms of quality. The goal of this study was to assess the amount of transparency and bias in scoring articles with a specific focus on AI based segmentation and treatment planning, using modified PROBAST and TRIPOD checklists, in order to provide recommendations for future guideline developers and reviewers. MATERIALS AND METHODS: The TRIPOD and PROBAST checklist items were discussed and modified using a Delphi process. After consensus was reached, 2 groups of 3 co-authors scored 2 articles to evaluate usability and further optimize the adapted checklists. Finally, 10 articles were scored by all co-authors. Fleiss' kappa was calculated to assess the reliability of agreement between observers. RESULTS: Three of the 37 TRIPOD items and 5 of the 32 PROBAST items were deemed irrelevant. General terminology in the items (e.g., multivariable prediction model, predictors) was modified to align with AI-specific terms. After the first scoring round, further improvements of the items were formulated, e.g., by preventing the use of sub-questions or subjective words and adding clarifications on how to score an item. Using the final consensus list to score the 10 articles, only 2 out of the 61 items resulted in a statistically significant kappa of 0.4 or more demonstrating substantial agreement. For 41 items no statistically significant kappa was obtained indicating that the level of agreement among multiple observers is due to chance alone. CONCLUSION: Our study showed low reliability scores with the adapted TRIPOD and PROBAST checklists. Although such checklists have shown great value during development and reporting, this raises concerns about the applicability of such checklists to objectively score scientific articles for AI applications. When developing or revising guidelines, it is essential to consider their applicability to score articles without introducing bias.

Stereotactic body radiotherapy for oligoprogression with or without switch of systemic therapy.

Background: Oligoprogression is defined as cancer progression of a limited number of metastases under active systemic therapy. The role of metastasis-directed therapy, using stereotactic body radiotherapy (SBRT), is controversial as is the continuation versus switch of systemic therapy. We report outcomes of oligoprogressive patients after SBRT, and compare those patients that continued or switched their current line of systemic therapy. Material/Methods: We included patients who developed up to 5 progressive extracranial metastases under systemic therapy for any solid organ malignancy and were treated with SBRT to all lesions at our institution between 01/2014 and 12/2019. Overall survival (OS) and progression-free survival (PFS) were analyzed using the Kaplan-Meier method, and the interval to the next systemic therapy line determined using cumulative incidence functions. Multivariable Cox regression models were used to analyze the influence of baseline and post-progression variables on OS, PFS and survival with the next systemic therapy after SBRT. Results: Among 135 patients with oligoprogressive disease of which the most common primary tumor was lung cancer (n = 46, 34.1 %), 96 continued their current line of systemic therapy after oligoprogression. Among 39 who switched systemic therapy, 28 (71.8 %) paused or discontinued, while 11 (28.2 %) immediately started another systemic treatment. After a median follow-up of 27.2 months, patients that switched and those who continued systemic therapy after oligoprogression had comparable median OS (32.1 vs. 38.2 months, p = 0.47) and PFS (4.3 vs. 3.4 months, p = 0.6). The intervals to the next systemic therapy line were comparable between both cohorts (p = 0.6). An ECOG performance status of 2 and immediately starting a new systemic therapy after oligoprogression were associated with a poorer survival without next systemic therapy, while the de-novo OMD state was associated with better survival without next systemic therapy compared to the induced state. Conclusion: Oncological outcomes of patients that continued or switched systemic therapy after SBRT for oligoprogression were comparable, potentially indicating that further lines of treatment may be safely delayed in selected cases.

Knowledge-based versus deep learning based treatment planning for breast radiotherapy.

Background and Purpose: To improve radiotherapy (RT) planning efficiency and plan quality, knowledge-based planning (KBP) and deep learning (DL) solutions have been developed. We aimed to make a direct comparison of these models for breast cancer planning using the same training, validation, and testing sets. Materials and Methods: Two KBP models were trained and validated with 90 RT plans for left-sided breast cancer with 15 fractions of 2.6 Gy. The versions either used the full dataset (non-clean model) or a cleaned dataset (clean model), thus eliminating geometric and dosimetric outliers. Results were compared with a DL U-net model (previously trained and validated with the same 90 RT plans) and manually produced RT plans, for the same independent dataset of 15 patients. Clinically relevant dose volume histogram parameters were evaluated according to established consensus criteria. Results: Both KBP models underestimated the mean heart and lung dose equally 0.4 Gy (0.3-1.1 Gy) and 1.4 Gy (1.1-2.8 Gy) compared to the clinical plans 0.8 Gy (0.5-1.8 Gy) and 1.7 Gy (1.3-3.2 Gy) while in the final calculations the mean lung dose was higher 1.9-2.0 Gy (1.5-3.5 Gy) for both KPB models. The U-Net model resulted in a mean planning target volume dose of 40.7 Gy (40.4-41.3 Gy), slightly higher than the clinical plans 40.5 Gy (40.1-41.0 Gy). Conclusions: Only small differences were observed between the estimated and final dose calculation and the clinical results for both KPB models and the DL model. With a good set of breast plans, the data cleaning module is not needed and both KPB and DL models lead to clinically acceptable results.

Patterns of metastatic spread and tumor burden in unselected cancer patients using PET imaging: Implications for the oligometastatic spectrum theory.

Introduction and background: Metastatic disease has been proposed as a continuum, with no clear cut-off between oligometastatic and polymetastatic disease. This study aims to quantify tumor burden and patterns of spread in unselected metastatic cancer patients referred for PET-based staging, response assessment of restaging. Materials and methods: All oncological fluorodeoxyglucose (FDG-) and prostate-specific membrane antigen (PSMA-) positron emission tomography (PET) scans conducted at a single academic center in 2020 were analyzed. Imaging reports of all patients with metastatic disease were reviewed and assessed. Results: For this study, 7,000 PET scans were screened. One third of PET scans (n = 1,754; 33 %) from 1,155 unique patients showed presence of metastatic disease from solid malignancies, of which 601 (52 %) and 554 (48 %) were classified as oligometastatic (maximum 5 metastases) and polymetastatic (>5 metastases), respectively. Lung and pleural cancer, skin cancer, and breast cancer were the most common primary tumor histologies with 132 (23.8 %), 88 (15.9 %), and 72 (13.0 %) cases, respectively. Analysis of the number of distant metastases showed a strong bimodal distribution of the metastatic burden with 26 % of patients having one solitary metastasis and 43 % of patients harboring >10 metastases. Yet, despite 43 % of polymetastatic patients having >10 distant metastases, their pattern of distribution was restricted to one or two organs in about two thirds of patients, and there was no association between the number of distant metastases and the number of involved organs. Conclusion: The majority of metastatic cancer patients are characterized by either a solitary metastasis or a high tumor burden with >10 metastases, the latter was often associated with affecting a limited number of organs. These findings support both the spectrum theory of metastasis and the seed and soil hypothesis and can support in designing the next generation of clinical trials in the field of oligometastatic disease.

Dosimetric Analysis of Proximal Bronchial Tree Subsegments to Assess The Risk of Severe Toxicity After Stereotactic Body Radiation Therapy of Ultra-central Lung Tumors.

•Stereotactic body radiation therapy (SBRT) for ultra-central lung tumors is associated with high toxicity rates.•To evaluate differences in radiosensitivity within the proximal bronchial tree (PBT), the PBT was sub-segmented into seven anatomical sections.•A risk-adapted SBRT regimen of EQD2_10 = 54.4 Gy in 8 or 10 fractions results in excellent local control and low rates of severe toxicity.•Data from a recent meta-analysis, the NORDIC Hilus trial and dosimetric data from this study were combined to create a NTCP model.•A dose threshold of EQD2_3 = 100 Gy to the PBT or any of its subsegments is expected to result in low rates of severe bronchial toxicity.

The Future of MR-Guided Radiation Therapy.

Magnetic resonance image guided radiation therapy (MRIgRT) is a relatively new technology that has already shown outcomes benefits but that has not yet reached its clinical potential. The improved soft-tissue contrast provided with MR, coupled with the immediacy of image acquisition with respect to the treatment, enables expansion of on-table adaptive protocols, currently at a cost of increased treatment complexity, use of human resources, and longer treatment slot times, which translate to decreased throughput. Many approaches are being investigated to meet these challenges, including the development of artificial intelligence (AI) algorithms to accelerate and automate much of the workflow and improved technology that parallelizes workflow tasks, as well as improvements in image acquisition speed and quality. This article summarizes limitations of current available integrated MRIgRT systems and gives an outlook about scientific developments to further expand the use of MRIgRT.

Intra- and inter-fraction breath-hold variations and margins for radiotherapy of abdominal targets.

Radiotherapy in expiration breath-hold (EBH) has the potential to reduce treatment volumes of abdominal targets compared to an internal target volume concept in free-breathing. The reproducibility of EBH and required safety margins were investigated to quantify this volumetric benefit. Pre- and post-treatment diaphragm position difference and the positioning variability were determined on computed tomography. Systematic and random errors for EBH position reproducibility and positioning variability were calculated, resulting in margins of 7 to 12 mm depending on the prescription isodose and fractionation. A reduced volume was shown for EBH for lesions with superior-inferior breathing motion above 4 to 8 mm.

ASTRA: Atomic Surface Transformations for Radiotherapy Quality Assurance.

Treatment for glioblastoma, an aggressive brain tumour usually relies on radiotherapy. This involves planning how to achieve the desired radiation dose distribution, which is known as treatment planning. Treatment planning is impacted by human errors, inter-expert variability in segmenting (or outlining) the tumor target and organs-at-risk, and differences in segmentation protocols. Erroneous segmentations translate to erroneous dose distributions, and hence sub-optimal clinical outcomes. Reviewing segmentations is time-intensive, significantly reduces the efficiency of radiation oncology teams, and hence restricts timely radiotherapy interventions to limit tumor growth. Moreover, to date, radiation oncologists review and correct segmentations without information on how potential corrections might affect radiation dose distributions, leading to an ineffective and suboptimal segmentation correction workflow. In this paper, we introduce an automated deep-learning based method: atomic surface transformations for radiotherapy quality assurance (ASTRA), that predicts the potential impact of local segmentation variations on radiotherapy dose predictions, thereby serving as an effective dose-aware sensitivity map of segmentation variations. On a dataset of 100 glioblastoma patients, we show how the proposed approach enables assessment and visualization of areas of organs-at-risk being most susceptible to dose changes, providing clinicians with a dose-informed mechanism to review and correct segmentations for radiation therapy planning. These initial results suggest strong potential for employing such methods within a broader automated quality assurance system in the radiotherapy planning workflow. Code to reproduce this is available at https://github.com/amithjkamath/astraClinical Relevance: ASTRA shows promise in indicating what regions of the OARs are more likely to impact the distribution of radiation dose.

Dose escalation for stereotactic arrhythmia radioablation of recurrent ventricular tachyarrhythmia - a phase II clinical trial.

BACKGROUND: Stereotactic arrhythmia radioablation (STAR) is delivered with a planning target volume (PTV) prescription dose of 25 Gy, mostly to the surrounding 75-85% isodose line. This means that the average and maximum dose received by the target is less than 35 Gy, which is the minimum threshold required to create a homogenous transmural fibrosis. Similar to catheter ablation, the primary objective of STAR should be transmural fibrosis to prevent heterogenous intracardiac conduction velocities and the occurrence of sustained ventricular arrhythmias (sVA) caused by reentry. We hypothesize that the current dose prescription used in STAR is inadequate for the long-term prevention of sVA and that a significant increase in dose is necessary to induce transmural scar formation. OBJECTIVE: A single arm, multi-center, phase II, dose escalation prospective clinical trial employing the i3 + 3 design is being conducted to examine the safety of a radiation dose-escalation strategy aimed at inducing transmural scar formation. The ultimate objective of this trial is to decrease the likelihood of sVA recurrence in patients at risk. METHODS: Patients with ischemic or non-ischemic cardiomyopathy and recurrent sVA, with an ICD and history of ≥ 1 catheter ablation for sVA will be included. This is a prospective, multicenter, one-arm, dose-escalation trial utilizing the i3 + 3 design, a modified 3 + 3 specifically created to overcome limitations in traditional dose-finding studies. A total of 15 patients will be recruited. The trial aims to escalate the ITV dose from 27.0 Gy to an ITV prescription dose-equivalent level of maximum 35.1 Gy by keeping the PTV prescription dose constant at 25 Gy while increasing the dose to the target (i.e. the VT substrate without PTV margin) by step-wise reduction of the prescribing isodose line (85% down to 65%). The primary outcome of this trial is safety measured by registered radiation associated adverse events (AE) up to 90 days after study intervention including radiation associated serious adverse events graded as at least 4 or 5 according to CTCAE v5, radiation pneumonitis or pericarditis requiring hospitalization and decrease in LVEF ≥ 10% as assessed by echocardiography or cardiac MRI at 90 days after STAR. The sample size was determined assuming an acceptable primary outcome event rate of 20%. Secondary outcomes include sVA burden at 6 months after STAR, time to first sVA recurrence, reduction in appropriate ICD therapies, the need for escalation of antiarrhythmic drugs, non-radiation associated safety and patient reported outcome measures such as SF-36 and EQ5D. DISCUSSION: DEFT-STAR is an innovative prospective phase II trial that aims to evaluate the optimal radiation dose for STAR in patients with therapy-refractory sVA. The trial has obtained IRB approval and focuses on determining the safe and effective radiation dose to be employed in the STAR procedure. TRIAL REGISTRATION: NCT05594368.

Efficacy and safety analysis in metastatic cancer patients treated with multiple courses of repeat radiation therapy.

Background and purpose: Due to advances in oncology, a growing proportion of patients is treated with repetitive courses of radiotherapy. The aim of this study is to analyze whether radiotherapy maintains its safety and efficacy profile in patients treated with multiple repeat courses of irradiation. Material and methods: All patients treated between 2011 and 2019 at our institution were screened for a minimum of five repeat irradiation courses, to analyze treatment characteristics, survival, safety and efficacy. The type of re-irradiation was classified according to ESTRO-EORTC consensus guidelines. Results: A total of n = 112 patients receiving n = 660 radiotherapy courses were included in this retrospective cohort study. The most frequent primary tumors were lung cancer in 41.9 % (n = 47) and malignant melanoma in 8.9 % (n = 10). The most frequent re-irradiation types were repeat irradiation and Type 2 re-irradiation in 309 (46.8 %) and 113 (17.1 %) cases, respectively. Median survival after the first course of radiotherapy was 3.6 (0.3-13.4) years. Response to radiotherapy was observed in 548 (83.0 %) cases and CTCAE toxicity grade ≥ 3 was observed in 21 (3.2 %) cases. An increasing number of RT courses (HR: 1.30, p=<0.0001), Type 1 re-irradiation (HR 3.50, p = 0.008) and KPS ≤ 80 % (HR: 2.02, p = 0.002) were associated with significantly worse treatment responses. Toxicity rates remained stable with increasing numbers of RT courses. Conclusion: Multiple courses of repeat radiotherapy maintain a favorable therapeutic ratio of high response combined with reasonable safety profile.

Re-irradiation in clinical practice: Results of an international patterns of care survey within the framework of the ESTRO-EORTC E2-RADIatE platform.

BACKGROUND: Re-irradiation is an increasingly utilized treatment for recurrent, metastatic or new malignancies after previous radiotherapy. It is unclear how re-irradiation is applied in clinical practice. We aimed to investigate the patterns of care of re-irradiation internationally. MATERIAL/METHODS: A cross-sectional survey conducted between March and September 2022. The survey was structured into six sections, each corresponding to a specific anatomical region. Participants were instructed to complete the sections of their clinical expertise. A total of 15 multiple-choice questions were included in each section, addressing various aspects of the re-irradiation process. The online survey targeted radiation and clinical oncologists and was endorsed by the European Society for Radiotherapy and Oncology (ESTRO) and the European Organisation for Research and Treatment of Cancer (EORTC). RESULTS: 371 physicians from 55 countries across six continents participated. Participants had a median professional experience of 16 years, and the majority (60%) were affiliated with an academic hospital. The brain region was the most common site for re-irradiation (77%), followed by the pelvis (65%) and head and neck (63%). Prolonging local control was the most common goal (90-96% across anatomical regions). The most common minimum interval between previous radiotherapy and re-irradiation was 6-12 months (45-55%). Persistent grade 3 or greater radiation-induced toxicity (77-80%) was the leading contraindication. Variability in organs at risk dose constraints for re-irradiation was observed. Advanced imaging modalities and conformal radiotherapy techniques were predominantly used. A scarcity of institutional guidelines for re-irradiation was reported (16-19%). Participants from European centers more frequently applied thoracic and abdominal re-irradiation. Indications did not differ between academic and non-academic hospitals. CONCLUSION: This study highlights the heterogeneity in re-irradiation practices across anatomical regions and emphasizes the need for high-quality evidence from prospective studies to guide treatment decisions and derive safe cumulative dose constraints.

Deep-Learning-Based Dose Predictor for Glioblastoma-Assessing the Sensitivity and Robustness for Dose Awareness in Contouring.

External beam radiation therapy requires a sophisticated and laborious planning procedure. To improve the efficiency and quality of this procedure, machine-learning models that predict these dose distributions were introduced. The most recent dose prediction models are based on deep-learning architectures called 3D U-Nets that give good approximations of the dose in 3D almost instantly. Our purpose was to train such a 3D dose prediction model for glioblastoma VMAT treatment and test its robustness and sensitivity for the purpose of quality assurance of automatic contouring. From a cohort of 125 glioblastoma (GBM) patients, VMAT plans were created according to a clinical protocol. The initial model was trained on a cascaded 3D U-Net. A total of 60 cases were used for training, 15 for validation and 20 for testing. The prediction model was tested for sensitivity to dose changes when subject to realistic contour variations. Additionally, the model was tested for robustness by exposing it to a worst-case test set containing out-of-distribution cases. The initially trained prediction model had a dose score of 0.94 Gy and a mean DVH (dose volume histograms) score for all structures of 1.95 Gy. In terms of sensitivity, the model was able to predict the dose changes that occurred due to the contour variations with a mean error of 1.38 Gy. We obtained a 3D VMAT dose prediction model for GBM with limited data, providing good sensitivity to realistic contour variations. We tested and improved the model's robustness by targeted updates to the training set, making it a useful technique for introducing dose awareness in the contouring evaluation and quality assurance process.

Potential of ChatGPT in facilitating research in radiation oncology?

PURPOSE: To evaluate the potential of the artificial intelligence (AI) chatbot ChatGPT in supporting young clinical scientists with scientific tasks in radio oncological research. MATERIALS AND METHODS: Seven scientific tasks were to be completed in 3 h by 8 radiation oncologists with different scientific experience working at a university hospital: creation of a scientific synopsis, creation of a research question and corresponding clinical trial hypotheses, writing of the first paragraph of a manuscript introduction, clinical trial sample size calculation, and clinical data analyses (multivariate analysis, boxplot and survival curve). No participant had prior experience with an AI chatbot. All participants were instructed in ChatGPT v3.5 and its use was provided for all tasks. Answers were scored independently by two blinded experts. The subjective value of ChatGPT was rated by each participant. Data were analyzed with regression-, t-test and Spearman correlation (p < 0.05). RESULTS: Participants completed tasks 1-3 with an average score of 50% and 4-7 with 56%. Scientific experience, number of original publications and of first/last authorships showed a positive correlation with overall scoring (p = 0.01-0.04). Participants with little to moderate scientific experience scored ChatGPT to be more helpful in solving tasks 4-7 compared to more experienced participants (p = 0.04), with simultaneously presenting lower scorings (p = 0.03). CONCLUSIONS: ChatGPT did not compensate for differences in scientific experience of young clinical scientists, with less experienced researchers believing false AI-generated scientific results.

Identifying core MRI sequences for reliable automatic brain metastasis segmentation.

BACKGROUND: Many automatic approaches to brain tumor segmentation employ multiple magnetic resonance imaging (MRI) sequences. The goal of this project was to compare different combinations of input sequences to determine which MRI sequences are needed for effective automated brain metastasis (BM) segmentation. METHODS: We analyzed preoperative imaging (T1-weighted sequence ± contrast-enhancement (T1/T1-CE), T2-weighted sequence (T2), and T2 fluid-attenuated inversion recovery (T2-FLAIR) sequence) from 339 patients with BMs from seven centers. A baseline 3D U-Net with all four sequences and six U-Nets with plausible sequence combinations (T1-CE, T1, T2-FLAIR, T1-CE + T2-FLAIR, T1-CE + T1 + T2-FLAIR, T1-CE + T1) were trained on 239 patients from two centers and subsequently tested on an external cohort of 100 patients from five centers. RESULTS: The model based on T1-CE alone achieved the best segmentation performance for BM segmentation with a median Dice similarity coefficient (DSC) of 0.96. Models trained without T1-CE performed worse (T1-only: DSC = 0.70 and T2-FLAIR-only: DSC = 0.73). For edema segmentation, models that included both T1-CE and T2-FLAIR performed best (DSC = 0.93), while the remaining four models without simultaneous inclusion of these both sequences reached a median DSC of 0.81-0.89. CONCLUSIONS: A T1-CE-only protocol suffices for the segmentation of BMs. The combination of T1-CE and T2-FLAIR is important for edema segmentation. Missing either T1-CE or T2-FLAIR decreases performance. These findings may improve imaging routines by omitting unnecessary sequences, thus allowing for faster procedures in daily clinical practice while enabling optimal neural network-based target definitions.

Dosimetric analysis of 17 cardiac Sub-structures, Toxicity, and survival in ultra central lung tumor patients treated with SBRT.

•Data on cardiac toxicity after SBRT for ultra-central lung tumors remains limited.•We analyzed the dose to 18 cardiac sub-structures and cardiovascular toxicity.•A SBRT regimen of 45 Gy in 8-10 fractions yields good local control and low toxicity.•The highest cardiac doses were observed in the pulmonary artery and left atrium.•Higher doses to the base of the heart seem to be associated with non-cancer deaths.

External validation of three prognostic scores for brain metastasis velocity in patients treated with intracranial stereotactic radiotherapy.

BACKGROUND AND INTRODUCTION: Brain metastasis velocity (BMV) has been proposed as a prognostic factor for overall survival (OS) in patients with brain metastases (BMs). In this study, we conducted an external validation and comparative assessment of the performance of all three BMV scores. MATERIALS AND METHODS: Patients treated with intracranial stereotactic radiotherapy (SRT) for BM at a single center between 2014 and 2018 were identified. Where possible, all three BMV scores were calculated. Log-rank tests and linear, logistic and Cox regression analysis were used for validation and predictor identification of OS. RESULTS: For 333 of 384 brain metastasis patients, at least one BMV score could be calculated. In a sub-group of 187 patients, "classic" BMV was validated as categorical (p < 0.0001) and continuous variable (HR 1.02; 95% CI 1.02-1.03; p < 0.0001). In a sub-group of 284 patients, "initial" BMV was validated as categorical variable (high-risk vs. low-risk; p < 0.01), but not as continuous variable (HR 1.02; 95% CI 0.99-1.04; p = 0.224). "Volume-based" BMV could not be validated in a sub-group of 104 patients. On multivariable Cox regression analysis, iBMV (HR 1.85; 95% CI 1.01-3.38; p < 0.05) and cBMV (HR 2.32; 95% CI 1.15 4.68; p < 0.05) were predictors for OS for intermediate-risk patients after first SRT and first DBFs, respectively. cBMV proved to be the dominant predictor for OS for high-risk patients (HR 2.99; 95% CI 1.30-6.91; p < 0.05). CONCLUSION: This study externally validated cBMV and iBMV as prognostic scores for OS in patients treated with SRT for BMs whereas validation of vBMV was not achieved.

Recurrences of ventricular tachycardia after stereotactic arrhythmia radioablation arise outside the treated volume: analysis of the Swiss cohort.

AIMS: Stereotactic arrhythmia radioablation (STAR) has been recently introduced for the management of therapy-refractory ventricular tachycardia (VT). VT recurrences have been reported after STAR but the mechanisms remain largely unknown. We analysed recurrences in our patients after STAR. METHODS AND RESULTS: From 09.2017 to 01.2020, 20 patients (68 ± 8 y, LVEF 37 ± 15%) suffering from refractory VT were enrolled, 16/20 with a history of at least one electrical storm. Before STAR, an invasive electroanatomical mapping (Carto3) of the VT substrate was performed. A mean dose of 23 ± 2 Gy was delivered to the planning target volume (PTV). The median ablation volume was 26 mL (range 14-115) and involved the interventricular septum in 75% of patients. During the first 6 months after STAR, VT burden decreased by 92% (median value, from 108 to 10 VT/semester). After a median follow-up of 25 months, 12/20 (60%) developed a recurrence and underwent a redo ablation. VT recurrence was located in the proximity of the treated substrate in nine cases, remote from the PTV in three cases and involved a larger substrate over ≥3 LV segments in two cases. No recurrences occurred inside the PTV. Voltage measurements showed a significant decrease in both bipolar and unipolar signal amplitude after STAR. CONCLUSION: STAR is a new tool available for the treatment of VT, allowing for a significant reduction of VT burden. VT recurrences are common during follow-up, but no recurrences were observed inside the PTV. Local efficacy was supported by a significant decrease in both bipolar and unipolar signal amplitude.

Saliency-Enhanced Content-Based Image Retrieval for Diagnosis Support in Dermatology Consultation: Reader Study.

BACKGROUND: Previous research studies have demonstrated that medical content image retrieval can play an important role by assisting dermatologists in skin lesion diagnosis. However, current state-of-the-art approaches have not been adopted in routine consultation, partly due to the lack of interpretability limiting trust by clinical users. OBJECTIVE: This study developed a new image retrieval architecture for polarized or dermoscopic imaging guided by interpretable saliency maps. This approach provides better feature extraction, leading to better quantitative retrieval performance as well as providing interpretability for an eventual real-world implementation. METHODS: Content-based image retrieval (CBIR) algorithms rely on the comparison of image features embedded by convolutional neural network (CNN) against a labeled data set. Saliency maps are computer vision-interpretable methods that highlight the most relevant regions for the prediction made by a neural network. By introducing a fine-tuning stage that includes saliency maps to guide feature extraction, the accuracy of image retrieval is optimized. We refer to this approach as saliency-enhanced CBIR (SE-CBIR). A reader study was designed at the University Hospital Zurich Dermatology Clinic to evaluate SE-CBIR's retrieval accuracy as well as the impact of the participant's confidence on the diagnosis. RESULTS: SE-CBIR improved the retrieval accuracy by 7% (77% vs 84%) when doing single-lesion retrieval against traditional CBIR. The reader study showed an overall increase in classification accuracy of 22% (62% vs 84%) when the participant is provided with SE-CBIR retrieved images. In addition, the overall confidence in the lesion's diagnosis increased by 24%. Finally, the use of SE-CBIR as a support tool helped the participants reduce the number of nonmelanoma lesions previously diagnosed as melanoma (overdiagnosis) by 53%. CONCLUSIONS: SE-CBIR presents better retrieval accuracy compared to traditional CBIR CNN-based approaches. Furthermore, we have shown how these support tools can help dermatologists and residents improve diagnosis accuracy and confidence. Additionally, by introducing interpretable methods, we should expect increased acceptance and use of these tools in routine consultation.