Automated segmentation in pelvic radiotherapy: A comprehensive evaluation of ATLAS-, machine learning-, and deep learning-based models

Physica Medica, Год журнала: 2024, Номер 125, С. 104486 - 104486

Опубликована: Авг. 3, 2024

Язык: Английский

---

Safety and efficiency of a fully automatic workflow for auto-segmentation in radiotherapy using three commercially available deep learning-based applications

Physics and Imaging in Radiation Oncology, Год журнала: 2024, Номер 31, С. 100627 - 100627

Опубликована: Июль 1, 2024

Advancements in radiotherapy auto-segmentation necessitate reliable and efficient workflows. Therefore, a standardized fully automatic workflow was developed for three commercially available deep learning-based applications compared to manual safety efficiency. The underwent evaluation with failure mode effects analysis. Notably, eight modes were reduced, including seven severity factors ≥7, indicating the effect on patients, two Risk Priority Number value >125, which assesses relative risk level. Efficiency, measured by mouse clicks, showed zero clicks workflow. This automation illustrated improvement both efficiency of

Язык: Английский

---

A fully automated machine-learning-based workflow for radiation treatment planning in prostate cancer

Clinical and Translational Radiation Oncology, Год журнала: 2025, Номер 52, С. 100933 - 100933

Опубликована: Фев. 11, 2025

Highlights•Clinically feasible RT plans by one-click ML-based workflow.•ML-based within investigator-dependant variability.•High potential to increase efficency and accuracy.AbstractIntroductionThe integration of artificial intelligence into radiotherapy planning for prostate cancer has demonstrated promise in enhancing efficiency consistency. In this study, we assess the clinical feasibility a fully automated machine learning (ML)-based "one-click" workflow that combines segmentation treatment planning. The proposed was designed create clinically acceptable plan inter-observer variation conventional plans.MethodsWe evaluated fully-automated on five low-risk patients treated with external beam compared results optimized inverse planned based contours six different experienced radiation oncologists. Both qualitative quantitative metrics were analyzed. Additionally, dose distribution segmentations (manual vs. manual automation).ResultsThe automatic deep-learning target volume revealed close agreement between expert referring Dice Similarity- Hausdorff index. However, CTVs had significantly smaller than (47.1 cm3 62.6 cm3). provide coverage range variability observed plans. Due CTV PTV (expert contours) lower plans.ConclusionOur study indicates tested is leads comparable This represents promising step towards efficient standardized treatment. Nevertheless, cohort, auto associated volumes contours, highlighting necessity improving models prospective testing automation therapy.

Язык: Английский

---

Performance of Commercial Deep Learning-Based Auto-Segmentation Software for Prostate Cancer Radiation Therapy Planning: A Systematic Review

Information, Год журнала: 2025, Номер 16(3), С. 215 - 215

Опубликована: Март 11, 2025

As yet, there is no systematic review focusing on benefits and issues of commercial deep learning-based auto-segmentation (DLAS) software for prostate cancer (PCa) radiation therapy (RT) planning despite that NRG Oncology has underscored such necessity. This article’s purpose to systematically DLAS product performances PCa RT their associated evaluation methodology. A literature search was performed with the use electronic databases 7 November 2024. Thirty-two articles were included as per selection criteria. They evaluated 12 products (Carina Medical LLC INTContour (Lexington, KY, USA), Elekta AB ADMIRE (Stockholm, Sweden), Limbus AI Inc. Contour (Regina, SK, Canada), Manteia Technologies Co. AccuContour (Jian Sheng, China), MIM Software ProtégéAI (Cleveland, OH, Mirada Ltd. DLCExpert (Oxford, UK), MVision.ai Contour+ (Helsinki, Finland), Radformation AutoContour (New York, NY, RaySearch Laboratories RayStation Siemens Healthineers AG AI-Rad Companion Organs RT, syngo.via Image Suite DirectORGANS (Erlangen, Germany), Therapanacea Annotate (Paris, France), Varian Systems, Ethos (Palo Alto, CA, USA)). Their results illustrate can delineate organs at risk (abdominopelvic cavity, anal canal, bladder, body, cauda equina, left (L) right (R) femurs, L R pelvis, proximal sacrum) four clinical target volumes (prostate, lymph nodes, bed, seminal vesicle bed) clinically acceptable outcomes, resulting in delineation time reduction, 5.7–81.1%. Although recommended each centre perform its own prior implementation, seems more important due methodological respective single studies, e.g., small dataset used, etc.

Язык: Английский

---

Development of Artificial Intelligence-Based Real-Time Automatic Fusion of Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasonography of the Prostate

Urology, Год журнала: 2025, Номер unknown

Опубликована: Март 1, 2025

ObjectivesTo report the development of artificial intelligence (AI)-based software to allow for autonomous fusion transrectal ultrasound and multiparametric magnetic resonance images prostate be used during transperineal biopsies.Materials MethodsThis study was approved by Institutional Review Board (protocol ID3167CESC). The automatic biopsy involved three steps: 1) Developing an AI component segment ultrasound; 2) anatomical structures in using labeled datasets from Cancer Imaging Archive in-house scans; 3) register segmented a three-step process: pre-alignment, rigid alignment, elastic fusion, validated measuring lesion distance between modalities. Statistical analysis included descriptive statistics Mann-Whitney U test, evaluating outcomes with Dice scores average precision metrics.ResultsThe showed score 0.87 test set 75,357 28,946 annotations. achieved 0.85 on 2,494 It also demonstrated mean Average Precision 0.80 bounding boxes 0.88 objects, both measured at 50% intersection over union threshold. reduced median resonance-ultrasound 8 mm (IQR 6–9) after 4 3–5) (p<0.001).ConclusionA data processing pipeline were created ideally biopsies.

Язык: Английский

---

Adaptive und automatisierte Strahlentherapie

Deleted Journal, Год журнала: 2025, Номер 31(5), С. 477 - 482

Опубликована: Март 5, 2025

Etwa die Hälfte aller Tumorpatienten erhält im Laufe der Erkrankung eine Strahlentherapie, entweder als alleinige, kurative Therapie, Rahmen multimodaler Konzepte neo(adjuvante) Therapie oder zur Palliation. Technologische Weiterentwicklungen haben dazu beigetragen, dass gewünschte Strahlendosis immer genauer an Zielstruktur angeschmiegt werden und damit Belastung gesunder Gewebe minimiert kann. Eine Limitation war jedoch bis vor einigen Jahren, Bestrahlungsplanung für gesamte Behandlung üblicherweise auf einer prätherapeutischen Planungs-Computertomographie beruhte anatomische Veränderungen Bereich des Zielvolumens unter meist nicht detektiert konnten. Diese Unsicherheit limitierte in manchen Körperregionen applizierbaren Bestrahlungsdosen. Mithilfe adaptiven Strahlentherapie können nun durch am Linearbeschleuniger integrierte, optimierte Bildgebung solche Veränderungen, wie z. B. Tumorschrumpfung Lageveränderung besonders strahlenempfindlicher Organe, etwa von Darmschlingen, unmittelbar jeder Bestrahlungsfraktion erkannt Bestrahlungsplan passend Anatomie Tages angepasst werden. Dies ist bei abdominellen Tumoren, dem Pankreaskarzinom Lebertumoren, aber auch zentral gelegenen Tumoren Lunge relevant ermöglicht Applikation höherer Der Beitrag gibt einen Überblick über Indikationen adaptive jeweiligen Vorteile, Herausforderungen dazu, künstliche Intelligenz Automatisierung hier Therapieoptimierung beitragen.

---

Medical machine learning operations: a framework to facilitate clinical AI development and deployment in radiology

European Radiology, Год журнала: 2025, Номер unknown

Опубликована: Май 8, 2025

Abstract The integration of machine-learning technologies into radiology practice has the potential to significantly enhance diagnostic workflows and patient care. However, successful deployment maintenance medical (MedML) systems in requires robust operational frameworks. Medical operations (MedMLOps) offer a structured approach ensuring persistent MedML reliability, safety, clinical relevance. are increasingly employed analyse sensitive radiological data, which continuously changes due advancements data acquisition model development. These can alleviate workload radiologists by streamlining tasks, such as image interpretation triage. MedMLOps ensures that stay accurate dependable facilitating continuous performance monitoring, systematic validation, simplified maintenance—all critical maintaining trust machine-learning-driven diagnostics. Furthermore, aligns with established principles protection regulatory compliance, including recent developments European Union, emphasising transparency, documentation, safe retraining. This enables implement modern tools control oversight at forefront, reliable within dynamic context practice. empowers deliver consistent, high-quality care confidence, aligned evolving standards needs. assist multiple stakeholders models available, monitored easy use maintain while preserving privacy. better serve patients implementation cutting-edge clinicians only utilised when they performing expected. Key Points Question applications becoming adopted clinics, but necessary infrastructure sustain these is currently not well-defined . Findings Adapting machine learning concepts enhances ecosystems improving interoperability, automating monitoring/validation, reducing burdens on informaticians Clinical relevance Implementing solutions eases faster safer adoption advanced models, consistent for clinicians, benefiting through streamlined

Язык: Английский

---

Real world AI-driven segmentation: Efficiency gains and workflow challenges in radiotherapy

Radiotherapy and Oncology, Год журнала: 2025, Номер unknown, С. 110977 - 110977

Опубликована: Июнь 1, 2025

Язык: Английский

---

Transforming Prostate Cancer Care: Innovations in Diagnosis, Treatment, and Future Directions

International Journal of Molecular Sciences, Год журнала: 2025, Номер 26(11), С. 5386 - 5386

Опубликована: Июнь 4, 2025

Prostate cancer remains a major global health challenge, ranking as the second most common malignancy in men worldwide. Advances diagnostic and therapeutic strategies have transformed its management, enhancing patient outcomes quality of life. This review highlights recent breakthroughs imaging, including multiparametric MRI PSMA-PET, which improved detection staging. Biomarker-based diagnostics, such PHI 4K Score, offer precise risk stratification, reducing unnecessary biopsies. Innovations treatment, robotic-assisted surgery, novel hormone therapies, immunotherapy, PARP inhibitors, are redefining care for localized advanced prostate cancer. Artificial intelligence (AI) machine learning (ML) emerging powerful tools to optimize prediction, treatment personalization. Additionally, advances radiation therapy, IMRT SBRT, provide targeted effective options high-risk patients. While these innovations significantly survival minimized overtreatment, challenges remain optimizing therapy sequencing addressing disparities care. The integration AI, theranostics, gene-editing technologies holds immense promise future management.

Язык: Английский

---

Clinical adoption of deep learning target auto-segmentation for radiation therapy: challenges, clinical risks, and mitigation strategies

Deleted Journal, Год журнала: 2024, Номер 1(1)

Опубликована: Янв. 1, 2024

Abstract Radiation therapy is a localized cancer treatment that relies on precise delineation of the target to be treated and healthy tissues guarantee optimal effect. This step, known as contouring or segmentation, involves identifying both volumes organs at risk imaging modalities like CT, PET, MRI guide radiation delivery. Manual however, time-consuming highly subjective, despite presence guidelines. In recent years, automated segmentation methods, particularly deep learning models, have shown promise in addressing this task. However, challenges persist their clinical use, including need for robust quality assurance (QA) processes risks associated with use models. review examines considerations adoption auto-segmentation radiotherapy, focused volume. We discuss potential (eg, over- under-segmentation, automation bias, appropriate trust), mitigation strategies human oversight, uncertainty quantification, education professionals), we highlight importance expanding QA include geometric, dose-volume, outcome-based performance monitoring. While offers significant benefits, careful attention rigorous measures are essential its successful integration practice.

Язык: Английский

---