Diese Seite wurde automatisch übersetzt und die Genauigkeit der Übersetzung wird nicht garantiert. Bitte wende dich an die englische Version für einen Quelltext.

Multimodal LLM for Breast Cancer Radiotherapy Target Auto-Contouring and Planning

11. November 2026 aktualisiert von: Xiaohui Zhao

Multimodal Data Fusion-Based Artificial Intelligence Large Language Model for Automated Target Volume Contouring and Intelligent Treatment Planning in Breast Cancer Radiotherapy: A Multicenter Ambidirectional Cohort Study

Breast cancer remains the most prevalent malignancy among women worldwide, with a continuously increasing incidence. As a crucial component of comprehensive cancer treatment, radiotherapy plays a key role in reducing local recurrence rates and improving overall survival outcomes. However, traditional radiotherapy target volume contouring and treatment planning design rely heavily on clinician expertise, presenting challenges such as prolonged processing time, significant inter-observer variability, and high subjectivity in dose optimization. Although artificial intelligence has made substantial progress in automated contouring and planning, conventional single-modal models still face major limitations. These models struggle to effectively integrate clinical text, tumor biological characteristics, and dose distribution factors, resulting in restricted generalizability. Therefore, developing an intelligent radiotherapy assistance system capable of multimodal data fusion is of high clinical significance to enhance precision and efficiency in breast cancer radiotherapy.

This study aims to develop an innovative breast cancer radiotherapy assistance system based on a multimodal large language model. By integrating multimodal imaging data (including CT, MRI, and ultrasound), pathology reports, and structured electronic health record text, the system realizes end-to-end intelligence from target volume contouring to treatment plan generation. Employing a multicenter, ambidirectional cohort design, the study collects comprehensive clinical data from breast cancer patients who have undergone breast-conserving surgery or mastectomy. Technologically, the system innovatively combines the natural language processing capabilities of large language models with the image analysis strengths of deep learning, utilizing attention mechanisms to achieve deep fusion and interactive comprehension of heterogeneous medical data. Notably, the system breaks through traditional radiotherapy planning workflows by jointly optimizing target contouring and dose planning as a unified pipeline, applying reinforcement learning algorithms to dynamically balance target volume coverage with organ-at-risk protection. The outcomes of this study will provide an important paradigm for the clinical application of AI in radiation oncology and advance the field of precision radiotherapy.

Studienübersicht

Studientyp

Beobachtungs

Einschreibung (Geschätzt)

51000

Kontakte und Standorte

Dieser Abschnitt enthält die Kontaktdaten derjenigen, die die Studie durchführen, und Informationen darüber, wo diese Studie durchgeführt wird.

Studienkontakt

Studienorte

    • Guangdong
      • Guangzhou, Guangdong, China, 510120
        • Sun Yat-sen Memorial Hospital, Sun Yat-sen University
        • Kontakt:
        • Hauptermittler:
          • Xiaohui Zhao, MD

Teilnahmekriterien

Forscher suchen nach Personen, die einer bestimmten Beschreibung entsprechen, die als Auswahlkriterien bezeichnet werden. Einige Beispiele für diese Kriterien sind der allgemeine Gesundheitszustand einer Person oder frühere Behandlungen.

Zulassungskriterien

Studienberechtigtes Alter

  • Kind
  • Erwachsene
  • Älterer Erwachsener

Akzeptiert gesunde Freiwillige

Nein

Studienberechtigte Geschlechter

Alle

Probenahmeverfahren

Nicht-Wahrscheinlichkeitsprobe

Studienpopulation

The study population includes patients with breast cancer from multiple medical centers. The retrospective cohort comprises patients treated at Sun Yat-sen Memorial Hospital (n=35,000) and external collaborating centers (Shenshan Medical Center, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Zhoukou Central Hospital, and Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, n=15,000) between January 1, 2002, and October 31, 2025. The prospective test cohort comprises patients prospectively enrolled at Sun Yat-sen Memorial Hospital starting from January 1, 2026 (n=1,000).

Beschreibung

Inclusion Criteria:

  • Patients with pathologically confirmed breast malignant tumors who require postoperative adjuvant radiotherapy following radical surgery.
  • Availability of complete pathology reports.
  • Availability of complete imaging examinations and reports (such as CT, MRI, ultrasound, X-ray).
  • Availability of standardized clinical medical records.

Exclusion Criteria:

  • Patients with missing data.
  • Patients with primary multiple cancers.
  • Patients receiving salvage or palliative radiotherapy.
  • Patients receiving accelerated partial breast irradiation (APBI).

Studienplan

Dieser Abschnitt enthält Einzelheiten zum Studienplan, einschließlich des Studiendesigns und der Messung der Studieninhalte.

Wie ist die Studie aufgebaut?

Designdetails

  • Beobachtungsmodelle: COHORT
  • Zeitperspektiven: OTHER

Kohorten und Interventionen

Gruppe / Kohorte
Intervention / Behandlung
Retrospective Training and Internal Validation Cohort

Retrospective collection of clinical data, multimodal images (MRI, CT, ultrasound, X-ray), pathology reports, and radiotherapy plans from patients with breast cancer treated at Sun Yat-sen Memorial Hospital between January 1, 2002, and October 31, 2025. This cohort (n=35,000) is used for model training and internal validation of the multimodal LLM-assisted automated contouring and radiotherapy planning system.

The study evaluates an innovative multimodal large language model (LLM) system designed for automated target volume delineation and intelligent treatment planning in post-operative breast cancer radiotherapy. The system integrates clinical text, multimodal imaging (CT, MRI, ultrasound, X-ray), and pathology data to joint-optimize contouring and dose planning without direct patient intervention.

Retrospective External Validation Cohort

Retrospective collection of data from external collaborative centers (Sun Yat-sen Memorial Hospital Shenzhen Hospital, Zhoukou Central Hospital, and Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology) between January 1, 2002, and October 31, 2025 (total n=15,000). This cohort is used to evaluate the external generalizability and robustness of the multimodal AI model.

The study evaluates an innovative multimodal large language model (LLM) system designed for automated target volume delineation and intelligent treatment planning in post-operative breast cancer radiotherapy. The system integrates clinical text, multimodal imaging (CT, MRI, ultrasound, X-ray), and pathology data to joint-optimize contouring and dose planning without direct patient intervention.

Prospective Test Cohort

Prospective collection of data from patients with breast cancer treated at Sun Yat-sen Memorial Hospital starting from January 1, 2026 (n=1,000). This cohort undergoes double-blind evaluation to prospectively test the performance of the multimodal LLM-driven target volume auto-segmentation and intelligent radiotherapy planning system in clinical practice.

The study evaluates an innovative multimodal large language model (LLM) system designed for automated target volume delineation and intelligent treatment planning in post-operative breast cancer radiotherapy. The system integrates clinical text, multimodal imaging (CT, MRI, ultrasound, X-ray), and pathology data to joint-optimize contouring and dose planning without direct patient intervention.

Was misst die Studie?

Primäre Ergebnismessungen

Ergebnis Maßnahme
Maßnahmenbeschreibung
Zeitfenster
Dice Similarity Coefficient (DSC) of Target Volume Auto-Segmentation
Zeitfenster: Baseline

Evaluation of the Dice Similarity Coefficient (DSC) between AI-generated automatic breast cancer target volume contours and expert manual contours.

Baseline

Sekundäre Ergebnismessungen

Ergebnis Maßnahme
Maßnahmenbeschreibung
Zeitfenster
95% Hausdorff Distance (95HD) of Target Volume Auto-Segmentation
Zeitfenster: Baseline

Evaluation of the 95% Hausdorff Distance (95HD) between AI-generated automatic breast cancer target volume contours and expert manual contours.

Baseline
Intersection over Union (IoU) of Target Volume Auto-Segmentation
Zeitfenster: Baseline

Evaluation of the Intersection over Union (IoU) between AI-generated automatic breast cancer target volume contours and expert manual contours.

Baseline
Clinical Evaluation Score of Target Volume Auto-Segmentation
Zeitfenster: Baseline

Blinded evaluation by radiation oncology experts using the Investigator-Developed Cancer Target Volume Contouring Performance Evaluation Scale, a structured 5-point clinical evaluation scale. The scale assesses five dimensions: laterality accuracy (0 to 1 point), radiotherapy type suitability (0 to 1 point), target volume definition (0 to 1.5 points), coverage (0 to 1 point), and structural integrity (0 to 0.5 points). The total score ranges from a minimum of 0 to a maximum of 5, where higher scores indicate better target volume contouring quality and clinical acceptability.

Baseline

Mitarbeiter und Ermittler

Hier finden Sie Personen und Organisationen, die an dieser Studie beteiligt sind.

Sponsor

Ermittler

  • Hauptermittler: Xiaohui Zhao, MD, Sun Yat-Sen Memorial Hospital of Sun Yat-Sen University

Studienaufzeichnungsdaten

Diese Daten verfolgen den Fortschritt der Übermittlung von Studienaufzeichnungen und zusammenfassenden Ergebnissen an ClinicalTrials.gov. Studienaufzeichnungen und gemeldete Ergebnisse werden von der National Library of Medicine (NLM) überprüft, um sicherzustellen, dass sie bestimmten Qualitätskontrollstandards entsprechen, bevor sie auf der öffentlichen Website veröffentlicht werden.

Haupttermine studieren

Studienbeginn (Geschätzt)

9. November 2026

Primärer Abschluss (Geschätzt)

1. Dezember 2027

Studienabschluss (Geschätzt)

31. Dezember 2027

Studienanmeldedaten

Zuerst eingereicht

2. November 2026

Zuerst eingereicht, das die QC-Kriterien erfüllt hat

11. November 2026

Zuerst gepostet (Tatsächlich)

17. November 2026

Studienaufzeichnungsaktualisierungen

Letztes Update gepostet (Tatsächlich)

17. November 2026

Letztes eingereichtes Update, das die QC-Kriterien erfüllt

11. November 2026

Zuletzt verifiziert

1. November 2026

Mehr Informationen

Begriffe im Zusammenhang mit dieser Studie

Andere Studien-ID-Nummern

  • SYSKY-2026-014-01

Plan für individuelle Teilnehmerdaten (IPD)

Planen Sie, individuelle Teilnehmerdaten (IPD) zu teilen?

NEIN

Arzneimittel- und Geräteinformationen, Studienunterlagen

Studiert ein von der US-amerikanischen FDA reguliertes Arzneimittelprodukt

Nein

Studiert ein von der US-amerikanischen FDA reguliertes Geräteprodukt

Nein

Diese Informationen wurden ohne Änderungen direkt von der Website clinicaltrials.gov abgerufen. Wenn Sie Ihre Studiendaten ändern, entfernen oder aktualisieren möchten, wenden Sie sich bitte an register@clinicaltrials.gov. Sobald eine Änderung auf clinicaltrials.gov implementiert wird, wird diese automatisch auch auf unserer Website aktualisiert .

Abonnieren