Automated analysis of whole body FDG PET/CT Data using Machine Learning methods

Recruiting

• Conditions: C34; C81; C83; C43; Malignant neoplasm of bronchus and lung; Hodgkin lymphoma; Non-follicular lymphoma; Malignant melanoma of skin

• Registration Number: DRKS00026991
• Lead Sponsor: Klinikum der Universität München, Campus Großhadern

Brief Summary

Not available

Detailed Description

Not available

Study Timeline

• Study Start: 2022-03-10
• Last Update Posted: 19 August 2024

Recruitment & Eligibility

• Status: Recruiting
• Sex: All
• Target Recruitment: 2000

Inclusion Criteria

• Histologically confirmed malignancy (bronchial carcinoma, melanoma, lymphoma)
• Baseline (therapy-naïve) whole-body 18F -FDG PET/CT with at least one PET-positive lesion
• Control group: Patients without oncological disease who have received a whole-body 18F -FDG PET/CT due to another indication and do not show pathological tracer accumulation.

Exclusion Criteria

- Patients who have already received therapy prior to imaging
- Missing or incomplete image records in PACS

Study & Design

• Study Type: observational
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Automation of lesion detection and segmentation on whole-body PET/CT image data.

Secondary Outcome Measures

Name	Time	Method
Increase availability of anonymized clinical image datasets to researchers and support advancements in AI-based automated image analysis through public accessibility of datasets.