Using AI to classify and predict outcomes of gynecological smooth muscle tumors
Deep Learning for Histopathological Classification and Prognostication of Gynaecologic Smooth Muscle Tumours
This study is testing whether artificial intelligence can help doctors better classify and predict outcomes for uncertain tumors in the uterus called STUMP.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 392 (estimated) |
| Sex | Female |
| Sponsor | Institut Bergonié Academic / other |
| Locations | 1 site (Bordeaux) |
| Trial ID | NCT06540846 on ClinicalTrials.gov |
What this trial studies
This study focuses on smooth muscle tumors of the uterus that are classified as STUMP, which are tumors of uncertain malignant potential. It aims to develop deep learning models utilizing convolutional neural networks to analyze histopathological features from digital images of tumor samples. By training these models on small pixel groups from hematoxylin and eosin (H&E) slides, the study seeks to improve the accuracy of subclassifying these tumors and predicting their prognosis. The ultimate goal is to create a diagnostic and prognostic algorithm that can assist pathologists in their evaluations.
Who should consider this trial
Good fit: Ideal candidates include patients diagnosed with uterine smooth muscle tumors, including leiomyomas, STUMP, and leiomyosarcomas, with available histopathological material.
Not a fit: Patients without a diagnosis of uterine smooth muscle tumors or those lacking available histopathological samples may not benefit from this study.
Why it matters
Potential benefit: If successful, this approach could enhance diagnostic accuracy and prognostic predictions for patients with gynecological smooth muscle tumors.
How similar studies have performed: Other studies utilizing AI for histopathological classification have shown promise, indicating potential success for this novel approach.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Patients with a diagnosis of uterine smooth muscle tumors (leiomyomas, smooth muscle tumors of uncertain malignancy and leiomyosarcomas), registered in the RRePS database and/or treated at Institut Bergonié or one of the participating centers. * Histopathological material available (kerosene blocks and/or slides). * The follow-up (outcome) is required for each LMS/ STUMP. Exclusion Criteria: * na
Where this trial is running
Bordeaux
- Institut Bergonie — Bordeaux, France (Recruiting)
Study contacts
- Study coordinator: Sabrina CROCE
- Email: s.croce@bordeaux.unicancer.fr
- Phone: +33556333333
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.