CAPI-WOMEN Workshop

Keynote by Prof. Lise Lecointre
Department of Gynecologic Surgery, University Hospitals of Strasbourg 

Augmenting Medicine: A Clinician-Scientist’s Vision of AI, Augmented Reality and Personalized Endometrial Cancer Care

​

​​

Program:

08:00 - 08:05    Opening remarks

08:05 - 09:35    Oral Session 1 (12 min + 3Q/A)

• Panda: Unsupervised Pelvic Anomaly Detection for Real-Time MR Imaging (ID_1)

• Comparing Problem Formulations for Endometriosis Lesion Detection on T2-weighted Pelvic MRI (ID_2)

• Interactive Foundation Models for 3D MR Uterus Segmentation: Performance and Limitations (ID_4)

• Pelvic MRI Patient Verification with Foundation-Model Embeddings (ID_5)

• PelviNeXt: A Modality-Agnostic Hybrid Network for Pelvic Imaging in Women's Health ((ID_7)

• Unsupervised Adversarial Domain Adaptation for Uterine layer Segmentation: From Labeled Cine to Unlabeled Dynamic EPI MRI (ID_8)

09:35 - 10:00    Poster teasers (3 min, No Q/A)

• Patient-Level Data Leakage in Colposcopy Classification: A Cautionary Study with Evidential Uncertainty Quantification for Cervical Cancer Screening ((ID_10)

• Multi-scale radiomics in pelvic MRI for endometriosis subtyping: highlighting data heterogeneity constraints (ID_3)

• A Multi-Site Label Trap in Shape-Based Endometriosis Detection: What Female Pelvic-Organ Shape Does (and Does Not) Encode ((ID_6)

• Cross-Modal MRI Ovary Segmentation in Endometriosis Using Unpaired TVUS Prototype Priors ((ID_9)

• An Integrated Ultrasound Pathway for Endometriosis Assessment (ID_11)

• Real-Time AI Assistance Improves Detection of Digestive Endometriosis on Transvaginal Ultrasound: A Within-Reader Study Across Experience Levels (ID_12)

• Task Definition and Multicenter Variability in Multimodal AI for Colposcopy (ID_13)

10:00 - 10:50    Coffee Break + Poster Session

10:50 - 11:20    Oral Session 2 (12 min + 3Q/A)

• Evaluating the Prompt-to-Automation Gap: A Comprehensive Benchmark for Laparoscopic Endometriosis Segmentation (ID_14)

• Pelvis-SigLIP: Benchmarking Vision-Language Models for Female Pelvic MRI Series Retrieval across Zero-Shot, Linear-Probe, and Fine-Tuning (ID_15)

11:20 - 12:20    Keynote: Augmenting Medicine: A Clinician-Scientist’s Vision of AI, Augmented Reality and Personalized Endometrial Cancer Care (Prof. Lise Lecointre, Department of Gynecologic Surgery, University Hospitals of Strasbourg)

12:20 - 12:30    Closing remarks

​​

​Posters:

1.    Panda: Unsupervised Pelvic Anomaly Detection for Real-Time MR Imaging
2.    Comparing Problem Formulations for Endometriosis Lesion Detection on T2-weighted Pelvic MRI
3.    Multi-scale radiomics in pelvic MRI for endometriosis subtyping: highlighting data heterogeneity constraints
4.    Interactive Foundation Models for 3D MR Uterus Segmentation: Performance and Limitations
5.    Pelvic MRI Patient Verification with Foundation-Model Embeddings
6.    A Multi-Site Label Trap in Shape-Based Endometriosis Detection: What Female Pelvic-Organ Shape Does (and Does Not) Encode
7.    PelviNeXt: A Modality-Agnostic Hybrid Network for Pelvic Imaging in Women's Health
8.    Unsupervised Adversarial Domain Adaptation for Uterine layer Segmentation: From Labeled Cine to Unlabeled Dynamic EPI MRI
9.    Cross-Modal MRI Ovary Segmentation in Endometriosis Using Unpaired TVUS Prototype Priors
10.    Patient-Level Data Leakage in Colposcopy Classification: A Cautionary Study with Evidential Uncertainty Quantification for Cervical Cancer Screening
11.    An Integrated Ultrasound Pathway for Endometriosis Assessment
12.    Real-Time AI Assistance Improves Detection of Digestive Endometriosis on Transvaginal Ultrasound: A Within-Reader Study Across Experience Levels
13.    Task Definition and Multicenter Variability in Multimodal AI for Colposcopy
14.    Evaluating the Prompt-to-Automation Gap: A Comprehensive Benchmark for Laparoscopic Endometriosis Segmentation
15.    Pelvis-SigLIP: Benchmarking Vision-Language Models for Female Pelvic MRI Series Retrieval across Zero-Shot, Linear-Probe, and Fine-Tuning