Computer-Aided Pelvic Imaging for Female Health (CAPI)

Motivation: Women’s health and particularly the application of sophisticated imaging and analysis tools to the female pelvis is recently attracting fast growing interest.

 

Taking endometriosis as an example, often referred to as the chameleon disease, research and improvements in care are clearly required: “Endometriosis is one of the most common gynecological diseases, but it can take up to 10 years to get a diagnosis [..] Roughly 10 percent globally—of reproductive-age women and girls have endometriosis.” Diana W. Bianchi, M.D., former NICHD Director. Common indications are furthermore cervical cancer, endometrial cancers, ovarian cancer, and fibroids among others.

 

However, the female reproductive organs display an extraordinary variability, both between individuals (eg retroverted/anteverted uteri), within the same individual with different stages of the menstrual cycle (eg uterine lining, ovarian appearance) and even more dramatically over the female patient’s life span - from the appearance reflecting the stage of sexual maturation, to reproductive age, to pregnancy and postpartum changes, to postmenstrual women, where the endometrium and ovaries undergo progressive involution.

​

The influence of hormones and the requirement to assess any imaging with respect to the phase to differentiate between normal and pathological findings further complicate diagnosis and decision making. 

 

Furthermore, the pelvis is an area of extraordinary dynamic events, from the natural peristalsis to breathing motion, to the influence of surrounding tissues such as the bowel on the imaging process. Both diagnosis and therapies often include laparoscopic interventions, amplifying the challenges stated above and placing further emphasis on the need for individualized imaging and analysis techniques. This area is thus becoming increasingly relevant as well to the interventional part of MICCAI (“CAI”) as robotic surgeries are becoming more complex and new surgical developments lead to highly specialized tools and advanced methods for surgical planning. 

 

This area therefore poses excellent challenges, both on the image acquisition side as well as on the analysis side, calling for expertise in advanced methods in these areas, such as perfectly represented in the MICCAI community.

​

Two keynote speakers - one focusing more on clinical and one on methodological developments, will deliver keynote talks at the workshop to stimulate discussions, present recent research and highlight future challenges in this field.

Two oral sessions, composed of reviewed high quality papers in the field, characterized by their methodological rigour, clinical relevance and innovative character, will ensure a high quality overview, will stimulate discussion, and thus lead to new collaborations within the MICCAI community and beyond.

 

Call for abstracts: Potential submissions will cover the full scope of medical image analysis applied to women’s reproductive health, from early detection and diagnosis, to computer-aided diagnosis, to handling data, labels, annotations and segmentations, evaluation methods, clinical validity like performance metrics, multimodal approaches, for example fusing ultrasound + MRI or other sensors + MRI, AI-based interventions for laparoscopic surgery (endometriosis, cysts, fibroids), uterine artery embolization prediction (for fibroids) and post-surgery analysis methods, applied for example to histopathological uterine or ovarian tissue.

​

Keywords - Reproductive Health, Women’s Health, Imaging, Endometriosis, Adenomyosis, Cervical Cancer, Uterine Myomas/Fibroids, Pelvic Inflammatory Diseases, PCOs/PCODs