Patient-Specific Modeling in Urogynecology: A Meshfree Approach

Mechanical deformation of tissues in the female pelvic floor is believed to be central to understanding a number of important aspects of women’s health, particularly pelvic floor dysfunction. A 2008 study of US women reported the prevalence of pelvic floor disorders in the 20 and 39 years range as 9.7% with the prevalence increasing with age until it reaches roughly 50% in the 80 and older age group [Nygaard, Barber, Burgio, and et al (2008)]. Clinical observation indicates a strong correlation between problems such as pelvic organ prolapse/urinary incontinence and vaginal childbirth. It is thought that childbirth parameters like fetal weight, duration of labor, and pelvic bony and soft tissue geometry can modulate the level of injury sustained during childbirth. However, it is difficult to study the impact of childbirth parameters non-destructively in living women. Therefore, realistic, efficient, computational modeling capabilities are necessary to study the mechanical response of the organs and muscles during childbirth under varying conditions, in order to develop and test hypotheses for childbirth related injury. Furthermore, manufacturers of embedded prosthetic devices, such as those used to treat prolapse, may benefit from the ability to predict the mechanical performance of their prostheses in situ, and this potential benefit highlights the need for a capability to rapidly develop analytical models of the pelvic floor. This paper discusses an algorithm to automatically generate an analysis-suitable geometry from medical images. The automated analysis capability is demonstrated in modeling vaginal contracture, as might occur in cases of women treated with radiation for cervical cancer.