MR-based three-dimensional modeling of the normal pelvic floor in women: Quantification of muscle mass

Julia R. Fielding, Huseyin Dumanli, Andreas G. Schreyer, Shigeo Okuda, David T. Gering, Kelly H. Zou, Ron Kikinis, Ferenc A. Jolesz

Research output: Contribution to journalArticlepeer-review

150 Citations (Scopus)


OBJECTIVE. Our objective was to use a combination of axial MR source images and three-dimensional (3D) models to describe the anatomy of the normal pelvic floor in young nulliparous women and to measure the volume of the levator ani. SUBJECTS AND METHODS. Ten healthy nulliparous female volunteers (average age, 27 years) underwent T2-weighted MR imaging of the pelvis. Three-dimensional color-coded models of the pelvic bones and organs and the three major components of the levator ani - puborectalis, iliococcygeus, and coccygeus - were created. Source images were used to measure muscle width and signal intensity and to identify ligamentous structures. Using 3D models, we measured the volume of the levator ani, the angle of the levator plate, the posterior urethrovesical angle, and the distance of the bladder neck from the symphysis pubis and the pubococcygeal line. RESULTS. In all volunteers, the signal intensity of the puborectalis exceeded that of the obturator externus. The average volume of the levator ani was 46.6 ml, the average width of the levator hiatus was 41.7 mm, and the average posterior urethrovesical angle was 143.5°. Vaginal shape in the volunteers followed no recognizable pattern. CONCLUSION. Muscle morphology, signal intensity, and volume is relatively uniform among healthy young women.

Original languageEnglish
Pages (from-to)657-660
Number of pages4
JournalAmerican Journal of Roentgenology
Issue number3
Publication statusPublished - 2000
Externally publishedYes

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'MR-based three-dimensional modeling of the normal pelvic floor in women: Quantification of muscle mass'. Together they form a unique fingerprint.

Cite this