Related Experiment Video
Updated: Apr 1, 2026
07:41
Quantification of Levator Ani Hiatus Enlargement by Magnetic Resonance Imaging in Males and Females with Pelvic Organ Prolapse
Published on: April 17, 2019
10.5K
Semi-automatic outlining of levator hiatus
N Sindhwani1,2, D Barbosa3, M Alessandrini3
1Department of Development and Regeneration, Cluster Organ Systems, Biomedical Sciences, KU Leuven, and Obstetrics and Gynaecology, University Hospitals Leuven, Leuven, Belgium.
Summary
A new automated hiatus segmentation (AHS) tool accurately outlines the levator hiatus, significantly reducing interobserver variability and analysis time. This AI-powered method improves consistency and speed for medical imaging analysis.
Area of Science:
- Medical Imaging Analysis
- Pelvic Floor Imaging
- Anatomical Segmentation
Background:
- Levator hiatus measurement is crucial in pelvic floor diagnostics.
- High interobserver variability exists in manual hiatal outlining.
- Faster and more consistent methods are needed for clinical analysis.
Purpose of the Study:
- To develop a semi-automated outlining tool for the levator hiatus.
- To reduce interobserver variability in hiatal analysis.
- To accelerate the process of hiatal segmentation.
Main Methods:
- An automated hiatus segmentation (AHS) algorithm was developed using C-plane images.
- The algorithm employs template fitting on an edge map and B-spline active surfaces for refinement.
- Segmentation accuracy was evaluated against manual outlines using MAD, Hausdorff distance, Dice, and Jaccard coefficients.
Main Results:
- The AHS algorithm demonstrated strong agreement with manual outlines (median MAD of 2.10 mm).
- It significantly reduced interobserver differences and improved intraclass correlation coefficient (ICC=0.93).
- AHS analysis was nearly three times faster than manual outlining (7.07s vs 21.31s).
Conclusions:
- The developed AHS method provides a fast, robust, and reliable way to trace the levator hiatal outline.
- It requires minimal user input, enhancing efficiency and consistency.
- The tool effectively improves interrater agreement in hiatal segmentation.
