Related Experiment Video
Updated: May 15, 2026
05:33
Three-Dimensional Shape Modeling and Analysis of Brain Structures
Published on: November 14, 2019
Pathology hinting as the combination of automatic segmentation with a statistical shape model
Pascal A Dufour1, Hannan Abdillahi, Lala Ceklic
1ARTORG Center for Biomedical Engineering Research, Ophthalmic Technologies, University of Bern, 3010 Bern, Switzerland. pascal.dufour@artorg.unibe.ch
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|January 5, 2013
Summary
This study introduces a novel method using segmentation and statistical shape models to quickly detect abnormalities in medical images. The system effectively visualizes pathologies like drusen in optical coherence tomography scans, aiding clinical practice.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Increasing volumes of medical image data pose challenges for clinicians.
- Accurate and rapid detection of pathologies is crucial for timely diagnosis and treatment.
Purpose of the Study:
- To develop a novel method for rapid visualization of medical image abnormalities.
- To apply this method for detecting drusen in optical coherence tomography (OCT) volumes for age-related macular degeneration (AMD) screening.
Main Methods:
- Combining segmentation algorithms with statistical shape models.
- Utilizing residual fitting error from a healthy statistical shape model to highlight deviations.
- Applying a segmentation technique for drusen and analyzing it with the statistical shape model.
Main Results:
- The developed system effectively visualizes potentially pathological areas.
- The method demonstrated high sensitivity in detecting drusen.
- All drusen with a height of at least 93.6 micrometers were detected, and most drusen of 85.5 micrometers height were identified.
Conclusions:
- The proposed method offers a fast and sensitive approach for identifying pathologies in medical images.
- This technique can aid clinicians in screening large datasets, particularly for conditions like AMD.
- Statistical shape models combined with segmentation provide a powerful tool for abnormality detection.
