Related Experiment Video
Updated: May 15, 2026
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Automated intraventricular septum segmentation using non-local spatio-temporal priors.
Mithun Das Gupta1, Sheshadri Thiruvenkadam, Navneeth Subramanian
1John F. Welch Technology Center, GE Global Research, Bangalore, India. mithun.dasgupta@ge.com
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|January 5, 2013
Summary
This study presents an automated method for segmenting the intra-ventricular septum (IVS) in echocardiograms. The approach enhances cardiac disease quantification by overcoming challenges like noise and variable boundaries for improved diagnostic accuracy.
Area of Science:
- Medical Imaging
- Computer Vision
- Cardiology
Background:
- Automated segmentation of the intra-ventricular septum (IVS) is crucial for early cardiac disease quantification.
- Ultrasound image segmentation is difficult due to intensity variations, speckle noise, and non-rigid boundary changes.
Purpose of the Study:
- To develop a robust and automated method for segmenting the intra-ventricular septum (IVS) from B-mode echocardiographic images.
- To address the challenges of noise and boundary variations in ultrasound cardiac imaging.
Main Methods:
- A novel 1-D active contour segmentation approach incorporating non-local (NL) temporal cues.
- Robust initialization using NL-means de-noising and Markov Random Field (MRF) based clustering with physiological cues.
Main Results:
- The method was validated on approximately 30 cardiac scan videos (2000 frames).
- Achieved fully automatic and near real-time performance at 0.1 seconds per frame.
Conclusions:
- The proposed approach effectively segments the intra-ventricular septum (IVS), overcoming common ultrasound imaging challenges.
- This automated, near real-time method facilitates efficient and accurate cardiac disease quantification.
