Related Experiment Videos
Automatic fetal head measurements from sonographic images
V Chalana1, T C Winter, D R Cyr
1Center for Bioengineering, University of Washington, Seattle 98195-2500, USA.
Academic Radiology
|August 1, 1996
Summary
An automated image processing technique accurately measures fetal biparietal diameter (BPD) and head circumference (HC) from prenatal ultrasounds, matching expert sonographer performance.
Area of Science:
- Medical Imaging
- Computer Vision
- Obstetrics
Background:
- Accurate fetal biometry is crucial for monitoring growth and development.
- Manual measurement of biparietal diameter (BPD) and head circumference (HC) can be subjective.
- Automated methods offer potential for increased objectivity and efficiency.
Purpose of the Study:
- To develop and evaluate an automated image processing technique for measuring fetal BPD and HC.
- To compare the accuracy of automated measurements against experienced sonographers.
Main Methods:
- Utilized an active contour model for detecting fetal skull boundaries in digitized prenatal sonograms.
- Automated computation of BPD and HC using the developed image processing algorithm.
- Comparison of automated measurements with those from six experienced sonographers.
Main Results:
- The algorithm achieved high accuracy, with mean absolute differences of 1.4% for BPD and 2.9% for HC compared to sonographers.
- Excellent correlations were observed: .999 for BPD and .994 for HC.
- Automated measurements showed no significant difference compared to inter-observer variability among sonographers.
Conclusions:
- The developed algorithm provides effective and accurate automatic measurement of fetal BPD and HC.
- The technique demonstrates potential for integration into clinical ultrasound machines for routine use.