Related Experiment Video
Updated: Feb 19, 2026
09:36
Outer-Boundary Assisted Segmentation and Quantification of Trabecular Bones by an Imagej Plugin
Published on: March 14, 2018
9.8K
CT image segmentation methods for bone used in medical additive manufacturing.
Maureen van Eijnatten1, Roelof van Dijk1, Johannes Dobbe2
1Department of Oral and Maxillofacial Surgery, 3D Innovation Lab, VU University Medical Center, P.O. Box 7057, 1007 MB Amsterdam, The Netherlands.
Medical Engineering & Physics
|November 4, 2017
Summary
Image segmentation accuracy is crucial for medical additive manufacturing. While thresholding is common, advanced methods are needed for better accuracy and reduced costs in patient-specific constructs.
Area of Science:
- Medical Additive Manufacturing
- Medical Image Analysis
- Biomedical Engineering
Background:
- Accuracy in additive manufactured medical constructs is significantly impacted by image segmentation errors.
- Developing precise patient-specific implants and surgical guides relies heavily on accurate 3D models derived from medical imaging.
Purpose of the Study:
- To conduct a comprehensive literature review of image segmentation methods applied in medical additive manufacturing.
- To evaluate the accuracy and limitations of various segmentation techniques for bone structures.
Main Methods:
- A systematic literature search was performed across PubMed, ScienceDirect, Scopus, and Google Scholar.
- Thirty-two publications focusing on bone segmentation accuracy using computed tomography (CT) images were identified and analyzed.
- The advantages, disadvantages, and reported accuracies of different segmentation methods were critically assessed.
Main Results:
- Reported accuracies for bone segmentation varied widely, ranging from 0.04 mm to 1.9 mm.
- Global thresholding, the most frequent method, achieved accuracies below 0.6 mm but required extensive manual post-processing.
- Advanced thresholding methods demonstrated potential for improved accuracy (under 0.38 mm) but are not yet integrated into commercial software.
- Statistical shape models offered accuracies between 0.25 mm and 1.9 mm, suitable for structures with moderate variations.
Conclusions:
- Thresholding remains the predominant segmentation technique in medical additive manufacturing despite its limitations.
- Enhancing the accuracy and cost-effectiveness of patient-specific additive manufactured constructs necessitates the adoption of more sophisticated segmentation methodologies.
- Further research and integration of advanced segmentation algorithms into clinical workflows are essential for improving medical additive manufacturing outcomes.
