Related Experiment Video
Updated: Mar 16, 2026
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.9K
Improving Spleen Volume Estimation Via Computer-assisted Segmentation on Clinically Acquired CT Scans
Zhoubing Xu1, Adam L Gertz2, Ryan P Burke3
1Electrical Engineering, Vanderbilt University EECS, 2301 Vanderbilt Pl., PO Box 351679 Station B, Nashville, TN 37235-1679.
Academic Radiology
|August 14, 2016
Summary
Multi-atlas segmentation accurately estimates spleen volumes from CT scans. An automated approach with outlier correction offers the best balance of accuracy and speed.
Area of Science:
- Medical Imaging
- Radiology
- Computer-Aided Diagnosis
Background:
- Multi-atlas fusion is a key technique for computer-assisted segmentation of anatomical structures.
- Accurate spleen volume estimation is crucial for clinical assessments.
Purpose of the Study:
- To evaluate the accuracy and time efficiency of multi-atlas segmentation for spleen volume estimation.
- To compare different segmentation pipelines using computed tomography (CT) scans.
Main Methods:
- Five pipelines were compared: manual segmentation, fully automated segmentation, automated with manual outlier correction, and two methods using spleen measurements.
- Accuracy was assessed using Dice similarity coefficient, Pearson correlation, R-squared, and volume deviation against manual segmentation as ground truth.
- Time cost for each pipeline was recorded.
Main Results:
- The automated segmentation with manual outlier correction (Pipeline 3) demonstrated superior accuracy (Pearson correlation 0.99, absolute volume deviation 23.7 cm³) and efficiency (1 minute/scan).
- A pipeline using 3D splenic index measurements (Pipeline 5) was the second-best (Pearson correlation 0.98, deviation 46.92 cm³, 1.5 minutes/scan).
- Manual segmentation (Pipeline 1) required significantly more time (11 minutes/scan).
Conclusions:
- Computer-automated segmentation with manual outlier correction provides accurate spleen volume estimations.
- This approach offers a practical and time-efficient solution for clinical applications.
