Related Experiment Video
Updated: Sep 3, 2025
10:23
Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
3.0K
A geometric alignment for human temporal bone CT images via lateral semicircular canals segmentation
Xiaoguang Li1, Peng Fu1, Hongxia Yin2
1Faculty of Information Technology, Beijing University of Technology, Beijing, China.
Medical Physics
|July 29, 2022
Summary
This study introduces an automated method for aligning temporal bone CT images, significantly reducing processing time. The new technique ensures anatomical symmetry, improving efficiency for radiologists.
Area of Science:
- Medical Imaging
- Radiology
- Computer-Aided Diagnosis
Background:
- Temporal bone CT images often exhibit bilateral asymmetry due to variations in patient posture and scanner settings.
- Manual alignment of these images is time-consuming and a critical step for subsequent computer-aided analysis.
- Ensuring geometric alignment is vital for accurate multiplanar reconstruction and assessing bilateral anatomical symmetry.
Purpose of the Study:
- To develop a fully automatic algorithm for geometric alignment of human temporal bone CT images.
- To address the challenge of bilateral asymmetry in raw temporal bone CT scans.
- To improve the efficiency of image preprocessing for computer-aided temporal bone analysis.
Main Methods:
- Segmentation of lateral semicircular canals (LSCs) using a proposed multifeature fusion network.
- Definition of a standard 3D coordinate system for alignment.
- Implementation of an automated alignment procedure based on segmented LSCs.
Main Results:
- The LSC segmentation network demonstrated high accuracy.
- An acceptable alignment rate of 85% was achieved across 910 temporal bone CT sequences.
- The automated alignment process reduced the time from 10 minutes (manual) to 60 seconds.
Conclusions:
- An efficient automatic geometric alignment method for temporal bone CT images has been developed.
- The proposed method effectively addresses bilateral asymmetry in raw CT images.
- This technique offers significant time savings for radiologists and enhances computer-aided analysis.
