Related Experiment Video
Updated: Mar 8, 2026
07:13
Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
Published on: October 27, 2023
1.7K
Automatic Marker-free Longitudinal Infrared Image Registration by Shape Context Based Matching and Competitive
Chia-Yen Lee1, Hao-Jen Wang1,2, Jhih-Hao Lai1
1Department of Electrical Engineering, National United University, Taiwan.
Scientific Reports
|February 2, 2017
Summary
This study introduces an automatic algorithm for longitudinal infrared image registration, crucial for tracking breast cancer growth. The method accurately registers images, outperforming manual techniques for enhanced tumor detection.
Area of Science:
- Biomedical Imaging
- Medical Image Analysis
- Cancer Research
Background:
- Long-term infrared image comparison aids breast cancer assessment and early tumor detection.
- Longitudinal infrared image registration is essential but challenging due to difficulties with external markers and fiducial detection.
Purpose of the Study:
- To develop an automatic longitudinal infrared registration algorithm.
- To overcome limitations of manual marker-based registration in infrared imaging for breast cancer monitoring.
Main Methods:
- An automatic vascular intersection detection method was developed.
- Feature descriptors were established using shape context for robust matching.
- A competitive winner-guided mechanism was employed for optimal correspondence and control point extraction for the deformation model.
Main Results:
- The proposed algorithm achieved accurate and rapid image registration.
- The automated method demonstrated superior effectiveness compared to manual registration.
- A mean error of 0.91 pixels was achieved, indicating high precision.
Conclusions:
- The developed registration algorithm is accurate and reliable for longitudinal infrared image analysis.
- This provides a novel method for extracting more valuable data from infrared images for breast cancer assessment.
- The findings support improved early tumor detection and monitoring of breast cancer progression.
