Related Experiment Video
Updated: Jul 6, 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
Segmenting lung fields in serial chest radiographs using both population-based and patient-specific shape statistics.
1Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai 200030, China. shi-yh@cs.sjtu.edu.cn
IEEE Transactions on Medical Imaging
|April 9, 2008
Summary
This study introduces a novel deformable model for accurate lung field segmentation in serial chest radiographs. It enhances segmentation by combining population and patient-specific shape statistics for improved robustness.
Area of Science:
- Medical Imaging
- Computer Vision
- Radiology
Background:
- Accurate segmentation of lung fields in serial chest radiographs is crucial for monitoring respiratory conditions.
- Existing deformable models often struggle with robustness and accuracy in longitudinal studies.
Purpose of the Study:
- To develop and validate a novel deformable model for precise lung field segmentation in serial chest radiographs.
- To improve the accuracy and robustness of lung field segmentation using combined statistical shape constraints.
Main Methods:
- A modified Scale Invariant Feature Transform (SIFT) local descriptor was employed for enhanced image feature characterization.
- A deformable contour model was constrained by both population-based and patient-specific shape statistics.
- Patient-specific statistics were updated online to refine segmentation across serial images.
Main Results:
- The proposed model demonstrated superior robustness and accuracy compared to other active shape models.
- The integration of patient-specific statistics improved segmentation refinement over time.
- Effective segmentation of lung fields was achieved even for initial time-point images.
Conclusions:
- The novel deformable model offers a significant advancement in segmenting lung fields from serial chest radiographs.
- The combined approach of population and patient-specific shape statistics enhances segmentation performance.
- This method provides a more reliable tool for quantitative analysis of lung changes in longitudinal imaging studies.
