Related Experiment Videos
Automated CT image evaluation of the lung: a morphology-based concept.
R A Blechschmidt1, R Werthschützky, U Lörcher
1University of Technology, Department of Electrical Engineering and Information Technology, Darmstadt, Germany. r.blechschmidt@emk.tu-darmstadt.de
IEEE Transactions on Medical Imaging
|June 14, 2001
Summary
A new algorithm, the bullae index (BI), reliably detects emphysema, even when coexisting with fibrosis. This method, combined with emphysema type classification, offers precise diagnosis and avoids interobserver variance in CT imaging.
Area of Science:
- Pulmonary imaging and diagnostics
- Medical image analysis
- Quantitative CT analysis
Background:
- Computed tomography (CT) is crucial for detecting emphysema in vivo.
- Current methods like pixel index (PI) may fail when emphysema coexists with fibrosis.
- PI averages low-attenuation pixels, potentially masking emphysema in mixed cases.
Purpose of the Study:
- To introduce a novel algorithm for thoracic CT image evaluation based on emphysema morphology.
- To extend the pixel index (PI) to differentiate and quantify bullae size.
- To develop a bullae index (BI) and emphysema type (ET) classification for improved emphysema diagnosis.
Main Methods:
- A new algorithm evaluates thoracic CT images based on pulmonary emphysema morphology.
- Bullae are classified into four size categories (classes 1-4).
- The bullae index (BI) is calculated from the area percentages of different bullae sizes; emphysema type (ET) is derived from the ratio of large to small bullae areas.
Main Results:
- The bullae index (BI) demonstrates reliability comparable to the pixel index (PI).
- BI successfully detects emphysema in cases where PI shows normal values due to coexisting fibrosis.
- The combination of BI and ET classification correlates well with expert radiological assessment.
Conclusions:
- The bullae index (BI) provides an objective and reliable measure for quantifying emphysematous destruction.
- BI helps avoid interobserver variability, making it valuable for patient follow-up.
- Emphysema type (ET) classification offers a unique and helpful approach for precise emphysema diagnosis.