Related Experiment Video
Updated: Nov 9, 2025
03:38
Unilateral Lung Volume Analysis Using Micro-CT for Enhanced Assessment of Pulmonary Fibrosis in Preclinical Models
Published on: June 20, 2025
534
Fully Automated Segmentation of Pulmonary Fibrosis Using Different Software Tools
Julia Ley-Zaporozhan1,2, Athanasios Giannakis3,4,5, Tobias Norajitra6
1Department Radiology, University Hospital, LMU Munich, Munich, Germany.
Respiration; International Review of Thoracic Diseases
|April 15, 2021
Summary
LUFIT software offers superior segmentation of pulmonary fibrosis in idiopathic pulmonary fibrosis (IPF) patients compared to YACTA. This shape model-based tool enhances CT
Area of Science:
- Radiology and Medical Imaging
- Pulmonary Medicine
- Computational Pathology
Background:
- Idiopathic pulmonary fibrosis (IPF) is a progressive lung disease characterized by fibrotic changes.
- Computed Tomography (CT) plays a crucial role in assessing disease extent and evolution in IPF.
- Accurate segmentation and quantification of fibrotic changes are essential for disease monitoring and treatment evaluation.
Purpose of the Study:
- To evaluate and compare the performance of three automated software tools (YACTA, LUFIT, IMBIO) for segmenting and quantifying fibrotic changes in IPF patients.
- To determine the most effective software for characterizing pulmonary fibrosis using CT imaging.
- To assess the potential of these tools in strengthening CT's role as a biomarker in IPF.
Main Methods:
- Analysis of 418 nonenhanced thin-section MDCTs from 127 IPF patients and 78 MDCTs from 78 healthy controls.
- Comparison of YACTA and LUFIT for lung volume segmentation and quantification of 80th (fibrosis) and 40th (ground-glass opacity) density percentiles using Bland-Altman plots.
- Inclusion of fibrosis and ground-glass opacity segmented by IMBIO in regression analyses.
Main Results:
- LUFIT demonstrated superior performance over YACTA in segmenting lung volume and quantifying higher 80th and 40th percentiles in IPF patients.
- No significant differences were observed between the tools in the control group.
- LUFIT's quantified 80th/40th percentiles showed a strong positive correlation with IMBIO's fibrosis/ground-glass opacity percentages (r = 0.78/0.92).
Conclusions:
- LUFIT, a shape model-based segmentation tool, is superior to the threshold-based YACTA for pulmonary fibrosis segmentation in IPF.
- Shape modeling, as implemented in LUFIT, is a valid approach for quantifying IPF, particularly given its subpleural involvement.
- These findings support the use of advanced software tools for robust CT-based biomarkers in IPF management.
