Data analysis of the Lung Imaging Database Consortium and Image Database Resource Initiative

Area of Science:

• Medical Imaging
• Radiology
• Data Science

Background:

• The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI) is a crucial resource for lung nodule research.
• Existing data formats and lack of uniform analysis hinder efficient use of this large computed tomography (CT) dataset.

Purpose of the Study:

• To develop a uniform data model for the LIDC/IDRI dataset.
• To create a software tool for processing and analyzing CT images and nodule characteristics.
• To facilitate in-depth understanding and efficient utilization of the LIDC/IDRI data for lung cancer investigations.

Main Methods:

• Designed a uniform data model to integrate diverse information from source files.
• Developed a software tool for automatic nodule outline alignment, characteristic extraction, and feature calculation (e.g., diameter, volume).
• Processed and analyzed all 1018 CT scans, summarizing nodule feature distributions.

Main Results:

• Integrated information into a new, uniform data model.
• Identified 2655 nodules ≥3 mm, 5875 nodules <3 mm, and 7411 non-nodules.
• Characterized nodules ≥3 mm: 85.7% <10.0 mm diameter (mean 6.72 mm), with malignancy scores ranging from 1 to 5.

Conclusions:

• The developed software tool enhances understanding and use of the LIDC/IDRI dataset for researchers.
• Analysis results reveal diverse nodule characteristics, serving as a reference for evaluating nodule detection and segmentation algorithms.