Related Experiment Video
Updated: Mar 17, 2026
Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
Published on: May 19, 2023
Cloud-Based NoSQL Open Database of Pulmonary Nodules for Computer-Aided Lung Cancer Diagnosis and Reproducible
José Raniery Ferreira Junior1, Marcelo Costa Oliveira2, Paulo Mazzoncini de Azevedo-Marques3
1Lab of Telemedicine and Medical Informatics, University Hospital Prof. Alberto Antunes, Institute of Computing, Federal University of Alagoas, Av. Lourival Melo Mota, Cidade Universitária, 57072-900, Maceió, Alagoas, Brazil. jrfj@ic.ufal.br.
A new cloud-based database offers public pulmonary nodule data with 3D texture attributes to improve computer-aided lung cancer diagnosis and detection. This resource aids researchers by providing essential data for developing and testing diagnostic tools.
Area of Science:
- Medical Imaging
- Radiology
- Computer-Aided Diagnosis
Background:
- Lung cancer is a leading cause of cancer deaths globally, with pulmonary nodules as a primary indicator.
- Accurate detection and classification of pulmonary nodules are challenging due to specialist interpretation variability.
- Existing computer-aided diagnosis research is hindered by a lack of shared medical reference data.
Purpose of the Study:
- To address the data scarcity in computer-aided diagnosis research for lung cancer.
- To present a public, cloud-based database of pulmonary nodules with detailed characterization.
- To enhance the development and evaluation of computational methods for nodule detection and classification.
Main Methods:
- Utilized data from the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI).
- Extracted image descriptors using volumetric texture analysis.
- Developed a document-oriented NoSQL database schema deployed on a cloud infrastructure using MongoDB.
Main Results:
- The database contains 379 exams, 838 nodules, and 8237 images (4029 CT scans, 4208 segmented nodules).
- Nodules are characterized by 3D texture attributes and classified by experienced radiologists into nine subjective characteristics.
- The database is accessible via a cloud Database as a Service framework.
Conclusions:
- The developed database provides a valuable, publicly accessible resource for advancing lung cancer diagnosis research.
- Facilitates the improvement of computer-aided detection and classification algorithms for pulmonary nodules.
- Aims to reduce errors in image interpretation and support radiologists in clinical practice.
