Radiomics integration into a picture archiving and communication system

Area of Science:

• Medical Imaging
• Radiology
• Computational Pathology

Background:

• Radiomics involves extracting quantitative imaging biomarkers for clinical outcome prognosis and tumor characterization.
• Integrating radiomics extraction with open-source Picture Archiving and Communication Systems (PACS) is crucial for clinical application.
• Existing workflows can be cumbersome, hindering the widespread adoption of radiomics.

Purpose of the Study:

• To develop and evaluate an integrated software solution (SQLite4Radiomics) for seamless radiomics extraction.
• To bridge the gap between radiomics analysis and clinical implementation by combining it with an open-source PACS.
• To assess the performance and efficiency of the integrated system across different imaging modalities and datasets.

Main Methods:

• Development of SQLite4Radiomics, an open-source software integrating radiomics extraction with PACS.
• Testing the software across three distinct imaging modalities.
• Benchmarking performance using lung cancer open datasets (RIDER and MMD) to measure extraction time per Region of Interest (ROI).

Main Results:

• The integrated SQLite4Radiomics software demonstrated effective radiomics extraction capabilities.
• Performance was validated across multiple imaging modalities.
• Median extraction time per ROI was found to be 10.7 seconds (percentiles 25-75: 8.9-18.7) in lung cancer datasets.

Conclusions:

• The developed SQLite4Radiomics offers an efficient and integrated solution for radiomics analysis.
• This integration facilitates the clinical application of radiomics by streamlining biomarker extraction within a PACS environment.
• The software shows promising performance for tumor characterization and outcome prediction using quantitative imaging biomarkers.