Automatic extraction of comprehensive cardiac CT angiography parameters: a novel program with high accuracy and...

Area of Science:

• Cardiovascular Imaging
• Radiology Informatics

Background:

• Clinical reports for cardiac computed tomography angiography (CTA) contain valuable quantitative data.
• Manual extraction of these parameters is time-consuming and prone to error.
• Standardization of CTA reporting is crucial for data analysis and clinical decision-making.

Purpose of the Study:

• To evaluate the accuracy and efficiency of a novel automated program for extracting cardiac CTA parameters.
• To assess the program's performance on a large cohort of clinical CTA reports.
• To determine the generalizability of the program based on reporting guidelines.

Main Methods:

• Retrospective collection and processing of 500 cardiac CTA reports.
• Pre-population of reports with a structured template adhering to established guidelines.
• Application of a novel automated program to extract 15 predefined cardiac CTA parameters.

Main Results:

• The automated program demonstrated high accuracy, extracting 15 parameters with 97.3% precision.
• The extraction process was highly efficient, completed in an average of 84 seconds per report.
• The program's performance was consistent across the analyzed dataset.

Conclusions:

• The developed automated program offers a highly accurate and efficient method for cardiac CTA parameter extraction.
• Adherence to Society of Cardiovascular Computed Tomography (SCCT) reporting guidelines is key for the program's successful implementation at other institutions.
• This tool has the potential to streamline data collection and improve the utility of cardiac CTA reports in clinical practice and research.