Detection and Correction of Laterality Errors in Radiology Reports.

Area of Science:

• Medical Imaging
• Radiology Informatics
• Software Engineering

Background:

• Laterality errors in radiology reports can lead to misdiagnosis and patient harm.
• Manual review processes are time-consuming and prone to human error.
• There is a need for automated solutions to ensure accuracy in radiological reporting.

Purpose of the Study:

• To develop supervising software for double-checking laterality errors in radiology reports.
• To evaluate the usefulness of detection and correction software in radiology report systems.
• To assess the accuracy and reliability of the developed software.

Main Methods:

• An AutoHotkey macro program was designed for detecting laterality discrepancies.
• Software detects discrepancies between examination names and report context, offering pop-up correction.
• Accuracy was evaluated using 300 radiologic examinations with known laterality variations.

Main Results:

• The software achieved a detection accuracy of 99.67% (95% CI; 99.01-%) in test examinations.
• Laterality errors were identified in 0.048% of a large database of previous radiology reports.
• The AutoHotkey-scripted macro functioned reliably as additional software in the reading workstation.

Conclusions:

• Developed detection and correction software effectively addresses laterality errors in radiology reports.
• The method is adaptable to existing hospital software, enhancing the radiologic reading environment.
• This automated approach is expected to improve overall diagnostic accuracy and patient safety.