Correlate: A PACS- and EHR-integrated Tool Leveraging Natural Language Processing to Provide Automated Clinical...

Area of Science:

• Radiology and Medical Informatics
• Artificial Intelligence in Healthcare

Background:

• Radiologists face challenges in obtaining timely clinical follow-up information.
• Current methods are time-intensive and cover only a small fraction of cases.

Purpose of the Study:

• To develop and evaluate an automated system for retrieving clinical follow-up data.
• To improve workflow efficiency and clinical outcome awareness for radiologists.

Main Methods:

• Development of Correlate, a program integrated with Picture Archiving and Communication System (PACS) and Electronic Health Records (EHR).
• Utilization of Natural Language Processing (NLP) for free-text query processing and data parsing from EHR.
• Implementation of relevance-based result ordering and email notifications.

Main Results:

• Correlate enables efficient, minimally disruptive access to clinical follow-up information.
• The system provides a customizable database of queries and results.
• NLP facilitates intelligent prospective searches within the EHR.

Conclusions:

• Correlate enhances radiologists' ability to obtain essential clinical follow-up data.
• The tool improves patient care, interdisciplinary collaboration, and radiologist learning.
• Automated clinical follow-up tools can significantly benefit radiology practice.