Novel Artificial Intelligence Tool for Real-time Patient Identification to Prevent Misidentification in Health Care

Area of Science:

• Medical Informatics
• Computer Vision
• Artificial Intelligence

Background:

• Patient misidentification in healthcare settings can lead to critical errors in treatment delivery, including incorrect dosages and procedures.
• Ensuring accurate patient identification is paramount for safe and effective medical interventions.

Purpose of the Study:

• To develop and implement a Python deep learning-based program for real-time patient identification.
• To reduce errors associated with incorrect patient identification in healthcare facilities.

Main Methods:

• The program was developed using Python (version 3.9.12) and the OpenCV library for face recognition.
• The system focuses on face detection within the field of view for patient identification.
• The development process involved image data collection, data transfer, and data analysis.

Main Results:

• The developed patient identification program achieved a precision of 0.92, recall of 0.80, and specificity of 0.90.
• The overall accuracy of the program was determined to be 0.84.
• The system outputs "Unknown" for unidentified individuals in restricted areas, enhancing security.

Conclusions:

• The Python-based program facilitates accurate, real-time patient identification, minimizing manual intervention.
• This tool helps prevent medical complications arising from patient misidentification before therapies, medication administration, and other procedures.