Establishment of an intelligent auto-verification system for infectious serological markers in a clinical laboratory from West China

This study aimed to evaluate an auto-verification system for serological markers of infectious diseases in a clinical laboratory, with the goal of ensuring reliable results while shortening the turnaround time for reports, reducing the need for human resources, and improving work efficiency. Based on the guidelines and clinical requirements, personalized auto-verification rules were established. Then, the serological marker results of 144,927 samples (including hepatitis B, hepatitis C, Treponema pallidum, and human immunodeficiency virus) were used for verification, and the performance of the auto-verification system was evaluated. The auto-verification system enabled our laboratory to shorten the turnaround time by 21.13% (from 369.54 minutes to 291.47 minutes) while increasing the sample volume. The pass rate, sensitivity, specificity, false positive rate, and false negative rate of the auto-verification system were 54.81%, 100%, 54.99%, 45.01% and 0%, respectively. Further analysis of the reasons for the failure of auto-verification revealed that a portion of the samples were blocked due to technical issues with the laboratory information system. Further improvements are needed in the future. Establishing an appropriate auto-verification system based on the actual situation of the laboratory is an important means to ensure medical safety, while improving the efficiency of the clinical laboratory and reducing labor costs.