Use of Artificial Intelligence for Automated Recognition of Uncommon Pathogens in a Clinical Laboratory Setting: A Case Study

Intoduction: AI systems will transform clinical laboratories by improving diagnostic precision, shortening TAT, and identifying uncommon pathogens. This research investigates the diagnostic AI integration in a clinical laboratory at a Riyadh tertiary hospital.Materials and Methods: A 1000 set of samples containing various biological tissues such as blood, sputum, cerebrospinal fluid, and tissue biopsies were used for an observational prospective study. The integration of the AI algorithms together with the NGS data enabled the identification of rare pathogens. The diagnostic efficiency, TAT and rare pathogen identification were evaluated in both the conventional and AI_ Assisted work flows. Their opinions were qualitatively analyzed regarding the use of AI.Results: In comparison with the conventional methods, where the p- value was less than 0.001, the AI system had an overall sensitivity of 97.3% and specificity of 95.8%. AI cut the average search TAT by 40%, and discovered approximately 41.7 % of rare pathogens. In qualitative analysis, user satisfaction was high, but problems with data quality and the initial costs were mentioned.Summary: AI has the potential to change diagnostic processes in laboratories, which significantly increases diagnostic ability, TAT, and lowest detection rate of rare pathogens. There are unresolved questions concerning data quality and cost issues.