Artificial intelligence algorithms in the development of novel tools for diagnostic and therapeutic interventions in occupational burn-out syndrome in healthcare professionals — pilot study and validation of the CUPIO application

Introduction: Occupational burn-out is an increasingly common phenomenon. While all occupations are susceptible to stress and burn-out, healthcare practitioners are among the particularly vulnerable. The widescale development of technological competencies and the progress of artificial intelligence algorithms allow for the creation of accessible, available, and personalized digital resources for prevention, diagnosis, and intervention. Material and methods : Original data from the pilot study and validation of the CUPIO AI application, implementing a fuzzy logic artificial intelligence algorithm, are presented. The study assessed the level of occupational burn-out using a standardized questionnaire before and after a 30-day intervention period with the CUPIO AI application. Results : Before the intervention, the questionnaire-measured level of occupational burn-out was high, with average scores of: total burn-out (M = 5.55, SD = 0.37), depersonalization (M = 5.60, SD = 0.58), emotional exhaustion (M = 5.92, SD = 0.48), and disengagement (M = 5.09, SD = 0.72). After 30 days of using the application, statistically significant changes were observed in three of the studied indicators. A decrease in the values was noted for total burn-out (M = 4.33, SD = 0.35; t = 7.05, p < 0.001), depersonalization (M = 4.64, SD = 0.59; t = 7.95, p < 0.001), and emotional exhaustion (M = 4.39, SD = 0.59; t = 4.87, p < 0.001). Discussion : The results suggest that AI-based digital tools, particularly those utilizing fuzzy logic algorithms, may play a role in reducing occupational burn-out symptoms among healthcare professionals. The observed improvements in key indicators indicate the potential of personalized digital interventions to support psychological well-being and resilience in demanding professional environments. Conclusions : Further research is necessary to assess the utilization of tools using similar algorithms in burn-out prevention, diagnosis, and treatment.