Background: This study aims to analyze the geographical distribution of different AI types and applications, document implementation challenges, and assess outcomes of interest as well as potential opportunities for increasing healthcare efficiency.
Methodology: A systematic review analyzed 24 studies (2019-2024) from IEEE Xplore, PubMed, and Google Scholar using MeSH keywords, following specific inclusion and exclusion criteria.
Results: Results show that AI was applied to almost all spheres of life, with multi-modal AI, deep learning and machine learning models having promising applications in precision medicine, early diagnostics and integration of work processes. Common challenges included data shortages, bias in the algorithm, ethics and regulation, which indicated the need for appropriate guidelines and cross-disciplinary partnerships. Trends, however, included multi-modal data integration, increased automation and international convergence of standards. AI's benefits, advanced diagnostic accuracy, greater clinical predictability, and clinical processing efficiency are evidence of its ability to change the face of healthcare while removing significant barriers to its broader use.
Conclusion: AI can improve diagnostic processes in medicine by increasing their accuracy, improving their speed, and further adapting them to individual patients.