USING ARTIFICIAL INTELLIGENCE FOR BIOMARKER ANALYSIS IN CLINICAL DIAGNOSTICS

Introduction. Artificial intelligence (AI) technologies are becoming crucial in clinical diagnostics due to their ability to process and interpret large volumes of data. The implementation of AI for biomarker analysis opens new opportunities in personalized medicine, offering more accurate and individualized approaches to disease diagnosis and treatment. The relevance of this review stems from the need to systematize recent advances in AI application for biomarker analysis, which is critical for early diagnosis and prediction of chronic non-communicable diseases (NCDs). Material and methods. The analysis of peer-reviewed scientific publications and reports from leading research centers over the past five years was conducted. Studies on the application of AI algorithms for analyzing genomic, proteomic, and metabolomic biomarkers were reviewed, including machine learning methods and deep neural networks. Special attention was paid to the integration of multi-marker panels for improving the accuracy of diagnosis and prediction of cardiovascular, digestive, respiratory, endocrine system diseases, as well as oncological and neurodegenerative pathologies. Results. The application of AI has significantly increased the sensitivity and specificity of diagnostics, especially in complex cases requiring analysis of multiple disease parameters. The effectiveness of AI has been demonstrated in early diagnosis of lung, breast, and colorectal cancer, prediction of cardiovascular complications and NCDs progression, including diabetes mellitus and Alzheimer’s disease. AI’s significant contribution to the discovery of new biomarkers, optimization of personalized treatment, and improvement of therapeutic strategies has been noted. Conclusion. The use of AI in biomarker analysis has become a significant breakthrough in medical diagnostics, particularly in oncology, cardiology, and neurodegenerative diseases. The technology allows integration of data about various biomarkers and contributes to creating more accurate models for disease diagnosis and prediction. Further development is associated with technology advancement and overcoming ethical and regulatory barriers, which will expand AI capabilities in clinical practice.