The Use of Machine Learning in MicroRNA Diagnostics: Current Perspectives

Area of Science:

• Biomedical Informatics
• Computational Biology
• Genomics

Background:

• MicroRNAs are key regulators of gene expression, implicated in numerous diseases.
• Artificial Intelligence (AI), particularly machine learning (ML), offers powerful tools for data analysis and prediction.
• Integrating microRNA data with ML models presents a novel approach for disease management.

Approach:

• This review comprehensively analyzes the application of microRNA-based ML models in healthcare.
• Studies focusing on cancer diagnostics, including gastric, pancreatic, colorectal, breast, lung, and ovarian cancers, were prioritized.
• Applications in non-cancerous diseases such as ulcerative colitis, cardiovascular diseases, and dementia were also examined.

Key Points:

• MicroRNA-based ML models demonstrate high diagnostic accuracy across a wide range of cancers (e.g., colorectal: 100%, lung: 100%) and other conditions (e.g., coronary heart disease: 96%, ulcerative colitis: 92.8%).
• These models show promise in enhancing diagnostic precision for diseases like dementia (82.9%) and gestational diabetes mellitus (86%).
• The integration of microRNA-based ML can potentially reduce the need for invasive diagnostic procedures.

Conclusions:

• MicroRNA-based ML models are effective tools for improving disease diagnosis accuracy.
• AI, powered by ML and microRNA data, is poised to significantly impact future clinical practices.
• Healthcare professionals should adapt to AI integration to remain current in medical practice.