Predictive Patient Stratification Using Artificial Intelligence and Machine Learning

Summary

Predictive patient stratification is rapidly advancing as, it offers the ability to proactively identify patients who will benefit from specific interventions before their condition deteriorates. In the field of biomedical research, various stratification methods have proven successful in aiding the treatment process. However, due to the inherent heterogeneity and complexity of medical data, integrating and implementing these methods in practical clinical settings presents significant challenges. In this chapter, we aim to address these challenges by first reviewing state‐of‐the‐art artificial intelligence‐based methods for patient stratification. Subsequently, we introduce an integrative approach for stratifying cancer and Alzheimer disease patients. This approach combines the robust principal component analysis‐based dimensionality reduction technique with multiple kernel learning, providing a comprehensive and effective solution.