Integrating Artificial Intelligence (AI) and Machine Learning (ML) Techniques With Real-World Data (RWD) and Real-World Evidence (RWE) to Inform Precision Medicine: A Scoping Review
Author(s)
Jatinder Kumar, MPharm1, Divya Tamminina, MPharm, MBA2, Neema Joseph, MPH2, Rachel Gamburg, BSc3, Javed Shaikh, MPharm, MBA2, Coby Martin, MSc4, Alexandra Koumas, BSc5, Navneet Kumar, PhD1.
1RWE/HEOR/ES, Axtria India Pvt. Ltd., Gurugram, India, 2RWE/HEOR/ES, Axtria India Pvt. Ltd., Hyderabad, India, 3RWE/HEOR/ES, Axtria Inc., Waltham, MA, USA, 4RWE/HEOR/ES, Axtria Inc., Toronto, ON, Canada, 5RWE/HEOR/ES, Axtria Inc., Berkeley Heights, NJ, USA.
1RWE/HEOR/ES, Axtria India Pvt. Ltd., Gurugram, India, 2RWE/HEOR/ES, Axtria India Pvt. Ltd., Hyderabad, India, 3RWE/HEOR/ES, Axtria Inc., Waltham, MA, USA, 4RWE/HEOR/ES, Axtria Inc., Toronto, ON, Canada, 5RWE/HEOR/ES, Axtria Inc., Berkeley Heights, NJ, USA.
OBJECTIVES: The integration of artificial intelligence (AI) and machine learning (ML) in healthcare has revolutionized various aspects of medical practice, from diagnostics to personalized treatment. This review highlights the applications of these AI/ML technologies in personalized treatment to individual needs.
METHODS: A comprehensive search was performed in the PubMed and Embase. Keywords and MeSH terms related to Real-World Data (RWD) and Real-World Evidence (RWE), and precision medicine were used. We analysed the existing literature to identify key areas where AI/ML has been successfully applied to improve the healthcare delivery.
RESULTS: We identified a total of 963 articles through database searches. After removing 275 duplicates, we screened the remaining articles and included 36 studies for detailed analysis. AI/ML algorithms can effectively process vast amounts of RWD derived from diverse sources such as electronic health records, patient registries, and wearable devices to uncover patterns and insights. Currently, AI/ML technologies are facilitating the development of predictive models that support optimizing therapeutic interventions and clinical decision-making by enhancing the understanding of disease heterogeneity and identifying novel therapeutic targets. AI/ML frameworks, such as utilizing Markov processes, have shown to reduce costs by approximately 62% and improve patient outcomes by 30-35%. Additionally, Al/ML models have demonstrated an improvement in predicting individual treatment effects and outcomes, while the application in medical imaging has increased diagnostic accuracy considerably. However, challenges such as data quality and ethical considerations limits the integration of AI/ML and RWD/RWE. Addressing these challenges are crucial for the effective use of AI/ML techniques in precision medicine.
CONCLUSIONS: Overall, the integration of AI/ML with RWD/RWE has shown promising results in enhancing precision medicine, improving patient outcomes, and optimizing healthcare to reduce overall cost. Future research should focus on addressing the associated challenges and developing AI/ML tools to fully harness the potential to improve patient care.
METHODS: A comprehensive search was performed in the PubMed and Embase. Keywords and MeSH terms related to Real-World Data (RWD) and Real-World Evidence (RWE), and precision medicine were used. We analysed the existing literature to identify key areas where AI/ML has been successfully applied to improve the healthcare delivery.
RESULTS: We identified a total of 963 articles through database searches. After removing 275 duplicates, we screened the remaining articles and included 36 studies for detailed analysis. AI/ML algorithms can effectively process vast amounts of RWD derived from diverse sources such as electronic health records, patient registries, and wearable devices to uncover patterns and insights. Currently, AI/ML technologies are facilitating the development of predictive models that support optimizing therapeutic interventions and clinical decision-making by enhancing the understanding of disease heterogeneity and identifying novel therapeutic targets. AI/ML frameworks, such as utilizing Markov processes, have shown to reduce costs by approximately 62% and improve patient outcomes by 30-35%. Additionally, Al/ML models have demonstrated an improvement in predicting individual treatment effects and outcomes, while the application in medical imaging has increased diagnostic accuracy considerably. However, challenges such as data quality and ethical considerations limits the integration of AI/ML and RWD/RWE. Addressing these challenges are crucial for the effective use of AI/ML techniques in precision medicine.
CONCLUSIONS: Overall, the integration of AI/ML with RWD/RWE has shown promising results in enhancing precision medicine, improving patient outcomes, and optimizing healthcare to reduce overall cost. Future research should focus on addressing the associated challenges and developing AI/ML tools to fully harness the potential to improve patient care.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
MSR72
Topic
Methodological & Statistical Research
Topic Subcategory
Artificial Intelligence, Machine Learning, Predictive Analytics
Disease
STA: Personalized & Precision Medicine