A Comprehensive Assessment of Clinical Studies on Artificial Intelligence-Enabled Interventions: Uncovering Trends and Potential Implications
Author(s)
Sugandh Sharma, MSc1, Suditi Gogna, MPH1, Sreejith Madathil, M. Pharm.2.
1Parexel International, Chandigarh, India, 2Parexel International, Bengaluru, India.
1Parexel International, Chandigarh, India, 2Parexel International, Bengaluru, India.
OBJECTIVES: Artificial intelligence (AI) and machine learning (ML) are revolutionizing the healthcare industry by enhancing drug discovery, diagnostics, prevention and disease management. To gain insight into upcoming applications of AI/ML in healthcare, an analysis of clinical trials was conducted.
METHODS: We searched ClinicalTrials.gov from inception to identify trials reporting AI/ML-enabled interventions.
RESULTS: 507 trials were identified, 272 met inclusion criteria, with most initiated from 2020 onwards. The majority were randomized controlled trials (82.7%), with 36.0% from Asia-Pacific, 25.3% from North America, and 23.5% from Europe. Among these, 146 (53.7%) were ongoing, 86 (31.6%) completed, and 40 (14.6%) terminated/unknown. Academic/government institutions funded 86% of ongoing/completed trials. It was also observed that overall, there has been a 6.6-fold increase in AI/ML usage in 2021-2025 as compared with 2015-2020. Oncology (62, 29%), cardiovascular (24, 11%), neuroscience including mental & neurological disorders (24, 11%), and diabetes (14, 7%) were the most common therapy areas. In oncology, 47% of trials focused on timely cancer detection, 16% on enhancing screening, and 8% on personalized radiotherapy, using AI tools like QuitBot® and Digi-Coach®. Cardiovascular trials primarily evaluated AI for electrocardiography screening, diagnosis, and personalized treatment, using tools like AI-LVEF® (guided assessment of cardiac function) and AI-SCREENDCM® (screening dilated cardiomyopathy). Neuroscience trials explored personalized treatment, prevention, and counseling, using AI/ML tools like SMART-AI® (automated reconstruction of intracranial vessel occlusion), while diabetes trials concentrated on patient management support.
CONCLUSIONS: Our analysis reveals a rapid increase in AI-focused clinical trials aimed at improving early detection, screening, and personalized treatment strategies. The prevalence of mobile apps and chatbots underscores a trend towards accessible, patient-centric AI solutions. As most trials are primarily academically funded, there is significant potential for future AI to be affordable and accessible for patients. However, further research is needed to evaluate the real-world applicability of these AI interventions across diverse healthcare settings.
METHODS: We searched ClinicalTrials.gov from inception to identify trials reporting AI/ML-enabled interventions.
RESULTS: 507 trials were identified, 272 met inclusion criteria, with most initiated from 2020 onwards. The majority were randomized controlled trials (82.7%), with 36.0% from Asia-Pacific, 25.3% from North America, and 23.5% from Europe. Among these, 146 (53.7%) were ongoing, 86 (31.6%) completed, and 40 (14.6%) terminated/unknown. Academic/government institutions funded 86% of ongoing/completed trials. It was also observed that overall, there has been a 6.6-fold increase in AI/ML usage in 2021-2025 as compared with 2015-2020. Oncology (62, 29%), cardiovascular (24, 11%), neuroscience including mental & neurological disorders (24, 11%), and diabetes (14, 7%) were the most common therapy areas. In oncology, 47% of trials focused on timely cancer detection, 16% on enhancing screening, and 8% on personalized radiotherapy, using AI tools like QuitBot® and Digi-Coach®. Cardiovascular trials primarily evaluated AI for electrocardiography screening, diagnosis, and personalized treatment, using tools like AI-LVEF® (guided assessment of cardiac function) and AI-SCREENDCM® (screening dilated cardiomyopathy). Neuroscience trials explored personalized treatment, prevention, and counseling, using AI/ML tools like SMART-AI® (automated reconstruction of intracranial vessel occlusion), while diabetes trials concentrated on patient management support.
CONCLUSIONS: Our analysis reveals a rapid increase in AI-focused clinical trials aimed at improving early detection, screening, and personalized treatment strategies. The prevalence of mobile apps and chatbots underscores a trend towards accessible, patient-centric AI solutions. As most trials are primarily academically funded, there is significant potential for future AI to be affordable and accessible for patients. However, further research is needed to evaluate the real-world applicability of these AI interventions across diverse healthcare settings.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
SA1
Topic
Methodological & Statistical Research, Study Approaches
Disease
Cardiovascular Disorders (including MI, Stroke, Circulatory), Mental Health (including addition), Neurological Disorders, Oncology