Exploring Indian research trends in artificial intelligence for human health: An analysis of the WHO trial registry data

Abstract

INTRODUCTION Artificial intelligence (AI) has rapidly evolved into a pivotal technology in health care.[1] There is a global surge in AI healthcare research.[2] In India, the adoption of AI in health care is gaining momentum due to a favorable tech ecosystem, a large and diverse patient population, and public health needs.[3,4] However, AI may face several challenges in India, encompassing issues such as data quality and availability, ethical and sociocultural considerations, regulatory and policy frameworks, and resource allocation constraints.[5,6] A previous study by Rajiv and Khurshed explored trends of AI-related research registered in ClinicalTrials.gov up to September 2019 where one study was from India.[7] After that, AI-related research has grown exponentially worldwide as well as in India. Hence, we aimed to conduct this study to elucidate the current trends, focus areas, and regional distribution of AI research in healthcare in India. METHODS This is a cross-sectional public domain data audit. The study data were collected from the ICTRP search portal – https://trialsearch.who.int on July 5, 2024. The database was searched with the following keywords- ”artificial intelligence” OR “AI” OR “machine learning” OR “Deep learning” OR “large language model” OR “neural network.” From the dataset, registries from the Clinical Trials Registry-India (CTRI) were filtered and saved for further search and analysis. Each trial was searched individually in the CTRI database and the data were extracted. RESULTS A global trial search yielded a total of 4766 entries and among that, CTRI had 224 entries (4.7%). Among the 224, a total of 22 trials were excluded as it was not related to AI. Hence, a total of 202 trials were analyzed. There has been a gradual increase in the number of studies since 2018. A total of 143 (73.76%) were observational studies, and 59 (26.24%) were interventional studies. The majority was cross-sectional studies (36.14%), followed by cohort studies (10.89%) and retrospective studies (8.42%). Most studies (81.19%) were conducted in private institutions and sponsored by private companies (75.25%). Karnataka, Maharashtra, Delhi, and Tamil Nadu accounted for 71.29% of the total studies. Medical specialty-wise analysis showed that oncology is the mostly researched (31 [14.98%]) specialty, followed by pulmonology (22 [10.63%]), gynecology and obstetrics (15 [7.25%]), ophthalmology (15 [7.25%]), and dentistry (14 [6.76%]). DISSCUSSION This study explored the trends in AI research within the Indian healthcare sector. The predominance of observational studies is due to their effectiveness in identifying patterns, associations, and trends in large datasets, which is particularly relevant for AI applications that rely on vast amounts of data for training and validation.[8] There was a predominance of AI studies being conducted in private institutions as they often have more flexibility, resources, and a faster decision-making process compared to public institutions.[9] In addition, private companies have a vested interest in driving innovation to gain competitive advantages and open new market opportunities in the healthcare sector.[10] The predominance of oncology (most common was breast cancer and cervical cancer) as the most researched specialty in AI healthcare studies reflects the critical need for advancements in cancer diagnosis, treatment, and management. This finding is corroborative of the global research trends in AI.[7] This analysis provides insights into recent research trends in AI within the Indian healthcare landscape. The findings highlight the necessity for a balanced distribution of studies across different states and the need to strengthen research in government-run institutions with appropriate funding to further advance AI in health care. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.

Ähnliche Arbeiten