Examining the Current Use of Artificial Intelligence in Clinical Research
While there is still potential left to be unlocked, current uses of artificial intelligence in clinical research include drug discovery, determining patient disposition, and dosing.
A commentary recently published in the Singapore Medical Journal examined opportunities for the application of artificial intelligence (AI) in healthcare. The authors reviewed use cases that were presented at the inaugural International Conference on AI in Medicine.1
“The quadruple aims of healthcare are widely recognized as better patient experience, better outcomes, lower costs, and team well-being. Recently, a fifth aim of healthcare has evolved, that of improving health equity, to emphasize the importance of social factors as determinants of health,” the authors wrote. “This is in line with HealthierSG, Singapore’s effort to pivot towards population health. AI has the potential to address the quintuple aims of healthcare by scaling the provision of care through democratization of professional expertise.”1
The authors first highlighted the use of AI in radiology, which they see as the area “best primed” for AI model development. Picture archiving and communication systems that collect data are stored in a standardized Digital Imaging and Communications in Medicine (DICOM) format, providing opportunities for computer vision. According to the authors, of the current 521 FDA-approved Software as a Medical Device AI solutions, more than half (393 in total) are radiology applications, emphasizing the area’s potential for AI intervention.1
“The incremental value from radiology AI solutions for radiologists is probably marginal when compared to its value to non-radiologists, such as clinicians working in primary care, emergency departments or intensive care units,” the authors wrote.1
While AI in radiology has great potential, the authors contend that human intervention is still necessary.1
Additionally, computer vision has successfully been developed for gastrointestinal endoscopy. In particular, AI has been used successfully for polyp detection through AI-assisted colonoscopy. The authors also noted that anatomical landmark identification, monitoring of speed of scope withdrawal, and objective assessment of bowel preparation ensure adequate visualization for quality assurance.1
Looking to the evolving uses of AI in clinical research, the authors first highlighted accelerated protein structure-based drug discovery, which includes utilizing medical record databases. They also highlighted the potential of outcome predictions through determining patient disposition at the emergency department and dynamic modulation of drug intervention allowing for personalized dosing regimens. Finally, the authors noted that recent advances in generative AI and large language models (LLMs) have made them intriguing options in clinical research.1
“These LLMs could accelerate the retrieval of key information from vast amounts of unstructured text data in electronic medical records, increasing clinician productivity,” the authors wrote. “With ambient AI, speech conversion of consultation dialogue to clinical notes facilitates automated information organization. This alleviates administrative demands on clinicians, allowing them to spend quality time with patients.”1
While there is great potential for AI in clinical research, the authors maintain that quality checks of potential AI models must be carefully considered. To accelerate adoption, efficacy of the models, regulatory standards, best practices for safety, and access to accurate data are all things that need to be considered.1
“In summary, AI has vast potential to change the practice of healthcare. This is most imminent in the areas of diagnosis and outcome prediction, which would aid clinical decision-making. More work needs to be done to educate practitioners and patients,” the authors concluded. “The most pressing challenge that we face today is in guiding our local clinical community to appropriately assess, adapt and adopt AI solutions to optimize patient care.”1
1. Tan, Cher Heng FRCR, MBA1,2; Goh, Wilson Wen Bin MSc, PhD2,3,4; So, Jimmy Bok Yan FRCS, MPH5,6,7; Sung, Joseph J Y MD, PhD2,8. Clinical use cases in artificial intelligence: current trends and future opportunities. Singapore Medical Journal 65(3):p 183-185, March 2024. | DOI: 10.4103/singaporemedj.SMJ-2023-193 https://journals.lww.com/smj/fulltext/2024/03000/clinical_use_cases_in_artificial_intelligence_.10.aspx
Improving Relationships and Diversifying the Site Selection Process
April 17th 2025In this episode of the Applied Clinical Trials Podcast, Liz Beatty, co-founder and chief strategy officer, Inato, discusses a number of topics around site engagement including community-based sites, the role of technology in improving site/sponsor relationships, how increased operational costs are impacting the industry, and more.
Behind the Buzz: Why Clinical Research Leaders Flock to SCOPE Summit
February 7th 2025In this episode, we meet with Micah Lieberman, Executive Conference Director for SCOPE Summit (Summit for Clinical Ops Executives) at Cambridge Innovation Institute. We will dive deep into the critical role of collaboration within the clinical research ecosystem. How do we bring together diverse stakeholders—sponsors, CROs, clinical trial tech innovators, suppliers, patients, sites, advocacy organizations, investors, and non-profits—to share best practices in trial design, program planning, innovation, and clinical operations? We’ll explore why it’s vital for thought leaders to step beyond their own organizations and learn from others, exchanging ideas that drive advancements in clinical research. Additionally, we’ll discuss the pivotal role of scientific conferences like SCOPE Summit in fostering these essential connections and collaborations, helping shape the future of clinical trials. Join us as we uncover how collective wisdom and cross-industry partnerships are transforming the landscape of clinical research.
