Fasten Your Seatbelts: Machine Learning Is Revolutionizing Clinical Trials

Imagine standing on a foggy runway in the 1920s, using little more than flags and flashing lights, maybe an occasional bonfire, to direct incoming airplanes. In these early days, aviation was a manual, low-tech operation, relying on human sight, ground markers, and rudimentary signals to prevent mid-air disasters. Development of radio in the 1930s and radar technology during World War II allowed for airplane tracking and direct communication between pilots and ground operators. Fast forward, today real-time automation and artificial intelligence (AI)-powered decision-making make air traffic control one of the most sophisticated systems in the world, operating around the clock with pinpoint accuracy. 

How Machine Learning is Modernizing Clinical Trial Operations

Similarly, clinical trials are complex ecosystems of many moving parts, involving numerous sites,

investigators, patients, and regulatory obligations, aimed at answering critical healthcare questions. Similar to the inception of the aviation industry, clinical trials initially relied on manual data collection and paper-based documentation, lacking standardization and unable to adapt in a timely manner to protocol deviations and safety concerns. The establishment of Good Clinical Practice (GCP) in the 1990s provided a framework for monitoring and reporting. Strengthened by the regulatory bodies and centralized data management systems, the process began to require more stringency in documentation of trial activities, which made tracking and analyzing clinical data more efficient. Electronic data capture reduced time spent on manual data entry and improved data accuracy, while risk-based and remote monitoring optimized resource allocation and allowed sponsors and contract research organizations to access data remotely.  

Why Traditional Monitoring Falls Short in an ML-Driven Landscape

Despite being the cornerstone of medical innovation, clinical trials can be burdened by inefficiencies that delay progress and inflate costs. Traditional monitoring and reporting methods still heavily rely on manual oversight, periodic assessments, and retrospective data analysis, making it challenging to promptly detect and respond to issues. Machine learning (ML) is transforming this landscape by enabling real-time monitoring and reporting, significantly improving trial responsiveness and delivering a strong return on investment. 

Machine Learning as The Air Traffic Control Tower of Clinical Trial Monitoring and Reporting 

Managing a clinical trial is much like overseeing an air traffic. At any given moment, multiple flights (patients), airports (trial sites), and traffic routes (regulatory requirements) are in motion. Air traffic controllers must constantly track these moving parts, ensuring smooth landings, avoiding mid-air conflicts, and responding quickly to changing weather conditions. Similarly, the use of AI in clinical trials acts as a control system. Clinical research monitoring consists of a variety of complex tasks and requires close collaboration between clinical research monitoring personnel and investigators, site coordinators, and study sponsors. Protocol adherence, site audits, review of case report forms, regulatory compliance, and adverse event reporting are some of the most critical responsibilities of trial monitoring, aimed at ensuring the trial is being properly conducted. Just as airports rely on advanced automation to manage air traffic, clinical trials particularly benefit from ML-driven remote monitoring. This approach helps alleviate obstacles associated with traditional, centralized study designs. 

Decentralization and the Rise of Machine Learning–Enabled Study Designs

Regulatory flexibilities that came with COVID-19 provided an opportunity to accelerate decentralization of the clinical trial process. The pandemic-era policy changes prompted researchers to investigate new avenues for collection of data, such as completing study questionnaires via telemedicine, that would usually require a clinic visit (Kadakia et al. 2021). Beyond easing the administrative tasks, machine learning is also aiding more complex tasks, such as administering experimental drugs, adherence to medication regimens, or early detection of adverse events. 

Predictive Insights: Machine Learning’s Expanding Role in Trial Oversight

Some fascinating examples that embraced ML technologies in the trial design include facial recognition technology to confirm whether patients adhere to their medication regimens (Labovitz et al. 2017) and a piloted use of ingestible sensors to monitor the ingestion of antipsychotic medications in patients with schizophrenia. By integrating multiple data streams (i.e., wearable devices and sensors), these algorithms can be trained to derive novel digital biomarkers that enable more accurate monitoring of patients in diseases with fluctuating symptomatology (i.e., neurodegenerative diseases like Parkinson’s or Alzheimer’s disease). To address under-reporting of adverse events (AEs), a recurrent issue raised during GCP inspections and audits, one company developed an ML model to augment traditional quality assurance (QA) approaches and improve the oversight of GCP-regulated activities in their clinical studies (Ménard et al. 2019). This predictive model enables real-time monitoring of safety reporting by detecting sites that report significantly fewer AEs than the predicted threshold, triggering timely mitigation strategies by the QA program leads. It also enables a data-driven risk assessment for selection of patients and study sites instead of relying on the traditional random assignment. In conclusion, by processing vast amounts of trial data in real time, AI-enabled technologies ensure faster identification of protocol deviations, safety concerns, and data inconsistencies and reduces costly delays by pointing trial managers to potential bottlenecks before they escalate. 

Regulators Respond to Machine Learning in Clinical Research

The FDA reports that between 2016 and 2024, they have reviewed approximately 300 submissions that reference the use of AI from discovery to clinical research, post-market safety surveillance and even manufacturing. In 2024, the European Union (EU) released the AI Act, the first regulatory and legal framework for AI within the EU, pertaining also to the use of AI in pharmaceutical industry. The FDA followed and earlier this year issued the first official guidance on the use of AI for the development of drug and biological products, providing “recommendations to industry on the use of AI to produce information or data intended to support regulatory decision-making regarding safety, effectiveness, or quality of drugs.” However, these recommendations are non-binding, and the regulation is still evolving. Governing agencies are adapting to this new landscape, and regulatory and reporting practices for the use of AI in medical research are still under development.

“Garbage In, Garbage Out”  

If you have an IT friend or, like me, are married to one, it is likely that you’ve already heard about the “Garbage In, Garbage Out (GIGO)” principle. In the beloved mantra of computing, this fundamental concept emphasizes that the quality of output at a maximum can only be as good as the quality of input. 

ML models analyze massive datasets from multiple sources. If the data entering these models are incomplete, inconsistent, or even biased, the system can generate unreliable predictions and flawed risk assessments, leading to regulatory noncompliance, and introduce bias in decision making. For example, if data from electronic health records or wearable devices are incorrect or outdated, ML algorithms may fail to detect early warning signs of adverse events. If training datasets are skewed (eg, underrepresenting certain patient populations), models may generate biased insights, affecting the generalizability of trial results. Mitigation strategies are numerous; for example, ML models are periodically retrained with updated, high-quality datasets to adapt to evolving trial conditions. However, depending on the scale of a model, retraining is often a costly and lengthy operation. While automation does enhance efficiency, it does not replace human ML-generated reports to verify accuracy and context.

What Machine Learning–Enabled Trial Strategies Mean for Today’s Sponsors

ML holds enormous potential for transforming clinical trial monitoring and reporting, but its effectiveness hinges on data integrity. Implementing ML for trial monitoring and reporting translates into significant resource savings by optimizing staff allocation, reducing site monitoring visits, and minimizing costly trial delays. One of the main advantages of ML is its ability to enhance decision-making speed. ML-enabled risk-based monitoring allows faster identification of recruitment bottlenecks and safety concerns, leads to quicker resolutions, and reduces overall trial duration and associated costs. To date, the main pitfalls of widespread use of AI in medical research pertain to technical challenges, ethics in data availability, and lack of firm regulatory structure, As the industry embraces digital transformation, leveraging ML for real-time trial management will become a competitive necessity rather than a luxury.

Driving ML Adoption with Clinical and Regulatory Expertise

As the industry continues its digital transformation, the organizations that learn to pair machine learning

with strong data governance, clear documentation standards, and regulatory-aligned processes will be the ones best positioned to accelerate development safely and responsibly.

At Synterex, we help sponsors and emerging biotechs bridge that gap—bringing together clinical expertise, regulatory precision, and AI-ready workflows to support more efficient, compliant, and data-driven trials. If you’d like support implementing ML-enabled monitoring or strengthening your documentation strategy, our team is here to help.

References

Air Traffic Control. Federal Aviation Administration. Published November 16, 2021. Accessed February 22, 2024. https://www.faa.gov/about/history/photo_album/air_traffic_control 

Artificial Intelligence for Drug Development. Food and Drug Administration. Published on February 20, 2025. Accessed on March 10, 2025. https://www.fda.gov/about-fda/center-drug-evaluation-and-research-cder/artificial-intelligence-drug-development 

Considerations for the Use of Artificial Intelligence to Support Regulatory Decision-Making for Drug and Biological Products. Food and Drug Administration. Published on January 06, 2025. Accessed on March 10, 2025. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-use-artificial-intelligence-support-regulatory-decision-making-drug-and-biological 

Kadakia KT, Asaad M, Adlakha E, Overman MJ, Checka CM, Offodile AC 2nd. Virtual Clinical Trials in Oncology-Overview, Challenges, Policy Considerations, and Future Directions. JCO Clin Cancer Inform. 2021;5:421-425. doi:10.1200/CCI.20.00169 

Labovitz DL, Shafner L, Reyes Gil M, Virmani D, Hanina A. Using Artificial Intelligence to Reduce the Risk of Nonadherence in Patients on Anticoagulation Therapy. Stroke. 2017;48(5):1416-1419. doi:10.1161/STROKEAHA.116.016281 

Ménard T, Barmaz Y, Koneswarakantha B, Bowling R, Popko L. Enabling Data-Driven Clinical Quality Assurance: Predicting Adverse Event Reporting in Clinical Trials Using Machine Learning. Drug Saf. 2019;42(9):1045-1053. doi:10.1007/s40264-019-00831-4 

Miller, MI, Shih, LC., Kolachalama, VB. Machine Learning in Clinical Trials: A Primer with Applications to Neurology. Neurotherapeutics. 2023;20(4):1066-1080. doi:10.1007/s13311-023-01384-2 

Shah P, Kendall F, Khozin S, et al. Artificial intelligence and machine learning in clinical development: a translational perspective. NPJ Digit Med. 2019;2:69. Published 2019 Jul 26. doi:10.1038/s41746-019-0148-3 

The Role of Artificial Intelligence in Clinical Trial Design and Research with Dr. ElZarrad. Food and Drug Administration. Published on May 30 2024. Accessed on March 10 2025. https://www.fda.gov/drugs/news-events-human-drugs/role-artificial-intelligence-clinical-trial-design-and-research-dr-elzarrad#:~:text=AI%2C%20including%20machine%20learning%2C%20is%20all%20gaining%20traction,design%2C%20digital%20health%20technologies%2C%20and%20real-world%20data%20analytics. 

Weissler EH, Naumann T, Andersson T, et al. The role of machine learning in clinical research: transforming the future of evidence generation [published correction appears in Trials. 2021 Sep 6;22(1):593. doi: 10.1186/s13063-021-05571-4.]. Trials. 2021;22(1):537. Published 2021 Aug 16. doi:10.1186/s13063-021-05489-x