5 Minute HealthTech Jargon Buster: Artificial Intelligence for Clinical Trials

The growth of technology and its use in clinical trials provides an abundance of opportunities to improve efficiency and management. Technology can be utilised in various areas including more efficient documentation, preclinical predictions, centralising data and improving consent.

Electronic health records

Documentation was previously on paper, however, electronic medical records are an electronic source of documentation, to help store and manage patient records on an online format [2].  Electronic health records (EHRs) offer numerous benefits for clinical trials, making research more efficient and cost-effective. They enable faster patient recruitment by identifying eligible participants through existing medical data. EHRs also streamline data collection, reducing manual entry errors and improving accuracy [3]. Additionally, they support real-world evidence studies by integrating research into routine clinical care, making findings more applicable to everyday practice. By automating processes, EHRs save time, lower costs, and enhance the overall quality of clinical trials [3].

Clinical trial management systems

Software platform developments such as the clinical trial management system (CTMS) offer multiple improvements to clinical trials. CTMS is a type of software platform which manages clinical trials, centralising the data collected from multiple sources.  They are often used in conjunction with electronic health records to provide integrated data. Additionally, CTMS provides compliance support by maintaining accurate documentations, which can be used during audits and proof of FDA compliance.

eConsent

Patient consent is critical in participating in trials; thus, the prevalence of electronic consent (eConsent) provides an opportunity for consent to be monitored via an electronic system. This increases participant engagement through interactive features, such as videos and diagrams. EConsent has real-time tracking features, which allows for continuous monitoring and updating of consent throughout the trial. This ensures compliance and patient tracking [2].

Artificial intelligence (AI) presents an opportunity to improve efficiency and time management of clinical trials throughout multiple stages. Figure 1 displays an overview of key opportunities, such as preclinical use and analysis and the challenges of AI in clinical trials.

Detection of drug toxicity

AI is significantly improving drug development, particularly in the preclinical stages, by enhancing both efficiency and safety. One of its most valuable applications is predicting drug toxicity, as highlighted in Figure 2, enabling researchers to identify high-risk compounds much earlier in the development process. By utilizing machine learning models trained on extensive datasets, AI can accurately assess potential toxicity, which helps reduce the need for traditional in vitro testing. This not only shortens the timeline for drug development but also lowers costs by filtering out unsuitable compounds before they advance to clinical trials.

Drug design and adverse effects

Furthermore, AI plays a critical role in identifying potential adverse effects early, helping to prevent harm to participants in clinical trials and reducing risks associated with post-market exposure. By providing these early insights, AI allows for better-informed clinical decision-making regarding the safety and design of drugs. This contributes to faster market access for drugs while minimizing financial losses that could arise from costly errors in drug development. As a result, AI’s integration into preclinical trials enhances the overall success rates and safety of clinical research. [5]

An example of utilising AI in preclinical stages is in oncology trials. AI was used to streamline participant recruitment. Cloud-based AI systems can scan electronic health records and databases to identify eligible participants based on predefined inclusion and exclusion criteria. This approach enhances recruitment efficiency while reducing the workload on clinical staff, making the process faster and less labour-intensive [2]. 

Trial classification

Additionally, AI clinical management systems can be utilised to organise patient visits. For example, via the use of an automated trial classification system. This system categorises clinical trials, based on specific criteria, using algorithms and machine learning models to do so. For example, automated trial classification systems have been used to facilitate patient-trial matching to find individuals living with HIV and pregnant women. Using AI is particularly valuable as there were large datasets for recruitment, which would take clinicians a long time to manage [6].

Regulatory and policy concerns are crucial for implementing AI in clinical trials, focusing on data integrity, patient safety, and compliance with guidelines. Agencies like the FDA and EMA are working to ensure AI methods align with existing frameworks, prevent biases, and maintain ethical standards in clinical research.

FDA Strategies for the US 

Regulatory and policy concerns surrounding the implementation of AI in clinical trials focus on ensuring data integrity, patient safety, and compliance with existing regulatory frameworks to offset the use of AI. According to the FDA [8] the use of AI in clinical trial design must align with established guidelines to prevent biases in data collection, ensure transparency in decision-making, and maintain ethical standards in patient recruitment and monitoring.

Additionally, the FDA highlight concern about the reliability and reproducibility of AI-driven models used in drug development, emphasizing the need for rigorous validation and regulatory oversight to ensure that AI-generated insights are scientifically robust and do not compromise patient outcomes [9]. As AI continues to evolve, regulatory agencies must establish clear policies that balance innovation with safety, ensuring that AI-driven methodologies enhance, rather than undermine, the integrity of clinical trials.

EU Strategies

The European Medicines Agency (EMA) has outlined its Regulatory Science Strategy to 2025, emphasizing the need for a robust regulatory framework for AI in clinical trials. Pharmaceutical sector stakeholders are working alongside academia and expert centers to develop guidelines for AI validation and assessment, ensuring the safe and effective integration of AI-driven technologies in clinical research [10].