Bridging Innovation and Trust with Artificial Intelligence

In a recent video interview with Applied Clinical Trials, Michel van Harten, MD, CEO, myTomorrows, discussed the integration of artificial intelligence (AI) in clinical trials, highlighting its potential to reduce costs, accelerate timelines, and improve inclusion. Looking forward, advancements with AI are expected to speed up drug discovery, predict drug efficacy, and develop digital twins for personalized treatment simulations, potentially reducing trial risks and costs.

ACT: Where do you think the clinical trials industry currently stands with AI integration?

van Harten: I think in general, there's a lot of optimism, especially on the industry side. Sponsors and also CROs (contract research organizations) are recognizing the potential of AI to reduce costs, accelerate timelines, and also improve inclusion. I think, for example, AI is already being actively used to design trials and protocols to identify trial sites with access to specific populations, and also to match eligible patients faster and enroll them faster.

On the healthcare provider side, besides early adopters, there's sometimes hesitancy. In general, in healthcare, you must be very careful with introducing technology. I think first, the quality of the data is everything. For example, for us, we work with sensitive patient and clinical data, and although we do not advise physicians and patients, we do give them the options, the pre-approval treatment options, so clinical trials, that are out there and they need to be correct, so our AI models are as good as the data we feed them. We invest also heavily in cleaning and validating the data before the AI part can even begin.

Then second, I think transparency and explainability is important, and they're crucial in building trust. In healthcare, patients and providers have to understand how decisions are made, so we always show them the reasoning in a very detailed way. In addition, we provide them the option to correct the AI if they disagree, so the physician or the patient is always in the driver's seat. Until the AI is perfect, you have to build in checks and balances and develop, together with those physicians and key opinion leaders to remove skepticism early on.