AI is rapidly transforming how clinical trials are designed, run, and analysed — unlocking faster patient recruitment, more accurate real-world data interpretation, and scalable insights that were once out of reach.

In this in-depth Q&A, Lucas Tanner, chief financial officer at Carta Healthcare, explains how AI-powered tools are helping healthcare organisations streamline trial processes, reduce bias, and extract value from unstructured clinical data. He also explores the growing importance of public–private partnerships, the risks of siloed innovation, and why sustained investment in medical research is more urgent than ever.

 

How is AI being used today to accelerate data analysis and reduce the time it takes to reach meaningful clinical insights?

There are five core ways AI is transforming data analysis in clinical settings:

• Automated abstraction: AI extracts key data from both structured and unstructured clinical records, dramatically reducing the time and cost of manual chart reviews.

• Faster, more accurate interpretation: By intelligently parsing clinical notes, lab results and more, AI delivers rapid, nuanced understanding of complex patient records.

• Enhanced data validation: Built-in rules help AI spot inconsistencies, gaps, or anomalies—improving reliability and data quality.

• Continuous learning models: Our AI systems refine themselves over time, improving with every interaction thanks to human-in-the-loop validation.

• Increased speed to insight: The end result is faster access to reliable insights, which supports quicker, more informed clinical decisions.

Can you give an example of how AI has directly improved clinical trial recruitment or site selection?

Absolutely. AI is transforming patient screening by identifying eligible participants in near real time—sometimes even before they walk into the clinic. This is especially valuable in community or rural healthcare settings, where trial access is limited.

AI enables proactive patient matching by flagging key clinical characteristics and treatment timelines, helping avoid the common issue of patients becoming ineligible after starting other therapies. This has the potential to dramatically improve recruitment speed and accuracy.

How do you address concerns about bias or lack of transparency in AI-driven healthcare tools?

It’s a crucial issue, and one we take seriously. We build transparency into every part of the process through five principles:

• Complete audit logs: Every step in the data journey is traceable and immutable.

• Source document traceability: Every AI-generated entry links directly to its original source—like a lab result or discharge summary.

• Human-in-the-loop validation: Clinical experts review and approve AI-suggested data for quality and accuracy.

• Bias mitigation: Our models are trained on diverse data to help prevent skewed outputs and ensure fair, representative results.

• Reproducibility and transparency: AI-generated insights come with clear explanations and can always be traced back to the original data.

What responsibility does the private sector have in maintaining momentum in medical discovery, especially as public funding declines?

Private companies are increasingly stepping in to fill the funding gap. Academic medical centres already have the infrastructure — they need the capital and innovation partners to unlock its full potential.

But it’s not just about money. Responsible innovation means investing in ethical data practices, championing open access, and aligning profit with long-term patient benefit. It’s about driving research forward, without compromising integrity.

You’ve mentioned that over 36% of the world’s data comes from healthcare. How can companies make better use of it?

The problem isn’t a lack of data — it’s that most of it is locked in unstructured formats: PDFs, handwritten notes, even faxes. AI changes that.

By rapidly transforming this “dark data” into structured, usable information, AI allows researchers to uncover patterns, understand drug efficacy, and stratify patients faster than ever before. It unlocks research that would have taken years or been financially unviable just a decade ago.

Can you share a real-world example where public–private collaboration has improved outcomes?

One example is our research into breast cancer care. Using AI, we analysed outcomes for patients undergoing sentinel lymph node biopsies across various demographics. The results led to changes in care pathways, reducing unnecessary interventions.

We also built clinical “nudges” into the workflow — subtle alerts that guide clinicians toward alternative, evidence-based options. These small shifts can lead to major improvements in patient care and outcomes.

What’s the long-term risk if we fail to adequately fund medical research?

The human cost is enormous. Life expectancy has risen significantly in the past 50 years thanks to sustained research investment. If we pull back now, we risk stalling — or reversing — that progress.

Research isn’t something that can be turned on and off. It takes years, often decades, and requires long-term thinking. If we lose a generation of researchers to short-term budget cuts, we may never recover the momentum.

Do you see a risk that private-sector innovation becomes siloed or inaccessible to those who need it most?

Yes. When innovation is driven purely by financial return, we risk neglecting conditions that aren’t seen as commercially attractive.

And when data is locked in proprietary “walled gardens,” collaboration suffers. We need interoperability and data-sharing standards to break these silos and drive real progress across the system.

If you could speak directly to policymakers, what would your message be?

Prioritise interoperability, patient data control, and secure digital infrastructure. The backbone of public health innovation is accessible, standardised, and ethically governed data.

Also, don’t underestimate the importance of public funding. Many breakthroughs — like the internet — came from publicly funded research. The private sector can’t, and shouldn’t, carry the full burden. We need both working together to move the industry forward.