Husband, Dad, Investor, Company Builder, Philanthropist.

How Artificial Intelligence Makes for Better Clinical Trials

Artificial intelligence (AI) is working its way into different industries all around the world, and health care is no exception. The marketplace is full of digital health tools designed to offer advice to people at home or assist doctors with dosage and diagnosis in the hospital.

But there’s another way that AI can help impact the healthcare industry: clinical trial selection. For new drugs entering the market, clinical trials are essential. However, the process of getting a new drug onto the public marketplace is quite slow and often very expensive. AI shows potential to help remedy that situation, as it can get drugs to the market quicker and at a fraction of the cost. Though there are a number of ways AI tools can help in the clinical trial process, including monitoring drug adherence, optimizing processes, and monitoring pre-emptive risk, one of the most important is assisting organizations in designing better protocols and improving the trial management workflow.

One of those companies aiming to help with these issues is Trials.ai. With a mission to speed up the clinical trial process using artificial intelligence tools, Trials.ai is hoping to make a significant impact on patients’ lives by improving the clinical trial process.

Inspiration for Trials.ai

TrialsAI

Trials.ai has a heartbreaking story behind its founding, but it’s one that motivated CEO and co-founder Kim Walpole to make a difference. Before starting Trials.ai, Kim Walpole spent nearly 12 years in strategic planning for pharmaceutical companies. Then, in 2015, her best friend was diagnosed with pancreatic cancer. Immediately, Kim began reaching out to some of her friends in a desperate attempt to get him involved in clinical trials that promised potentially helpful therapies. Unfortunately, she was repeatedly told that though these treatment and therapy options looked promising, by the time they would be ready for clinical trials, he would already be gone. Heartbreakingly, that is exactly what happened. Her friend Paul passed away within eight months of his diagnosis.

Until that moment, Kim Walpole had simply viewed these issues as problems that would be solved by someone else. She knew what the issues were: archaic manual processes, logistical challenges, and more. But after her friend’s death, Kim decided that if no one else was going to try and fix things, then she would do it herself. She pulled together a small team and began working on the issue. Luckily, her team was approached by physicians at an academic institution who were actively looking to accelerate their trial process. They had heard about what Kim wanted to do and offered a grant to help with her work early on.

How Does Trials.ai Work?

In the world of clinical trials, it’s a sad reality that drugs, new devices, and therapies get bogged down in the process due to poor protocol and a lack of effective tools. Many organizations are still using spreadsheets and siloed systems (valuable data that is not being used), and even attempting to conduct trials without using real-time data.

Though sponsors spend nearly $65 billion dollars on clinical trials each year, 45 percent of these trials end up failing specifically due to poor design. Not only does that result in billions of dollars lost, but patients need the drugs and therapies that clinical trials can provide. Many of them simply don’t have the time to wait.

The amount of data collected in the process of designing and conducting clinical trials is overwhelming—it’s impossible for a human being to read and comprehend in its entirety. With nearly all organizations experiencing “dark” or “siloed” data (almost 90 percent of collected data is labeled in this way), pharmaceutical companies are making decisions based on incomplete information. Trials.ai aims to change this, leveraging AI tools to harness huge amounts of data and then using that information to provide unique insights into different therapeutic areas for clinical trial sponsors. This information is collected into a codified clinical trial database, giving organizations a holistic overview of everything they need to make intelligent decisions about the design and optimization of their clinical trial protocols.

Trials.ai works with three layers of data. The first is publicly available data taken from past trials, both successful and unsuccessful, along with research papers, journal articles, and anything else available. The second layer is generated from Trails.ai’s own data (both typical trial data and logistical data) captured from ongoing trials. The third layer comes from a built-in clause in the service agreements with trial partners that allows Trials.ai to use both historical and real-time data generated in trials to continuously inform their algorithms, making the system smarter over time.