One area where AI is making significant strides is in optimizing clinical trial strategies. In a recent interview with Wing Lon Ng, director of AI engineering at IQVIA's Applied Data Science Centre, we delved into the work being done to revolutionize clinical trials through the power of AI.
The interview took place during the 2024 SCOPE Summit held in Orlando, Florida, and was to discuss the company’s strategies and where they were going with AI in clinical trials.
At the heart of IQVIA's strategy lies the use of generative AI, a cutting-edge technology that excels in processing unstructured text data. Wing Lon highlighted how this technology enables the development of algorithms that can swiftly analyze vast amounts of data to create optimized clinical trial strategies. What sets generative AI apart is its ability to understand natural language commands, making it accessible even to non-experts in machine learning.
He said: “I head up a team of 10 machine learning data scientists who develop machine learning algorithms that power software applications to help IQVIA create optimized clinical trial strategies. The new strategy my team is trying to follow is to double down on the use of AI to make our processes faster and more effective – and the main component is generative AI.”
Algorithms and generative AI synergy
The synergy between IQVIA's algorithms and generative AI empowers stakeholders across the healthcare spectrum to effortlessly tailor clinical trial strategies to their specific needs. This democratization of the process opens doors for more inclusive and effective trials, ultimately benefiting patients and advancing medical research.
In our discussion, Wing Lon emphasized the importance of retaining human oversight in an era increasingly driven by automation.
He said: “While AI streamlines processes and enhances efficiency, human expertise remains indispensable in ensuring the ethical and safe implementation of algorithms. Additionally, the ethical use of AI in addressing issues such as diversity in clinical trials emerges as a pivotal consideration. By leveraging AI to overcome historical disparities in trial participation, IQVIA aims to foster more representative and inclusive research outcomes.”
One of IQVIA's key strengths lies in its extensive repository of clinical trial data. With access to comprehensive information on trial outcomes and patient demographics, IQVIA is able to train machine-learning models with high accuracy. Wing Lon says this data advantage positions IQVIA as a leader in the field, enabling the development of highly effective and tailored solutions for optimizing clinical trials.
Expedited timelines for trial setup and execution
IQVIA's adoption of generative AI isn't just about efficiency—it's about fundamentally transforming the way clinical trials are conducted. By automating tasks such as site planning and resource allocation, AI accelerates processes that traditionally relied on manual intervention. The result Wing Lon says is expedited timelines for trial setup and execution, ultimately expediting the delivery of life-saving treatments to patients worldwide.
He told OSP: “There is a key theme in this area of work, and that is that AI is going to remove the human in the loop, and the main consensus that I fully agree with is we don’t do it. We still require subject matter experts to be human – to oversee everything and make sure it is running effectively, doing what it is meant to be doing, and not creating sudden solutions that could have adverse effects.
“The processes do require additional monitoring and the setting up of reasonable guard rails to make sure the algorithm is working better than before but also in a sensible manner.”
Another hot topic for IQVIA is ethics and diversity and following guidance from the US Food and Drug Administration, (FDA), this is a key area those involved in clinical trials are keen to keep abreast of.
Diversity and ethnic representation
Wing Lon said: “Certain populations have historically always been underrepresented in clinical trials. And if we teach a machine how to overcome this, we can achieve this goal much faster than we could have without it. AI has not created this issue; it has always been an issue, but we can use AI to solve it.
“For example, if the FDA says it can not see any Hispanic representation, one way we can use AI is to tell the machine to prioritize hospitals that are closer to a Hispanic population and of course, you can do this for any ethnic group or population you are interested in – we can then gradually solve this problem.
“I also think where IQVIA excels is we have access to more clinical data than a lot of companies which gives us the ability to train our machines better.”
Wing Lon went on to say that at the core of IQVIA's mission is a commitment to driving positive change in healthcare. While his background may not be rooted in medicine, his passion for leveraging data to improve patient outcomes is palpable. From contributing to the rapid rollout of vaccines during the COVID-19 pandemic to advancing the efficiency and inclusivity of clinical trials, IQVIA's approach to data-driven healthcare, he emphasized, is paving the way for a brighter future.
In conclusion, Wing Lon strongly believes IQVIA's embrace of generative AI exemplifies the transformative potential of technology in revolutionizing healthcare. He said that by harnessing the power of AI, IQVIA is not only streamlining clinical trial processes but also fostering a more equitable and patient-centric approach to medical research.
He said: “As we look towards the future, the marriage of data science and healthcare holds immense promise for improving lives and driving innovation in the years to come.”
