The key to success is using data effectively
One of the treads that ran through many of the conference discussions was that optimized use of data remains a key for success. Alexander Krupp noted that Bayer placed the emphasis on the secondary use of data that revolves around addressing the pain points of the organization, rather than simply placing all the data in a data lakehouse. This approach helps the company tackle some of the main challenges in the clinical data space, such as improving access to data, ensuring seamless governance, promoting a collaborative environment and making access to AI tools easy.
Utilizing data efficiently can be done in multiple ways. The first one is by using AI tools to extract insights from it. Machine learning models are very effective at finding “the needle in the haystack”. Therefore, they can help researchers to more deeply understand the patient population, epidemiology, as well as the current standard of care, and treatment guidelines to design more effective methods.
The second way is to utilize publicly available unstructured information. This can help researchers inform the design of clinical trials and design smaller-scale trials that require less resources to execute and deliver the same statistical powers. As clinical trials are still one of the most costly steps of drug development, companies should focus on optimizing this step to bring down the costs of therapies.
Scaling AI solutions remains a challenge for organizations
For many companies, dealing with the large amounts of data that AI solutions require remains a challenge. Scaling up data infrastructures requires specialized knowledge, especially when it comes to often-overlooked things like operations data. Robert McGregor gave an example of how Novartis focuses its efforts on a specific user group that needs to solve a very specific question. To do so effectively, his team brings insights from multiple systems and combines them to provide actionable information for planning operations: “That allows us to stay reactive and to stay ahead of the curve,” he said.
Bringing in real-world evidence will deliver additional impact
In his short talk, Patrick Loerch, Senior Vice President, Clinical Data Science at Gilead, highlighted the value of incorporating real-world evidence base into medical research. Patient records, for example, hold a wealth of data that can now be fully tapped into thanks to the shift from manual to electronic chart extraction in the pharmaceutical industry. As an example, Loerch used a recent study where researchers tested ChatGPT on oncology charts. The model was able to extract data features with 80% accuracy. But because AI can provide the results within minutes and at a fraction of the cost (around $10), it provides an opportunity to process the charts of hundreds of patients and extract powerful, population-scale insights.
Bringing in healthcare transformation together
It is no accident that BiotechX Europe has secured its spot as one of the biggest gatherings of researchers and innovators working to transform healthcare. This year’s conference was an outstanding event, full of insightful discussions and community building. This is the reason why Ardigen keeps coming back to this conference every year.
As the BioTechX website states, “We have to work together in an interdisciplinary manner, also across borders of any kind.” Ardigen embodies that ethos in everything we do: from building a world-class team of multi-disciplinary experts to the diversity of partners we serve. If you want to know what it’s like working with one of the top AI CROs, reach out to us to learn more about our services.
