You are opening our English language website. You can keep reading or switch to other languages.
09.07.2024
6 min read

The Role of AI in Clinical Trials: Insights from Industry Experts

AI has the potential to significantly enhance clinical trials by speeding up research and substantially reducing duration. Companies interested in adopting the technology focus on finding seamless ways to integrate it into their operational processes and, more importantly, identifying cases with the highest ROI.

The Role of AI in Clinical Trials: Insights from Industry Experts

Article by

Daniel Piekarz
Daniel Piekarz
Andrey Sorokin
Andrey Sorokin

Recently, DataArt and Microsoft partnered to host a webinar, "The Role of AI in Clinical Trials”, to discuss how this innovative technology can save time and money in clinical trials. The panel of experts included:

  • Gene Buckley, Senior Director for Customer Success for Health and Life Sciences, Microsoft
  • Andrey Sorokin, AI Expert and Solution Architect for Healthcare and Life Science Practice, DataArt
  • Daniel Peikarz, Senior Vice President, Head of Healthcare and Life Science Practice, DataArt

In this article, we summarized the key insights from their discussion.

Implementing AI Solutions Just Got Easier

Just a few years ago, building and implementing AI tools was a time-consuming and expensive process. Businesses had to hire a team of engineers to customize AI models to their needs and provide operational support, with no guarantee of success. For example, in 2017, IBM's Watson attempted to create a clinical decision support tool using AI at a cost of 62 million dollars, but the project failed.

AI has changed; it's not what it was before. All the heavy lifting and research that had to go into it has been packaged down to the point where it's now a tool that almost anyone can implement.
Daniel Peikarz
Daniel Peikarz

Today, AI tools are much more accessible and user-friendly. Thanks to advancements in natural language processing, RAG (retrieval augmented generation), and prompt engineering, we can now interact with large language models simply by asking questions in plain English and receiving answers.

These developments offer new opportunities in the clinical trial industry. While traditionally, clinical trials have been time-consuming, costly, and complex, the latest advancements in AI tools make the process much more efficient by reducing costs and improving the accuracy and speed of data analysis.

Image
We have seen a remarkable uptake of AI tools among our clients, particularly in the last year. The advancement of large language models has been a great help for them, especially in areas like medical documentation and patient engagement.
Gene Buckley
Gene Buckley

Addressing Data Privacy Concerns

As AI becomes more powerful and widespread, data privacy concerns are one of the hot topics raised within the clinical trial industry, given the sensitive data used during these trials. The main concerns focus on whether cloud providers can access the clinical data uploaded and if this data will be used to train AI models.

We comply with stringent healthcare regulations, ensuring that Microsoft neither accesses nor utilizes your data for model training or any other purpose. Microsoft is here to provide cloud-based capabilities as defined by the client — nothing else.
Gene Buckley
Gene Buckley

Using Azure Cloud for data ensures customer security and privacy. Microsoft’s commitment to comply with strict privacy standards in handling sensitive medical information helps the company maintain trust among Azure Cloud users.

Achieving Regulatory Compliance with Ease

Remaining compliant poses an ongoing challenge for clinical trial businesses, as they need to adhere to numerous regulations and reference documents. AI can be used to accelerate clinical trials, significantly reducing the time required to get from the design phase to the regulatory submission. For example, AI can analyze past research and local regulations to generate sections of a trial protocol. Another example is an annotation of protocol documents: AI tools can automate the extraction of eligibility criteria and cut preparation time in half.

Image
In clinical trials, we often deal with at least a hundred-page documents, and it takes hundreds of man-hours just to read through the existing research papers. AI can turn this static content into an interactive Q&A tool.
Daniel Peikarz
Daniel Peikarz

Maximizing Success in Patient Recruitment

Patient recruitment is one of the most difficult and time-consuming parts of clinical trials. With over 80% of clinical trials failing to meet their patient enrolment timelines, the need for efficient recruitment strategies is more critical than ever.

At DataArt, we leveraged AI to extract inclusion and exclusion criteria for trials from databases like ClinicalTrials.gov and then mix them with electronic health record data. It allowed us to quickly find patients for trials, including those with rare diseases that only 0.1% of the population has.
Daniel Peikarz
Daniel Peikarz

Simplifying Data Processing

Processing a large volume of data is arguably the biggest challenge for companies involved in clinical trials. Cloud technologies, such as Microsoft Azure, ease this challenge as they allow processing data from thousands of sources and analyzing it in real-time. AI tools can further help businesses standardize patient data and provide insights.

I believe that AI tools can do the heavy lifting when analyzing multi-modal data, including medical imaging and unstructured documents. AI can help close the gap between patients’ electronic data in different formats and then use this data for further analysis.
Andrey Sorokin
Andrey Sorokin

DataArt’s AI Use Cases in Clinical Trials

  • Recommendation System for Clinical Trials.
    DataArt solved the expensive and time-consuming issue of distributing patient quotas across hospitals and countries by using advanced algorithms and an automated process.
  • GenAI-Powered ICF Processing Solution.
    DataArt created an innovative Generative AI solution that automates the population workflow for Informed Consent Forms (ICFs), solving the client's problems with manual filling and related inefficiencies.
  • Matching Cancer Patients with Clinical Trials.
    Using advanced NLP methods and human expertise, DataArt developed a model that speeds up the process of finding a personalized clinical trial for a specific patient.

Learn More about DataArt’s AI Expertise in Clinical Trials

DataArt and Microsoft

DataArt has been a Microsoft partner for over 20 years, equipped with the expertise in Microsoft tools necessary to enhance the effectiveness of clinical trials with AI. By leveraging AI capabilities within Microsoft's ecosystem, DataArt can streamline participant recruitment, optimize trial designs, improve data analysis, and contribute to faster and more accurate outcomes in medical research.

I'd like to reinforce the value of the partnership we see with DataArt. The collaboration between the client, DataArt, and Microsoft on various projects, and the remarkable reduction in implementation timing, for example, from 9 to 12 months down to under a quarter, is exciting for our clients.
Gene Buckley
Gene Buckley

Learn More about DataArt's Services on Microsoft Azure

Final Thoughts and Next Steps

AI has a transformative impact on clinical trials, particularly in areas such as compliance, patient recruitment, and regulatory submissions. As it continues to advance, the collaboration between industry experts and technology providers promises to drive significant improvements in efficiency, accuracy, patient safety, and the overall clinical trial experience.

To see the full webinar and get more insights on AI in clinical trials, request the recording here.

 

Interested in DataArt's AI Solutions for Clinical Trials?

Discover how our innovative AI solutions can transform the landscape of clinical trials. Join us in advancing healthcare through cutting-edge technology and transformative insights.

Contact Us!