Harnessing AI Technology to Identify Adverse Events with Great Accuracy and Efficiency

The use of technology in clinical trials, including artificial intelligence (AI), is evolving rapidly. According to the consulting firm Gartner, “The life science industry stands at an inflection point driven by emerging technologies spanning AI, automation, and advanced analytics.” While the hype of AI in clinical trials has largely been focused on participant-related functions, including recruitment and enrollment, the possible applications of AI across the entire drug development timeline are nearly infinite.

At the same time, increased development of advanced therapeutics and vaccines as well as greater use of accelerated approval pathways require closer surveillance for safety concerns. In “Forward Thinking for the Introduction of AI into Clinical Trials,” the Association of Clinical Research Professionals writes that “patient monitoring can be improved by incorporating the use of AI, which in turn can help improve patient safety and reduce the risk of adverse events.” By harnessing AI tools and models, researchers can identify adverse events (AEs) more accurately and efficiently.

At Veranex, we are committed to leveraging AI to significantly improve pharmacovigilance and postmarketing surveillance. Our Biometrics department has enhanced its clinical monitoring capabilities by integrating a sophisticated algorithm designed for detailed categorization of Adverse Events (AEs). This AI-powered method analyzes textual data associated with AEs including, descriptions and related queries, to classify their “tone” into distinct categories. This technology is highly promising for more effectively identifying, analyzing, and organizing diverse signals in postmarketing surveillance activities.

Based on natural language processing (NLP) technology, this clinical monitoring algorithm tool leverages an acclaimed BERT embedding model, that has been finetuned on an extensive dataset of clinical records and medical texts. Our algorithm analyzes the descriptions of adverse events, data management queries, and similar texts, allowing for the precise classification of records. The classification process prioritizes the most critical cases and enhances the quality of data by identifying potential mismatches in coding. The deployment of this innovative algorithm marks a significant step forward in improving AE management within clinical trials.

Veranex knows that patients come first. That’s why our industry-leading experts prioritize high-quality, timely, and reliable surveillance of safety data. By incorporating scientifically validated, AI-based technology into our clinical monitoring services, we are adding to the robustness of our safety system to better address the unique challenges posed by ever-evolving regulatory requirements, increased cost and complexity of clinical trials, and the overall need for closer safety surveillance.