Core Concepts
The CT-ADE dataset provides a comprehensive resource for developing advanced predictive models to forecast adverse drug events (ADEs) by integrating drug, patient population, and contextual information from clinical trial results.
Abstract
The CT-ADE dataset was developed to enhance the predictive modeling of adverse drug events (ADEs). It encompasses over 12,000 instances extracted from clinical trial results, integrating drug, patient population, and contextual information for multilabel ADE classification tasks in monopharmacy treatments.
Key highlights:
The dataset is structured to support multilabel ADE classification, reflecting the complex nature of ADEs. Annotations are standardized at the system organ class (SOC) level of the Medical Dictionary for Regulatory Activities (MedDRA) ontology.
It includes detailed information on drug molecular structures (SMILES notation), patient eligibility criteria, and treatment regimen descriptions, enabling a comprehensive analysis of factors influencing ADE occurrence.
The dataset is divided into training, validation, and test sets with no overlap in drug compounds, ensuring robust model evaluation and generalization.
Baseline models achieved promising results, with the best-performing model achieving a 73.33% F1-score and 81.54% balanced accuracy, highlighting the dataset's potential to advance ADE prediction research.
The dataset's coverage spans a wide range of System Organ Classes (SOCs) and Anatomical Therapeutic Chemical (ATC) drug classifications, demonstrating its comprehensive representation of the ADE landscape.
Feature attribution analysis using Integrated Gradients revealed that patient eligibility criteria and treatment regimen details are the most influential factors in the model's ADE predictions, underscoring the importance of contextual information beyond just drug molecular structures.
The CT-ADE dataset provides an essential tool for researchers aiming to leverage the power of artificial intelligence and machine learning to enhance patient safety and minimize the impact of ADEs on pharmaceutical research and development.
Stats
About 96% of drug candidates do not receive market approval, underscoring the inefficiencies and financial risks in drug development.
The average investment to bring a new drug to market is estimated at $1.3 billion.
Safety concerns are responsible for 17% of clinical trial failures.
Quotes
"ADEs are unexpected medical occurrences in patients administered a pharmaceutical product, potentially caused by the drug's pharmacological properties, improper dosage, or interactions with other medications."
"Recent advancements in artificial intelligence and machine learning have created a significant shift in this area, with research now intensely focused on these technologies to forecast ADEs with greater accuracy."