#drug-safety News & Analysis

4 articles tagged with #drug-safety. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

4 articles

AINeutralarXiv – CS AI · Jun 237/10

🧠

DrugBench: Evaluating AI Control Protocols for Medication Harm Mitigation

Researchers introduce DrugBench, a benchmark for evaluating AI safety protocols in medical LLM applications, combining 3,671 medical conversations with FDA drug data to test systems against medication-related harms. The study reveals that existing AI control mechanisms can be circumvented and proposes severity-based monitoring to better account for the potential consequences of unsafe outputs in clinical contexts.

AINeutralarXiv – CS AI · Jun 55/10

🧠

Unsupervised Pattern Analysis in Japanese Veterinary Toxicology: A Regulatory-Compliant Framework for Cross-Species Risk Assessment

Japanese researchers developed an unsupervised machine learning framework for analyzing adverse drug events in veterinary medicine, identifying species-specific toxicity patterns from 4,120 ADE reports. The regulatory-compliant approach achieved 83% alignment with pharmacological classes and discovered distinct toxicity profiles across companion animals, ruminants, and sheep, offering improved interpretability for drug safety assessment.

AIBullisharXiv – CS AI · May 286/10

🧠

From Detection to Mechanism: Cross-Attention Graph Neural Networks Enable Drug-Drug Interaction Type Prediction An Ablation Study with Acetylsalicylic Acid Validation

Researchers demonstrate that Cross-Attention Graph Neural Networks significantly outperform traditional architectures for predicting drug-drug interaction mechanisms, improving multi-class classification by 45% while showing minimal gains in binary detection. Validation on acetylsalicylic acid pairs confirms the approach's effectiveness, suggesting atom-level inter-molecular communication is critical for mechanism-type prediction rather than simple interaction detection.

AINeutralarXiv – CS AI · Mar 34/105

🧠

Exploring Drug Safety Through Knowledge Graphs: Protein Kinase Inhibitors as a Case Study

Researchers developed a knowledge graph framework that integrates diverse data sources to predict adverse drug reactions for protein kinase inhibitors. The system combines drug-target data, clinical literature, trial metadata, and safety reports into a unified network for better drug safety analysis and pharmacovigilance.