Leveraging Machine Learning for Predicting Molecular Refractivity in Drug Toxicology

• Published in: International Conference on Signal Processing and Communication (Online) pp. 475 - 480
• Main Authors: Joshi, Saloni, Gabrani, Reema, Beniwal, Ruby
• Format: Conference Proceeding
• Language: English
• Published: IEEE 20.02.2025
• Subjects: Accuracy; Drug; Drugs; Machine learning; Machine learning algorithms; Metabolism; Prediction; Predictive models; QSAR; Real-time systems; Refractive index; Safety; Toxicity; Toxicology
• ISSN: 2643-444X
• DOI: 10.1109/ICSC64553.2025.10968736

The integration of electronics engineering into molecular refractivity prediction is transforming drug toxicology studies. By combining biosensors, machine learning algorithms, and high-speed data systems, scientists can better predict toxicological profiles of drugs. This interdisciplinary approach improves safety and efficacy in drug development. Electronics engineering enables real-time monitoring of drug metabolism through wearable devices and predictive models, identifying toxic metabolites early and reducing adverse reactions. These technologies enhance the assessment of metabolic processes, guiding safer drug development. Machine learning algorithms further refine predictions by analyzing large datasets, identifying toxicity patterns, and improving personalized treatment plans, ultimately reducing time and costs in toxicology testing. In this research work, machine learning classifier accuracy has been achieved 85% to identify toxic drugs and toxicity prediction accuracy by using the Bayesian Neural Network was 95.8%.