Abstract

Drug-induced convulsion may be a severe adverse event; but, no helpful biomarkers for it are discovered. We tend to propose a replacement methodology for predicting drug-induced convulsions in monkeys supported pulse variability (HRV) and a machine learning technique. As a result of involuntary nervous activities area unit altered round the time of a convulsion and such alterations have an effect on HRV, they’ll be expected by watching HRV. Within the planned methodology, abnormal changes in multiple HRV parameters area unit monitored by means that of a convulsion prediction model and convulsion alarms are issued once abnormal changes in HRV area detected. The convulsion prediction model is constructed based on multivariate statistical process control (MSPC), a well-known anomaly detection algorithm in machine learning.