Role of Automation, Natural Language Processing, Artificial Intelligence, and Machine Learning in hospital settings to identify and prevent Adverse Drug Reactions

Abstract

Patient Safety is at the center of all pharmacovigilance activities. As several covariates can impact the safety of a medicinal product in patients, a large amount of data is required for an accurate assessment of the safety and therefore, the benefit-risk balance of a medicinal product. Natural language processing, Artificial Intelligence, and Machine Learning are being popularly used to facilitate various pharmacovigilance activities in the Pharma industry. Artificial Intelligence and Machine learning if properly used in hospital settings can also facilitate the identification of adverse events from hospital records and discharge summaries and prescription errors, thus, alerting treating physicians regarding the same. However, the potential of using these techniques needs to be fully explored in hospital settings to facilitate the collection and evaluation of safety data.

Introducing AI to the molecular tumor board: one direction toward the establishment of precision medicine using large-scale cancer clinical and biological information

“…The first is toward the utilization of data from physicians’ daily practice, represented by medical records (electronic) [ 113 – 115 ]. For example, automatic extraction of adverse drug reaction signals from electronic medical records is being attempted [ 116 – 118 ]. The second major trend is the use of NLP to analyze published data, such as medical articles and case reports, to extract important information for clinical applications [ 119 – 121 ].…”