Industry Perspective on Artificial Intelligence/Machine Learning in Pharmacovigilance

Kassekert, Raymond; Grabowski, Neal; Lorenz, Denny; Schaffer, Claudia; Kempf, Dieter; Roy, Promit; Kjoersvik, Oeystein; Saldana, Griselda; ElShal, Sarah

doi:10.1007/s40264-022-01164-5

Cited by 17 publications

(11 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Even before the pandemic, automation and its potential benefits for PV activities were well recognized (Kassekert et al, 2022). Rules based automation, also known as robotic process automation (RPA), are well established and have been routinely used for several years by companies to assist in the processing of individual case safety reports [ICSRs; see (Kassekert et al, 2020) for an example].…”

Section: Introductionmentioning

confidence: 99%

An industry perspective on the use of machine learning in drug and vaccine safety

Painter

Kassekert²,

Bate³

2023

Front. Drug Saf. Regul.

Self Cite

View full text Add to dashboard Cite

In recent years there has been growing interest in the use of machine learning across the pharmacovigilance lifecycle to enhance safety monitoring of drugs and vaccines. Here we describe the scope of industry-based research into the use of machine learning for safety purposes. We conducted an examination of the findings from a previously published systematic review; 393 papers sourced from a literature search from 2000–2021 were analyzed and attributed to either industry, academia, or regulatory authorities. Overall, 33 papers verified to be industry contributions were then assigned to one of six categories representing the most frequent PV functions (data ingestion, disease-specific studies, literature review, real world data, signal detection, and social media). RWD and social media comprised 63% (21/33) of the papers, signal detection and data ingestion comprised 18% (6/33) of the papers, while disease-specific studies and literature reviews represented 12% (4/33) and 6% (2/33) of the papers, respectively. Herein we describe the trends and opportunities observed in industry application of machine learning in pharmacovigilance, along with discussing the potential barriers. We conclude that although progress to date has been uneven, industry is very interested in applying machine learning to the pharmacovigilance lifecycle, which it is hoped may ultimately enhance patient safety.

show abstract

Section: Introductionmentioning

confidence: 99%

An industry perspective on the use of machine learning in drug and vaccine safety

Painter

Kassekert²,

Bate³

2023

Front. Drug Saf. Regul.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Artificial intelligence technologies used in pharmacovigilance systems [3,6,8] Технология Technology [10]. Таким образом, помимо качественной обработки сообщений о нежелательных реакциях, поступающих от участников системы фармаконадзора, использование прогрессивных технологий позволяет осуществлять обзор источников литературы, мониторинг информации социальных сетей и оказать поддержку специалистам в процессах принятия решений [11,12]. Наиболее эффективно применение представленных технологий автоматизации в комплексе: интеллектуальная автоматизация должна дополняться автоматизацией на основе правил для облегчения рабочих процессов и элементами машинного обучения, оптимизированными для определенных наборов данных или задач [12].…”

Section: уровни автоматизации рабочих процессов в электронной системе...unclassified

“…Таким образом, помимо качественной обработки сообщений о нежелательных реакциях, поступающих от участников системы фармаконадзора, использование прогрессивных технологий позволяет осуществлять обзор источников литературы, мониторинг информации социальных сетей и оказать поддержку специалистам в процессах принятия решений [11,12]. Наиболее эффективно применение представленных технологий автоматизации в комплексе: интеллектуальная автоматизация должна дополняться автоматизацией на основе правил для облегчения рабочих процессов и элементами машинного обучения, оптимизированными для определенных наборов данных или задач [12]. Кроме того, по мнению исследователей, сокращение объема рутинной работы специа-листов по фармаконадзору при использовании AI позволит повысить ценность их деятельности, поскольку их квалификация, опыт и время будут использоваться в основном для экспертной оценки полученных результатов [11].…”

Section: уровни автоматизации рабочих процессов в электронной системе...unclassified

New Technologies in Electronic Pharmacovigilance Systems for Marketing Authorisation Holders

Loginovskaya

Kolbatov²,

Sukhov³

et al. 2022

Bezopasnost' i risk farmakoterapii

View full text Add to dashboard Cite

Emergent ways to obtain information on the safety of medicinal products give relevance to the implementation of new information technologies into big data analysis in pharmacovigilance.The aim of the study was to systematise data on the use of information technologies for pharmacovigilance process automation and identify problems and limitations that may arise when introducing the technologies.Materials and Methods: the authors analysed literature on the subject matter and the practical experience of Flex Databases with the development of the electronic system for pharmacovigilance data processing designed for marketing authorisation holders.Results: using the electronic pharmacovigilance system by Flex Databases as an example, the authors demonstrated the feasibility of basic, robotic, and cognitive automation and artificial intelligence technologies for data processing. Automation technologies allow the users to streamline information entry, process and analyse data, create reports and metrics, timely submit the reports and metrics to regulatory authorities, and manage risks and safety signals; they also help specialists in decision making. Artificial intelligence technologies (a wide range of technologies including machine learning, neural networks, and automatic natural language processing) are used to collect safety reports, amongst other things, through real-world clinical data analysis; prepare summary reports; and manage risks and safety signals. Moreover, human involvement is necessary only at certain stages, particularly to process the data on exceptional cases and to analyse the results in an expert capacity.Conclusions: there is demand for process automation and artificial intelligence technologies at all stages of collection and analysis of pharmacovigilance information, from receiving a safety report to submitting it to regulatory authorities and identifying a safety signal. The deployment of the technologies within pharmacovigilance systems helps to increase the amount of data processed, among other things as a result of the inclusion of real-world clinical data into the search process. As the technologies reduce the degree of human involvement into routine processes of data collection, entry, verification, and analysis, the likelihood of errors reduces as well, whereas the quality and accuracy of the obtained results improve.

show abstract

“…A recent industry survey on the use of AI and machine learning in pharmacovigilance indicated that pharmacovigilance organizations intend to move rapidly with the planning, piloting, and production implementation of intelligent automation, while also pointing out that regulatory guidance is a potential challenge 8 …”

Section: Introductionmentioning

confidence: 99%

“…[2][3][4][5][6][7] A recent industry survey on the use of AI and machine learning in pharmacovigilance indicated that pharmacovigilance organizations intend to move rapidly with the planning, piloting, and production implementation of intelligent automation, while also pointing out that regulatory guidance is a potential challenge. 8 Medicines Regulators also need innovation that can deliver for public health. The recent coronavirus SARS-CoV-2 vaccination campaign is a good example of this potential application of artificial intelligence.…”

Section: Introductionmentioning

confidence: 99%

Artificial intelligence in pharmacovigilance: A regulatory perspective on explainability

Kurz

2022

Pharmacoepidemiology and Drug

View full text Add to dashboard Cite

Displacement or augmentation of human pharmacovigilance activities by algorithms can introduce unexpected risks and ethical concerns that must be identified and managed.• Deep learning, or black-box, models are particularly difficult to scrutinise as their mathematical properties prevent them from being directly interpretable.• Explainable models help interrogate deep learning models, but merely provide an approximate understanding of the relationship between the data and the predictions.• By default, pharmacovigilance models should require explainability, however pharmacovigilance activities have some error tolerance where explainability might be traded for better model performance, on a risk-based approach.

show abstract

Industry Perspective on Artificial Intelligence/Machine Learning in Pharmacovigilance

Cited by 17 publications

References 13 publications

An industry perspective on the use of machine learning in drug and vaccine safety

An industry perspective on the use of machine learning in drug and vaccine safety

New Technologies in Electronic Pharmacovigilance Systems for Marketing Authorisation Holders

Artificial intelligence in pharmacovigilance: A regulatory perspective on explainability

Contact Info

Product

Resources

About