Wednesday Research Seminar Series - AI for Smarter Post-Market Drug Surveillance

We are pleased to invite you to our Wednesday Research Seminar. It will be held in-person on 26th February at Oasis Theatre. Wednesday Research Seminar Series was launched in 2008 and has featured more than 360 presentations to date. The seminars provide a forum for researchers to share their work. Presenters include faculty from Middlesex University Dubai and other universities in the United Arab Emirates, as well as researchers from other global institutions. Dr. Elena and Dr Evgenii will present their work on:

“AI for Smarter Post-Market Drug Surveillance: Adjusting Online Ratings for Fair Comparisons and Analyzing Side Effect Impact on Patient Satisfaction”
Elena Pokryshevskaia & Evgenii Antipov

Abstract

Can online drug ratings be trusted? While millions of patients rely on platforms like WebMD and Drugs.com, these ratings often reflect hidden biases—shaped by patient demographics, condition severity, and treatment expectations—making fair drug comparisons difficult. At the same time, side effects significantly influence patient satisfaction, yet their impact is often overlooked in traditional analyses. This research presents an AI-driven framework for improving post-market drug surveillance by adjusting online ratings for fairness and analyzing the nuanced relationship between side effects and satisfaction. Using machine learning, statistical modeling, and natural language processing (NLP), we identify and correct biases, extract meaningful insights from patient reviews, and quantify how side effects shape adherence and overall satisfaction. Preliminary findings highlight significant distortions in existing ratings and reveal new patterns in patient-reported experiences. Our ongoing work aims to refine these methods, integrate electronic health records for validation, and explore predictive modeling for patient satisfaction. This research contributes to a more accurate and patient-centered approach to drug evaluation, with implications for healthcare providers, regulators, and pharmaceutical companies. We will also discuss directions for future studies, including potential NIH-funded research in AI-driven pharmacovigilance.

Presenter Bios

Dr. Elena Pokryshevskaia, PhD is an Associate Professor and a senior research fellow at HSE University specializing in AI-driven business analytics, machine learning, and data-driven decision-making. With a strong background in optimization, econometrics, and predictive modeling, her research addresses computational challenges in business, healthcare, and smart systems. She has published extensively in top peer-reviewed journals and led multiple industrydriven research projects, including AI-powered forecasting, demand modeling, and automated decision-support tools. An award-winning educator, Dr. Pokryshevskaia integrates real-world applications into her teaching, drawing from her experience in developing AI-based decisionsupport systems, web applications, and predictive models for major industries. She equips students with practical skills in programming languages and advanced analytics, guiding them to build data-driven solutions that mirror industry challenges.

Dr Evgenii Antipov, PhD is an Associate Professor at Canadian University Dubai specializing in AI-powered business analytics, high-dimensional econometrics, and causal impact evaluation. His research applies machine learning, optimization, and structural econometric modeling to decision-making problems in business, economics, and operations research. He has published extensively in leading peer-reviewed journals and has led research projects in demand estimation, pricing optimization, predictive modeling, and market intelligence. Beyond academia, Dr. Antipov is the director of a data analytics consultancy, where he has collaborated with government agencies and multinational corporations on AI-driven analytics and impact evaluation. He has also worked on fiscal policy, tax compliance, fraud detection, and strategic decision-making projects with top-tier consulting firms. His teaching integrates Python, R, and simulation-based methods to train students in data-driven decision-making and computational modeling, aligning with the evolving needs of AI-driven research in econometrics, business intelligence, and operations research.