Title: Role of low-code automation in healthcare delivery, with a focus on improving operational efficiency and outcomes
Presenter: Emmanuel Fagbenle
Abstract: Persistent healthcare access inequities continue to affect medically underserved areas (MUAs) globally, driven by a combination of clinician shortages, geographic isolation, fragmented systems, and poor digital infrastructure. Administrative bottlenecks and the limited availability of adaptable health information systems further exacerbate disparities in care delivery, particularly in rural and low resource settings. As healthcare systems transition toward digital first models, there is a growing need for scalable, inclusive technologies that can be rapidly deployed by non-technical stakeholders. This review explores the potential of low code development platforms, with a focus on the Microsoft Power Platform, as tools for advancing healthcare equity in underserved contexts. Using a conceptual framework and a literature-based analysis, the paper critically examines the health-related applications of key Power Platform components Power BI, Power Apps, Power Automate, Power Pages, Power Virtual Agents.
Title: Learning-based distributed ambulatory care scheduling
Presenter: Amirhossein Moosavi
Abstract: This work studies an ambulatory care scheduling problem for a geographically distributed healthcare center offering multi-appointment, multi-class, and multi-priority treatments. The problem is considered in a dynamic environment characterized by uncertain patient arrivals and emergency department utilization, formulated as an infinite-horizon Markov decision process. Given the limitations of conventional methods in solving large-scale instances, a neural network is integrated within the Markov decision process model to simplify feasibility constraints while ensuring adherence to the problem’s assumptions. Additionally, two straightforward, easy-to-implement scheduling policies are derived from this approach. Simulation results indicate that the approximate optimal policy and derived heuristic rules significantly outperform alternative scheduling methods. Through a case study, we show that our approach offers booking clerks with efficient, data-driven scheduling rules that substantially improve over scheduling templates.