The self-aware diabetic patient software agent model

Multi-Agent Optimization for Residential Demand Response under Real-Time Pricing

Demand response (DR) programs encourage consumers to adapt the time of using electricity based on certain factors, such as cost of electricity, renewable energy availability, and ancillary request. It is one of the most economical methods to improve power system stability and energy efficiency. Residential electricity consumption occupies approximately one-third of global electricity usage and has great potential in DR applications. In this study, we propose a multi-agent optimization approach to incorporate residential DR flexibility into the power system and electricity market. The agents collectively optimize their own interests; meanwhile, the global optimal solution is achieved. The agent perceives its environment, predicts electricity consumption, and forecasts electricity price, based on which it takes intelligent actions to minimize electrical energy cost and time delay of using household appliances. The decision-making action is formulated into a convex program (CP) model. A distributed heuristic algorithm is developed to solve the proposed multi-agent optimization model. Case studies and numerical analysis show promising results with low variation of the aggregated load profile and reduction of electrical energy cost. The proposed approaches can be utilized to investigate various emerging technologies and DR strategies.

A Systematic Literature Review of Agents Applied in Healthcare

Intelligent agents and healthcare have been intimately linked in the last years. The intrinsic complexity and diversity of care can be tackled with the flexibility, dynamics and reliability of multi-agent systems. The purpose of this review is to show the feasibility of applying intelligent agents in the healthcare domain and use the findings to provide a discussion of current trends and devise future research directions. A review of the most recent literature (2009-2014) of applications of agents in healthcare is discussed, and two classifications considering the main goal of the health systems as well as the main actors involved have been investigated. This review shows that the number of published works exhibits a growing interest of researchers in this field in a wide range of applications.

A signal processing application for evaluating self-monitoring blood glucose strategies in a software agent model

We propose the signal processing technique of calculating a cross-correlation function and an average deviation between the continuous blood glucose and the interpolation of limited blood glucose samples to evaluate blood glucose monitoring frequency in a self-aware patient software agent model. The diabetic patient software agent model [1] is a 24-h circadian, self-aware, stochastic model of a diabetic patient's blood glucose levels in a software agent environment. The purpose of this work is to apply a signal processing technique to assist patients and physicians in understanding the extent of a patient's illness using a limited number of blood glucose samples. A second purpose of this work is to determine an appropriate blood glucose monitoring frequency in order to have a minimum number of samples taken that still provide a good understanding of the patient's blood glucose levels. For society in general, the monitoring cost of diabetes is an extremely important issue, and these costs can vary tremendously depending on monitoring approaches and monitoring frequencies. Due to the cost and discomfort associated with blood glucose monitoring, today, patients expect monitoring frequencies specific to their health profile. The proposed method quantitatively assesses various monitoring protocols (from 6 times per day to 1 time per week) in nine predefined categories of patient agents in terms of risk factors of health status and age. Simulation results show that sampling 6 times per day is excessive, and not necessary for understanding the dynamics of the continuous signal in the experiments. In addition, patient agents in certain conditions only need to sample their blood glucose 1 time per week to have a good understanding of the characteristics of their blood glucose. Finally, an evaluation scenario is developed to visualize this concept, in which appropriate monitoring frequencies are shown based on the particular conditions of patient agents. This base line can assist people in determining an appropriate monitoring frequency based on their personal health profile.

Fast and frugal trees: translating population-based pharmacogenomics to medication prioritization

Aim: Fast and frugal decision trees (FFTs) can simplify clinical decision making by providing a heuristic approach to contextual guidance. We wanted to use FFTs for pharmacogenomic knowledge translation at point-of-care. Materials & Methods: The Pharmacogenomics for Every Nation Initiative (PGENI), an international nonprofit organization, collects data on regional polymorphisms as a predictor of metabolism for individual drugs and dosages. We advanced FFTs to work with PGENI pharmacogenomic data to produce medication recommendations that are accurate, transparent and straightforward to automate. Results: By streamlining medication selection processes in the PGENI workflow, information technology applications can now be deployed. Conclusion: We developed a decision tree approach that can translate pharmacogenomic data to provide up-to-date recommended care for populations based on their medication-specific markers.