Search

This article presents a system for predicting a human's risk of contracting infectious diseases based on their personal attributes and environments (region, specific location features and climate contexts). This system is also intended to help human experts in the domain (i.e. epidemiologists) to represent their knowledge and ease their jobs related to personalized infectious disease risk prediction. The system consists of a knowledge representation to encode epidemiological knowledge about infectious disease risk, and an algorithm that auto-converts the encoded knowledge into a model that predicts the risk as a probability. The knowledge representation, Infectious Disease Risk (IDR), consists of an ontology and rules to represent the knowledge structure and its quantification in a way that allows auto-generation to a prediction model, Bayesian Network (BN). The algorithm, BN-Builder, converts the IDR knowledge-base to an infectious disease risk BN, including populating the basis of predictive reasoning from the IDR rules. A user interface facilitates encoding of epidemiological knowledge into the IDR knowledge-base. The system's output, personalized infectious disease risk prediction, is validated for three disease-country contexts: Dengue Fever and Tuberculosis in Indonesia, and Cholera in India. The personalized infectious disease risks are reliable (p values > 0.05) for each population parameter. The personalized infectious disease risk probability can be reliably predicted using this system. Inclusion of more granularity on contexts in this domain will be considered in further development of this system.

Infectious diseases are a major cause of human morbidity, but most are avoidable. An accurate and personalized risk prediction is expected to alert people to the risk of getting exposed to infectious diseases. However, as data and knowledge in the epidemiology and infectious diseases field becomes available, an updateable risk prediction model is needed. The objectives of this article are (1) to describe the mechanisms for generating a Bayesian Network (BN), as risk prediction model, from a knowledge-base, and (2) to examine the accuracy of the prediction result. The research in this paper started by encoding declarative knowledge from the Atlas of Human Infectious Diseases into an Infectious Disease Risk Ontology. Automatic generation of a BN from this knowledge uses two tools (1) a Rule Converter generates a BN structure from the ontology (2) a Joint & Marginal Probability Supplier tool populates the BN with probabilities. These tools allow the BN to be recreated automatically whenever knowledge and data changes. In a runtime phase, a third tool, the Context Collector, captures facts given by the client and consequent environmental context. This paper introduces these tools and evaluates the effectiveness of the resulting BN for a single infectious disease, Anthrax. We have compared conditional probabilities predicted by our BN against incidence estimated from real patient visit records. Experiments explored the role of different context data in prediction accuracy. The results suggest that building a BN from an ontology is feasible. The experiments also show that more context results in better risk prediction.