Data Integration and Predictive Analysis System (IPAS)

KBSI proposes to design and develop a Data Integration and Predictive Analysis System (IPAS) for the prediction of incidents of human infectious diseases. IPAS will utilize innovative collection of data from open data sources, veterinary and medical professionals, and public observations, together with data cleaning, harmonization, spatio-temporal pattern extraction, factor analysis, and predictive models to provide comprehensive disease incidence prediction. The project will collect and integrate a comprehensive dataset of previous disease occurrences and potential influencing factors like environmental conditions, regional health status and practices, demographics, ethnicity and cultural practices, veterinary and zoonotic indicators, and vector prevalence. Natural language processing (NLP) will be used to extract disease, syndromic, and zoonotic details from news feeds, public health reports, and medical publications. A smartphone app will be used to collect data from situated public, veterinary, and health officials on veterinary, zoonotic, and human signs and symptoms, and on the prevalence of vectors. Machine learning and predictive-analytics-based models will be developed to predict the probability of occurrence of disease for specific geographical locations and times. Phase I will focus on a select CDC Category A disease.