Doctoral Dissertations

Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Data Science and Engineering

Major Professor

Arash Shaban-Nejad

Committee Members

Russell Zaretzki, Kristina Kintziger, Haileab Hilafu


Chronic and infectious diseases have a profound impact on the quality and length of life of populations that suffer from these conditions. Scientists, physicians, and health officials are seeking innovative approaches to decrease the morbidity and mortality of deadly diseases. Incorporating artificial intelligence and data science techniques across the health science domain could improve disease surveillance, intervention planning, and policymaking. In this dissertation, we describe the design and development of the Urban Population Health Observatory (UPHO), an explainable knowledge-based multimodal big data analytics platform. A common challenge for conducting multimodal big data analytics is integrating multidimensional heterogeneous data sources, which often exist in different levels of granularity. These sources range from individual-level to group-level data, as well as structured and unstructured data sources. The UPHO platform is designed to foster the integration of multimodal big data including social determinants of health, observation of daily living, and population health data consistently across jurisdictions to estimate the incidence and prevalence of different health conditions, as well as related risk factors. Leveraging semantics, the UPHO platform provides contextual knowledge from relevant domains of interest by reusing several ontologies focusing on clinical and public health (diseases, geography, etc.). UPHO encompasses an Explainable Artificial Intelligence (XAI) and Interpretability feature which increases trust and trustworthiness, justifies actions and decisions, improves usability, aids in locating sources of error, and can minimize the chance for human error. Developing a multimodal scalable surveillance system to monitor and detect trends as well as deliver rapid early warnings and recommendations could assist health officials, physicians, and researchers in responding, and mitigating a public health crisis.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."