2016 Ontology Engineering Class Project: Comparing two Child Asthma Studies in support of the Child Health Exposure Analysis Repository

Printer-friendly version
Research Areas: Data Science, Health Informatics, Semantic eScience, Web Science
Principal Investigator: Deborah L. McGuinness
Co Investigator: Kristin Bennett
Concepts: Data Management, Ontology, Health informatics, Linked Data
Description:
Child Health Exposure Analysis Repository Data Science Semantics Meeting Notes

Rensselaer and Mount Sinai have created a unique partnership sponsored by the National Institutes of Health (NIH) to support groundbreaking work in the emergent field of “Exposomics”, which is the comprehensive study of environmental exposures in humans from conception through development. This newly formed program is called the Child Health Exposure Analysis Repository program, or CHEAR. As part of this effort, a team of Rensselear graduate students working in the 2016 Ontology Engineering Course with Profs. Deborah McGuinness and Elisa Kendall have created a new Ontology which encodes surveys related to childhood asthma to assist CHEAR researchers in gaining new insights for their work.

Epidemiology is the study and analysis of the patterns, causes, and effects of health and disease conditions in defined populations. Studies and surveys are one of the most popular ways of collecting data for any epidemiological project. Structuring this information becomes essential when we want to use information from current or past studies to help make decisions in the current project or about potential new studies in the future. For this purpose we need a way to encode, surveys using an ontology. We have defined an ontology for the representation of epidemiological study questionnaires which can be used to investigate how study results might meaningfully be combined. We have also developed an approach to encode questions in a survey enabling information from past surveys to be leveraged and reused in any new studies that may be conducted. This would greatly help in expediting the process of creating surveys and the epidemiologists can focus on evaluating the results from the survey to draw conclusions on the study being conducted.

The use case for our work is to utilize an ontology for comparing and contrasting two studies with the goal of determining how study results might meaningfully be combined. The first study examines relationships between different body size measurements and asthma in ethnic minority children while the other examined the association between prenatal and postnatal maternal stress and wheeze in 417 children enrolled in a prospective birth cohort in Mexico City see references. During these two studies, a collection of variables were created and used to address hypothesis proposed by epidemiologists . For instance, smoking in the home increases the risk of asthma in the study participant; or, increased body size is associated with increased risk of asthma in the study participant. The data for these variables were collected by means of survey and anthropology measurement. We encoded these variables with their related concepts into our ontology, such as body composition calculations, body measure, disease, symptom, instrument, question and response. After that, we answered a set of competency questions using our ontology.

Demo


View Static Demo

Demo

Call to Action

We welcome input from the semantic and epidemiology communities to help us improve our Ontology. Access our Github Repository or share your feedback and suggestions here: CHEAR Project Forum


Project Participants


Principal Investigator:

Deborah L. McGuinness

Dr. Deborah McGuinness is a leading expert in knowledge representation and reasoning languages and systems and has worked in ontology creation and evolution environments for over 20 years. Most recently, Deborah is best known for her leadership role in semantic web research [...]

Co Investigator:

Kristin Bennett

Kristin P. Bennett's research interests include: combining operations research and artificial intelligence problem solving methods; mathematical programming approaches to problems in data science, data mining, artificial intelligence and machine learning such as machine learning, support vector mach [...]

Collaborators:

Ian Gross

Ian Gross is a Senior Undergraduate Student at Rensselaer Polytechnic Institute, working towards a major in Computer Science and a minor in Information Technology and Web Science (ITWS). He joined Tetherless World Constellation in the summer of 2016, working on the CHEAR project.

Zhicheng Liang

PhD student at Rensselaer Polytechnic Institute

Yue Liu

Second year Masters Student in the IT department at Rensselaer Polytechnic Institute working on projects with the Tetherless World Constellation.

Anirudh Prabhu

Anirudh Prabhu is a Masters student in the ITWS department at Rensselaer Polytechnic Institute. His research interests include Semantic E-science, Data Visualization and Ontologies. He is currently doing his masters project under Prof. Peter Fox.

John Sheehan

John Sheehan is a Computer Science PhD student and graduate student researcher at the Tetherless World Constellation (TWC) at Rensselaer Polytechnic Institute. His research interests are in Healthcare Informatics and medical applications of semantic technologies. John is also an instructor and Com [...]