projects by the databrew team

Monitoring the effectiveness of an indoor residual spraying for malaria in real-time

In 2016-17, we helped researchers in rural Mozambique roll out and evaluate a trial testing the efficacy of a house fumigation program for the prevention of malaria in children. This collaboration, still in progress, has lead to advances in algorithmic data cleaning, and insights into malaria risk and surveillance.

Training Mozambican medical researchers in data proficiency

In 2017, we carried out a 12-week training course for medical research professionals at the Centro de Investigação em Saude de Manhiça. Using the R statistical language, we taught data management, visualization, and analysis, so that our students - largely doctors - were equipped to glean insight from their data without having to waste funds on software and licensing fees.

Genomics and Big Data Modelling

Ben is involved in numerous projects through his work as a bioinformatician in genetics and genome biology at the Hospital for Sick Children (SickKids) in Toronto. Currently Ben is using germline methylation data to predict age of onset in patients with Li Fraumeni Syndrome, a rare hereditary cancer predisposition disorder. He is also currently using machine learning to predict tumor growth in patient derived xenograft models, testing the efficacy of lung cancer chemotherapy drugs.

Predicting hazardous waste violations in facilities regulated by the EPA

In 2015, as part of the Data Science for Social Good Fellowship, Ben’s team helped the US Environmental Protection Agency (EPA) identify hazardous material facilities at risk of major violations. They implemented several machine learning methods and used historical data to predict the risk of serious violations, helping the EPA prioritize their inspections of hazardous material facilities around the United States.

Estimating the effect of a school immunization program on student absenteeism

In 2014, we helped the Florida Department of Health in Alachua County quantify the reduction in absenteeism associated with a mass influenza immunization program. The results of our analysis were used to help school leaders and decision-makers better understand the costs and benefits of in-school immunizations.

Helping the Chicago Department of Public Health predict lead poisoning before it happens

In 2014, as part of his Data Science for Social Good Fellowship, Joe worked with the Chicago Department of Public Health and University of Chicago to build a statistical model for predicting which Chicago residents were most at risk of lead poisoning.

Academic research

We are active researchers in the academic world as well, with focuses on economics, epidemiology, biostatistics, and data science. Our peer-reviewed research ranges from the effectiveness of flu vaccination to tuberculosis trends, cancer diagnostics, and childhood obesity. For a full list of our publications, click here .