Technologies to manage waiting lists in a health delivery network for low-income patients in Chile
In Chile, 73% of the population receive government-funded healthcare. This population is subject to waiting lists for surgeries and specialist consultation that are not managed by the government for many disease conditions. This has created health inequities in the timeliness of attention of patients with higher risk of disease deterioration. In 2016, 22,459 patients in waiting lists that were not managed by the government died before their first medical consultation, and 2,358 died before the surgery they were waiting. These numbers compare to the 993 deaths in the waiting lists of diseases managed by the government. To address these inequities, our overall objective is to develop a data driven methodology for prioritizing patients with diseases not regulated by the government.
Specimen sampling criteria for influenza emergencies
Influenza viruses create emergencies almost every year, and existing surveillance systems are key to address uncertainty in the disease detection, monitoring and control. An essential tool for monitoring influenza activity is the trend of cases confirmed with an influenza virus type. Confirmed case trends are useful to estimate when the season starts and which viruses are circulating, as each flu virus targets different age groups and produces symptoms of different types and severity. During influenza emergencies, specimens from infected cases massively arrive at the testing laboratories, which leads to chaos and increased operational costs as the labs increase their capacity in response to the overwhelming testing demand. Moreover, confirmed case data produced under this chaotic environment is being used intently by decision makers and the research community. I seek to understand how conventional and non-conventional methods for specimen sampling minimize the publication delay and uncertainty in the daily number of influenza confirmed cases.
Data driven monitoring of medical recommendations for breast cancer treatment
Medical recommendations for breast cancer healthcare may result in overtreatment or undertreatment. Overtreatment may occur when aggressive courses are prescribed to patients with low risk of cancer spreading, while undertreatment may be generated by patients' behaviors, as well as by social, economic, or racial disparities. My research team investigates techniques for the monitoring of over or undertreatment using data mining methods and traditional quality control philosophy.
Students and Collaborators
Collaborators
Alumni
- Bolanle Akinyele, MD, Cardiology Fellow at Johns Hopkins
- Sammy Sakaria, MD, Assistant Professor of Cardiology at Johns Hopkins
- Binu Koirala, PhD, Assistant Professor of Nursing at Johns Hopkins
- Rodrigo Martinez, MD, Manager of Clinical Processes and Hospital Operations at MINSAL (Chile)
- Diego Martinez, PhD, Assistant Professor of Emergency Medicine and Health Informatics at Johns Hopkins
- Gillian Stoltman, PhD, MPH, Associate Professor of Community Health at Western Michigan University
- Rajib Paul, PhD, Associate Professor of Public Health at UNC Charlotte
- Elise Dedoncker, PhD, Professor of Computer Science at Western Michigan University
- Richard A. Van Enk, PhD, Director of Infection Control at Bronson Methodist Hospital
Alumni
- Yuwen Gu, PhD in Industrial Engineering at Western Michigan University
- Milton Soto-Ferrari, PhD in Industrial Engineering at Western Michigan University
- Eric Meisheri, MS in Industrial Engineering at Western Michigan University
- Peter Holvenstot, MS in Industrial Engineering at Western Michigan University
- Greg Ostroy, BS in Computer Science at Western Michigan University