Engineering students, faculty focus research to mitigate incontinence
Twenty-five percent of the United States suffers from urinary incontinence, and several Auburn University students and faculty are working to help make their lives better.
Industrial and systems engineering, or ISE, doctoral student Carlos Olivos, along with recent master’s graduate Khaled Khamis, have published their first paper on the subject in collaboration with Haneen Ali, director of the Health Services Administration Program, and ISE Assistant Professor Jia (Peter) Liu. The paper, “Mitigating Urinary Incontinence Condition Using Machine Learning,” was recently published in the BioMed Central medical journal and is the first study that proposes a model to mitigate urinary incontinence, or UI, based on predictive analytics. It is the first step in developing a machine-learning model to predict when a person will need to urinate.
“This is the first study that presents a machine-learning model to predict when people need to go to the bathroom,” explained Olivos. “Even though that may look trivial, it is not for people with urinary incontinence. Thus, this model could be embedded in a mobile app to assist people with urinary incontinence in reliable, prompted voiding.”
This framework can influence the ability of a UI patient to interact in social and professional situations and to avoid social embarrassment and psychological distress caused by their medical condition. In addition, a precise predictive instrument can enable health care providers and caregivers to assist people with various forms of UI in reliable, prompt voiding. The future goal of the research team is to turn the model into a smartphone application that can help individuals with UI conditions.
Olivos said he is proud to be a part of this research because the goal goes beyond monetary benefit.
“In this area, our goal is to improve people’s lives,” he said. “Second, this research included real data, which is challenging in our field. It is not always the case where you can collect real data and apply it in our models.”
The published paper is available here.
Submitted by: Carla Nelson