Presenter: Myrriah Jones − Biology
Faculty Mentor(s): Molly Jud, Edouard Hay
Session: (In-Person) Poster Presentation
Uroflowmetry measures data points like the max and average flow rate, volume, and duration of urination. Pediatric urologists use uroflowmetry to aid in diagnosing disorders of the urinary system like pediatric voiding dysfunction, a disorder that affects the sphincter control of the urethra.
Our purpose is to create a cost-effective tool for urologists to use to collect these data points more frequently and more accurately, in a more comfortable environment for patients. We used a combination of machine learning techniques and audio recordings of simulated urinations to train an algorithm to accurately predict the data points in people who urinate in a standing position. The data from the simulated urinations had similar trends in the data as the machine learning predictions and could reasonably work as a tool for urologists. By having a tool like this app, we can work towards increasing accessibility for necessary medical testing and improve both the accuracy and precision of uroflowmetry testing which helps provide better differential diagnoses and proper treatment to pediatric patients with similar symptoms yet distinct disorders.