Alexandria M. Hertz, MD, presented “Microscopic Hematuria Evaluation Stratification: A Machine Learning Approach” for the Grand Rounds in Urology audience in February, 2020.
How to cite: Hertz, Alexandria M. “Microscopic Hematuria Evaluation Stratification: A Machine Learning Approach” February, 2020. Accessed Apr 2024. https://dev.grandroundsinurology.com/microscopic-hematuria-evaluation-stratification-a-machine-learning-approach/
Microscopic Hematuria Evaluation Stratification: A Machine Learning Approach – Summary:
Alexandria M. Hertz, MD, a resident physician and clinical teaching fellow at the Madigan Army Medical Center, discusses her team’s investigation of machine learning algorithms to improve risk evaluation for microscopic hematuria. She briefly reviews the current outdated approaches and their results, including as low as a 1% detection of malignancy with a 5-7% chance of UTIs, radiation exposure, or other negative outcomes. Dr. Hertz then outlines her retrospective review of 229 microscopic hematuria patients, which used a variety of classifications and algorithms to improve the timing and selection of patients for microscopic hematuria evaluation. The results of this study, specifically the criteria in the medium tree classifier, provide a promising first step in using machine learning to improve this process.
This post is the first in a curated series of lectures originally presented at the Society of Government Service Urologists 2020 Kimbrough Meeting in Charlotte, North Carolina. Other lectures in this series include:
