Expanding AI-Based Optical Urinalysis with Synthetic Data: A Large-Scale 849-Patient Study

Igor Balashov, Sofya Stroganova, Yuri Poimanov, Vahan Gevorgyan, Armen Charchyan, Andrey Grunin, Andrey A. Fedyanin

Keywords: Optical transmission spectroscopy, Urinalysis, Machine learning, Synthetic data augmentation, Data imbalance

TL;DR: Synthetic data augmentation improves machine-learning classification accuracy for optical spectroscopy-based urinalysis, especially when addressing data imbalance in rare clinical parameters.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 257