MACHINE LEARNING-BASED PREDICTION OF NO-SHOW TO TELEMEDICINE ENCOUNTERS IN PERU

Poster Number: P22
Presentation Time: 05:00 PM - 06:30 PM

Abstract Keywords: Telemedicine, Machine Learning, Global Health, Diversity, Equity, Inclusion, Accessibility, and Health Equity
Primary Track: Applications
Programmatic Theme: Clinical Informatics

This study evaluates the effectiveness of machine learning models in predicting no-shows to telemedicine appointments in Peru. Using data from the "Teleatiendo" platform, we explored models like XGBoost, and Random Forest. Results indicate that cost sensitive XGBoost balance well between identifying no-shows and shows, with the Balanced Random Forest model showing the highest recall. The findings highlight the potential of ML in improving telemedicine appointment adherence in resource-limited settings.

Speaker(s):
Christian Reategui Rivera, PhD student
University of Utah

Author(s):
Wanting Cui, Masters - University of Utah; Stefan Escobar Agreda, MD - Ministry oh Health (MINSA - Perú); Leonardo Rojas-Mezarina, MD - Telehealth Unit - Universidad Nacional Mayor de San Marcos; Joseph Finkelstein, MD, PhD - University of Utah;