Explainable Artificial Intelligence applied to the Classification of Severity of Sleep Apnea through Heart Rate Variability

• Grant number: 24/04927-6
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): July 01, 2024
• Effective date (End): June 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Renato Tinós
• Grantee: Gabriel Branco Vitorino
• Host Institution: Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto (FFCLRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil

Abstract

We are interested in Explainable Artificial Intelligence to explain decisions of Machine Learning (ML) models in the problem of Sleep Apnea severity classification. (Obstructive) Sleep Apnea is one of the most common sleep disorders, consisting of the cessation or reduction of airflow during sleep. The most common and accurate test to diagnose sleep disorders is polysomnography which, despite its effectiveness, is a complex and relatively expensive test. In this project we will apply the LOcal Rule-based Explanations with fitness sharing (LOREfs) method to explain decisions of ML models in the problem of classifying the severity of Sleep Apnea through Heart Rate Variability. LOREfs will be able to explain decisions made by ML models that are efficient, but that generate decisions that are not interpretable by human experts. Understanding which attributes are important and how they are used by ML models for decision making is of great relevance for the understanding, by specialists in the field of Medicine, of how Heart Rate Variability is related to Sleep Apnea.