To access the session recording, navigate to the content tab and click the view video button. 

Session Schedule

Find a specific presentation in the session by navigating to the timestamp indicated below.

0:00:00
Introduction

0:01:08
When sleep apnea speaks: Voice as a biomarker of sleep apnea
Azadeh Yadollahi (Canada)

0:22:18
AI analysis of EEG patterns in polysomnography- An insight into the brain
Haoqi Sun (United States)

0:47:57
AI-based endo-phenotyping of sleep apnea
Lucia Pinilla (Australia)

1:11:24
Question and answer

Summary
We are living in an age when artificial intelligence (AI) is transforming sleep medicine by enabling advanced diagnostics, personalized treatments, and improved health outcomes (Topol, 2019). AI-powered wearable devices facilitate analysis of continuous sleep monitoring data in real-world settings, offering insights into sleep patterns and behaviors (Ko et al., 2021). Similarly, AI algorithms applied to EEG data streamline the detection and analysis of sleep stages, disorders, and microarousals with high accuracy, reducing reliance on labor-intensive manual scoring (Sun et al., 2020). In sleep apnea, AI may enhance measurement precision by integrating polysomnographic data with predictive models to assess severity and potential cardiovascular risks (Kapur et al., 2022). Moreover, AI enables the development of novel metrics beyond traditional indices, such as the apnea-hypopnea index, for improved risk stratification and treatment efficacy prediction. This integration of AI can also reduce diagnostic costs and time, addressing accessibility challenges in underserved populations. By harnessing AI technologies, sleep medicine can advance toward a future of more individualized, efficient, and impactful care for patients with sleep disorders. In this symposium, we plan to discuss applications of AI in sleep medicine, with particular focus on wearable devices, EEG analysis, sleep disordered breathing, and risk prediction.
Why is this symposium needed now?
The rapidly evolving field of AI presents a unique opportunity to address long-standing challenges in sleep medicine, making this symposium both timely and necessary. Advances in wearable technology and machine learning have enabled continuous, real-world sleep monitoring addressing barriers in traditional clinical settings (Walch et al., 2019; Goldstein et al. 2020). Additionally, AI-driven methods for analyzing polysomnography and EEG data are transforming diagnostic accuracy and risk prediction. For example, these methods have proven to be useful for predicting risk of Alzheimer Disease (Sun et al. 2025; Sun et al. 2024). With the increasing prevalence of conditions such as obstructive sleep apnea and their link to cardiovascular disease and other comorbidities, there is increasing demand for innovative, efficient, and scalable endophenotyping tools (Dutta et al. 2021; Parekh et al. 2021; Pinilla et al. 2023; Hajipour et al. 2023; Azarbarzin et al. 2024). We have assembled a global panel of experts to review the current research landscape, and discuss these issues. Dr. Goldstein is a recognized leader in the field of digital sleep medicine and AI using wearable devices. Dr. Sun has greatly contributed to novel computational approaches (including AI) for EEG analyses. Dr. Pinilla has published high impact papers to better phenotype and risk-stratify individuals with OSA. The chairs (Dr. Najib Ayas from Canada and Dr. Ali Azarbarzin from USA) are recognized global experts in obstructive sleep apnea; however, each provides different perspectives as Dr. Ayas is a practicing sleep physician/scientist, and Dr. Azarbarzin is a biomedical engineer/physiologist with experience in advanced signal processing and machine learning.
Diversity: The chairs/speakers are also diverse with respect to gender/ethnicity/career level (2/5 self-identify as women, 3/5 as a visible minority), and represent multiple continents and countries.