Matheus Araújo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Matheus Araújo has a total of 64 co-authors. Here are the top co-authors by number of shared publications:
Fabrício Benevenuto
Federal University of Minas Gerais, Belo Horizonte, Brazil
4 shared publications- Catherine M. Heinzinger
Sleep Disorders Center, Neurological Institute Cleveland Clinic Foundation, Cleveland, USA
3 shared publications - Fabrício Benevenuto
Federal University of Minas Gerais, Belo Horizonte, Brazil
3 shared publications - Nancy Foldvary-Schaefer
Cleveland Clinic, OH
3 shared publications Pollyanna Gonçalves
Computer Science Department, Federal University of Minas Gerais, Belo Horizonte, Brazil
3 shared publicationsReena Mehra
University of Washington
3 shared publications
Latest Publications
- 22 September 2026
- Published by Elsevier in Epilepsy & Behavior
- Vol. 185
Performance characteristics of obstructive sleep apnea screening instruments in adults with epilepsy
- 22 September 2026
- Published by Elsevier in Epilepsy & Behavior
- Vol. 185
Performance characteristics of obstructive sleep apnea screening instruments in adults with epilepsy
- Open Access
- 3 August 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1) , 7603
A foundation model for sleep-based risk stratification and clinical outcomes
Clinical sleep studies capture multiple physiologic signals, yet interpretation is often reduced to single summary measures of limited prognostic value, such as the apnea–hypopnea index. We present a foundation model that learns rich representations of sleep physiology from more than 10,000 clinical sleep recordings linked to electronic medical records. Here we show that sleep physiology contains latent risk structure invisible to conventional metrics, identifying five patient risk groups with markedly different trajectories for mortality, cardiovascular, and neurological disease. The highest-risk group shows more than double the mortality risk of the lowest, whereas apnea–hypopnea index severity categories show limited predictive value. The framework generalizes to the independent Sleep Heart Health Study, distinguishing high- and low-risk patients despite lower-resolution data. We demonstrate that foundation models recover clinically meaningful risk information embedded in routine sleep recordings that conventional metrics systematically miss, providing a scalable path to precision sleep medicine.
- 3 August 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1) , 7603
Open AccessA foundation model for sleep-based risk stratification and clinical outcomes
Clinical sleep studies capture multiple physiologic signals, yet interpretation is often reduced to single summary measures of limited prognostic value, such as the apnea–hypopnea index. We present a foundation model that learns rich representations of sleep physiology from more than 10,000 clinical sleep recordings linked to electronic medical records. Here we show that sleep physiology contains latent risk structure invisible to conventional metrics, identifying five patient risk groups with markedly different trajectories for mortality, cardiovascular, and neurological disease. The highest-risk group shows more than double the mortality risk of the lowest, whereas apnea–hypopnea index severity categories show limited predictive value. The framework generalizes to the independent Sleep Heart Health Study, distinguishing high- and low-risk patients despite lower-resolution data. We demonstrate that foundation models recover clinically meaningful risk information embedded in routine sleep recordings that conventional metrics systematically miss, providing a scalable path to precision sleep medicine.
- 4 March 2025
- Published by Oxford University Press (OUP) in SLEEP
- Vol. 48 (6)
Polysomnographic sleep architectural disruption associated with atrial fibrillation development
Study Objectives: To examine the unclear, inconsistent role of sleep architectural disruption in atrial fibrillation (AF) development. Methods: Patients (age ≥ 18 years) who underwent in-laboratory sleep studies at Cleveland Clinic 2000–2015 were examined (follow-up: 7.8 ± 3.5 years). Primary predictors were arousal index and total sleep time. Secondary predictors included sleep efficiency, wakefulness after sleep onset, sleep and REM latency, and percentage of each sleep stage. Predictors were fit to Cox proportional hazard models predicting time from sleep study to AF by diagnosis code. Covariates included demographics, anthropometrics, tobacco use, sleepiness, apnea–hypopnea index, sleep apnea-specific hypoxic impact, cardiovascular risk factors and disease, mood disorders, medications, and positive airway pressure. Results: In our cohort (n = 27 232, age: 49.4 ± 14.5 years, 43.7% male, 73.9% white), 2077 (7.6%) developed incident AF. Arousal index was not associated with AF incidence. For every hour of decreased total sleep time, AF incidence increased 8% (HR = 1.08, 95% CI = 1.04 to 1.11). For every 10-unit decrease in sleep efficiency, AF incidence increased 6% (HR = 1.06, 95% CI = 1.04 to 1.09). For every hour of increased wakefulness after sleep onset, AF incidence increased 11% (HR = 1.11, 95% CI = 1.05 to 1.16). For every 10-unit increase in percent N1, AF incidence increased 6% (HR = 1.06, 95% CI = 1.01 to 1.10). Conclusions: Less sleep time and greater sleep disruption were associated with increased incident AF in this large clinical cohort. These results suggest that sleep macro-architecture can influence AF development. Mechanistic and prospective studies are needed to verify whether sleep disruption is a novel target for AF prevention.
- 4 March 2025
- Published by Oxford University Press (OUP) in SLEEP
- Vol. 48 (6)
Polysomnographic sleep architectural disruption associated with atrial fibrillation development
Study Objectives: To examine the unclear, inconsistent role of sleep architectural disruption in atrial fibrillation (AF) development. Methods: Patients (age ≥ 18 years) who underwent in-laboratory sleep studies at Cleveland Clinic 2000–2015 were examined (follow-up: 7.8 ± 3.5 years). Primary predictors were arousal index and total sleep time. Secondary predictors included sleep efficiency, wakefulness after sleep onset, sleep and REM latency, and percentage of each sleep stage. Predictors were fit to Cox proportional hazard models predicting time from sleep study to AF by diagnosis code. Covariates included demographics, anthropometrics, tobacco use, sleepiness, apnea–hypopnea index, sleep apnea-specific hypoxic impact, cardiovascular risk factors and disease, mood disorders, medications, and positive airway pressure. Results: In our cohort (n = 27 232, age: 49.4 ± 14.5 years, 43.7% male, 73.9% white), 2077 (7.6%) developed incident AF. Arousal index was not associated with AF incidence. For every hour of decreased total sleep time, AF incidence increased 8% (HR = 1.08, 95% CI = 1.04 to 1.11). For every 10-unit decrease in sleep efficiency, AF incidence increased 6% (HR = 1.06, 95% CI = 1.04 to 1.09). For every hour of increased wakefulness after sleep onset, AF incidence increased 11% (HR = 1.11, 95% CI = 1.05 to 1.16). For every 10-unit increase in percent N1, AF incidence increased 6% (HR = 1.06, 95% CI = 1.01 to 1.10). Conclusions: Less sleep time and greater sleep disruption were associated with increased incident AF in this large clinical cohort. These results suggest that sleep macro-architecture can influence AF development. Mechanistic and prospective studies are needed to verify whether sleep disruption is a novel target for AF prevention.
- Open Access
Status and Opportunities of Machine Learning Applications in Obstructive Sleep Apnea: A Narrative Review
Background: Obstructive sleep apnea (OSA) is a prevalent and potentially severe sleep disorder characterized by repeated interruptions in breathing during sleep. Machine learning models have been increasingly applied in various aspects of OSA research, including diagnosis, treatment optimization, and developing biomarkers for endotypes and disease mechanisms. Objective: This narrative review evaluates the application of machine learning in OSA research, focusing on model performance, dataset characteristics, demographic representation, and validation strategies. We aim to identify trends and gaps to guide future research and improve clinical decision-making that leverages machine learning. Methods: This narrative review examines data extracted from 254 scientific publications published in the PubMed database between January 2018 and March 2023. Studies were categorized by machine learning applications, models, tasks, validation metrics, data sources, and demographics. Results: Our analysis revealed that most machine learning applications focused on OSA classification and diagnosis, utilizing various data sources such as polysomnography, electrocardiogram data, and wearable devices. We also found that study cohorts were predominantly overweight males, with an underrepresentation of women, younger obese adults, individuals over 60 years old, and diverse racial groups. Many studies had small sample sizes and limited use of robust model validation. Conclusion: Our findings highlight the need for more inclusive research approaches, starting with adequate data collection in terms of sample size and bias mitigation for better generalizability of machine learning models in OSA research. Addressing these demographic gaps and methodological opportunities is critical for ensuring more robust and equitable applications of artificial intelligence in healthcare.
Open AccessStatus and Opportunities of Machine Learning Applications in Obstructive Sleep Apnea: A Narrative Review
Background: Obstructive sleep apnea (OSA) is a prevalent and potentially severe sleep disorder characterized by repeated interruptions in breathing during sleep. Machine learning models have been increasingly applied in various aspects of OSA research, including diagnosis, treatment optimization, and developing biomarkers for endotypes and disease mechanisms. Objective: This narrative review evaluates the application of machine learning in OSA research, focusing on model performance, dataset characteristics, demographic representation, and validation strategies. We aim to identify trends and gaps to guide future research and improve clinical decision-making that leverages machine learning. Methods: This narrative review examines data extracted from 254 scientific publications published in the PubMed database between January 2018 and March 2023. Studies were categorized by machine learning applications, models, tasks, validation metrics, data sources, and demographics. Results: Our analysis revealed that most machine learning applications focused on OSA classification and diagnosis, utilizing various data sources such as polysomnography, electrocardiogram data, and wearable devices. We also found that study cohorts were predominantly overweight males, with an underrepresentation of women, younger obese adults, individuals over 60 years old, and diverse racial groups. Many studies had small sample sizes and limited use of robust model validation. Conclusion: Our findings highlight the need for more inclusive research approaches, starting with adequate data collection in terms of sample size and bias mitigation for better generalizability of machine learning models in OSA research. Addressing these demographic gaps and methodological opportunities is critical for ensuring more robust and equitable applications of artificial intelligence in healthcare.
- Open Access
- 1 January 2025
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 28, 167-174
Status and opportunities of machine learning applications in obstructive sleep apnea: A narrative review
Background: Obstructive sleep apnea (OSA) is a prevalent and potentially severe sleep disorder characterized by repeated interruptions in breathing during sleep. Machine learning models have been increasingly applied in various aspects of OSA research, including diagnosis, treatment optimization, and developing biomarkers for endotypes and disease mechanisms. Objective: This narrative review evaluates the application of machine learning in OSA research, focusing on model performance, dataset characteristics, demographic representation, and validation strategies. We aim to identify trends and gaps to guide future research and improve clinical decision-making that leverages machine learning. Methods: This narrative review examines data extracted from 254 scientific publications published in the PubMed database between January 2018 and March 2023. Studies were categorized by machine learning applications, models, tasks, validation metrics, data sources, and demographics. Results: Our analysis revealed that most machine learning applications focused on OSA classification and diagnosis, utilizing various data sources such as polysomnography, electrocardiogram data, and wearable devices. We also found that study cohorts were predominantly overweight males, with an underrepresentation of women, younger obese adults, individuals over 60 years old, and diverse racial groups. Many studies had small sample sizes and limited use of robust model validation. Conclusion: Our findings highlight the need for more inclusive research approaches, starting with adequate data collection in terms of sample size and bias mitigation for better generalizability of machine learning models in OSA research. Addressing these demographic gaps and methodological opportunities is critical for ensuring more robust and equitable applications of artificial intelligence in healthcare.
- 1 January 2025
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 28, 167-174
Open AccessStatus and opportunities of machine learning applications in obstructive sleep apnea: A narrative review
Background: Obstructive sleep apnea (OSA) is a prevalent and potentially severe sleep disorder characterized by repeated interruptions in breathing during sleep. Machine learning models have been increasingly applied in various aspects of OSA research, including diagnosis, treatment optimization, and developing biomarkers for endotypes and disease mechanisms. Objective: This narrative review evaluates the application of machine learning in OSA research, focusing on model performance, dataset characteristics, demographic representation, and validation strategies. We aim to identify trends and gaps to guide future research and improve clinical decision-making that leverages machine learning. Methods: This narrative review examines data extracted from 254 scientific publications published in the PubMed database between January 2018 and March 2023. Studies were categorized by machine learning applications, models, tasks, validation metrics, data sources, and demographics. Results: Our analysis revealed that most machine learning applications focused on OSA classification and diagnosis, utilizing various data sources such as polysomnography, electrocardiogram data, and wearable devices. We also found that study cohorts were predominantly overweight males, with an underrepresentation of women, younger obese adults, individuals over 60 years old, and diverse racial groups. Many studies had small sample sizes and limited use of robust model validation. Conclusion: Our findings highlight the need for more inclusive research approaches, starting with adequate data collection in terms of sample size and bias mitigation for better generalizability of machine learning models in OSA research. Addressing these demographic gaps and methodological opportunities is critical for ensuring more robust and equitable applications of artificial intelligence in healthcare.
- Open Access
- 21 November 2023
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 12 (22) , e031462
Sleep‐Disordered Breathing, Hypoxia, and Pulmonary Physiologic Influences in Atrial Fibrillation
Background: We leverage a large clinical cohort to elucidate sleep‐disordered breathing and sleep‐related hypoxia in incident atrial fibrillation (AF) development given the yet unclear contributions of sleep‐related hypoxia and pulmonary physiology in sleep‐disordered breathing and AF. Methods and Results: Patients who underwent sleep studies at Cleveland Clinic January 2, 2000, to December 30, 2015, comprised this retrospective cohort. Cox proportional hazards models were used to examine apnea hypopnea index, percentage time oxygen saturation 2 , and 1947 (4.6%) developed AF over 5 years. A 10‐unit apnea hypopnea index increase was associated with 2% higher AF risk (hazard ratio [HR], 1.02 [95% CI, 1.00–1.03]). A 10‐unit increase in percentage time oxygen saturation <90% and 10‐unit decreases in mean and minimum oxygen saturation were associated with 6% (HR, 1.06 [95% CI, 1.04–1.08]), 30% (HR, 1.30 [95% CI, 1.18–1.42]), and 9% (HR, 1.09 [95% CI, 1.03–1.15]) higher AF risk, respectively. After adjustment for spirometry (n=9683 with available data), only hypoxia remained significantly associated with incident AF, although all coefficients were stable. Conclusions: Sleep‐related hypoxia was associated with incident AF in this clinical cohort, consistent across 3 measures of hypoxia, persistent after adjustment for pulmonary physiologic impairment. Findings identify a strong role for sleep‐related hypoxia in AF development without pulmonary physiologic interdependence.
- 21 November 2023
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 12 (22) , e031462
Open AccessSleep‐Disordered Breathing, Hypoxia, and Pulmonary Physiologic Influences in Atrial Fibrillation
Background: We leverage a large clinical cohort to elucidate sleep‐disordered breathing and sleep‐related hypoxia in incident atrial fibrillation (AF) development given the yet unclear contributions of sleep‐related hypoxia and pulmonary physiology in sleep‐disordered breathing and AF. Methods and Results: Patients who underwent sleep studies at Cleveland Clinic January 2, 2000, to December 30, 2015, comprised this retrospective cohort. Cox proportional hazards models were used to examine apnea hypopnea index, percentage time oxygen saturation 2 , and 1947 (4.6%) developed AF over 5 years. A 10‐unit apnea hypopnea index increase was associated with 2% higher AF risk (hazard ratio [HR], 1.02 [95% CI, 1.00–1.03]). A 10‐unit increase in percentage time oxygen saturation <90% and 10‐unit decreases in mean and minimum oxygen saturation were associated with 6% (HR, 1.06 [95% CI, 1.04–1.08]), 30% (HR, 1.30 [95% CI, 1.18–1.42]), and 9% (HR, 1.09 [95% CI, 1.03–1.15]) higher AF risk, respectively. After adjustment for spirometry (n=9683 with available data), only hypoxia remained significantly associated with incident AF, although all coefficients were stable. Conclusions: Sleep‐related hypoxia was associated with incident AF in this clinical cohort, consistent across 3 measures of hypoxia, persistent after adjustment for pulmonary physiologic impairment. Findings identify a strong role for sleep‐related hypoxia in AF development without pulmonary physiologic interdependence.
- Open Access
- 1 October 2023
- Published by Oxford University Press (OUP) in American Journal of Respiratory and Critical Care Medicine
- Vol. 208 (7) , 752-754
The Influence of Physiologic Burdens Related to Obstructive Sleep Apnea on Cardiovascular Outcomes
- 1 October 2023
- Published by Oxford University Press (OUP) in American Journal of Respiratory and Critical Care Medicine
- Vol. 208 (7) , 752-754
Open AccessThe Influence of Physiologic Burdens Related to Obstructive Sleep Apnea on Cardiovascular Outcomes
- Open Access
- 18 February 2020
- Published by Revista Engenharia na Agricultura in REVISTA ENGENHARIA NA AGRICULTURA - REVENG
- Vol. 28, 129-139
AUTOMATIC SYSTEM FOR PSYCHROMETRIC CALCULATIONS
Air psychrometric properties are important in several areas of agricultural engineering, such as calculation of evaluation of the animal environment and air control in grain storage units. Due to their relevance and the complexity and the uncertainties in the use of psychrometric charts, the objective of this paper was to develop the software, PsyCalculator, which stands out for the automated function that allows presenting the data acquired through sensors and the respective psychrometric properties of the air, either in graph or tables. The automatic data acquisition system consists of a microcontroller that performs readings of sensors that measure the dry bulb temperature and another variable, which may be: wet bulb temperature, relative air humidity or dew point temperature. Then, the values are sent to the PsyCalculator software system. In addition to the values of dry bulb temperature, the software system presents wet bulb temperature and dew point temperature, values of saturation vapor pressure, vapor pressure, mixing ratio, specific volume, enthalpy and degree of saturation. Among the psychometric variables analyzed, the software developed in the study presented a maximum error of 2.14% for the calculation of dew point temperature. The developed software allows the automation of systems that depend on the psychrometric parameters in a friendly and precise fashion.
- 18 February 2020
- Published by Revista Engenharia na Agricultura in REVISTA ENGENHARIA NA AGRICULTURA - REVENG
- Vol. 28, 129-139
Open AccessAUTOMATIC SYSTEM FOR PSYCHROMETRIC CALCULATIONS
Air psychrometric properties are important in several areas of agricultural engineering, such as calculation of evaluation of the animal environment and air control in grain storage units. Due to their relevance and the complexity and the uncertainties in the use of psychrometric charts, the objective of this paper was to develop the software, PsyCalculator, which stands out for the automated function that allows presenting the data acquired through sensors and the respective psychrometric properties of the air, either in graph or tables. The automatic data acquisition system consists of a microcontroller that performs readings of sensors that measure the dry bulb temperature and another variable, which may be: wet bulb temperature, relative air humidity or dew point temperature. Then, the values are sent to the PsyCalculator software system. In addition to the values of dry bulb temperature, the software system presents wet bulb temperature and dew point temperature, values of saturation vapor pressure, vapor pressure, mixing ratio, specific volume, enthalpy and degree of saturation. Among the psychometric variables analyzed, the software developed in the study presented a maximum error of 2.14% for the calculation of dew point temperature. The developed software allows the automation of systems that depend on the psychrometric parameters in a friendly and precise fashion.
- Top Cited Papers
- 1 February 2020
- Published by Elsevier in Information Sciences
- Vol. 512, 1078-1102
A comparative study of machine translation for multilingual sentence-level sentiment analysis
- 1 February 2020
- Published by Elsevier in Information Sciences
- Vol. 512, 1078-1102
Top Cited PapersA comparative study of machine translation for multilingual sentence-level sentiment analysis
- Open Access
- 30 October 2018
- Published by Association for Computing Machinery (ACM)
- p. 45-48
Visualizing Geo-Demographic Urban Data
- 30 October 2018
- Published by Association for Computing Machinery (ACM)
- p. 45-48
Open AccessVisualizing Geo-Demographic Urban Data