Strengths, weaknesses, opportunities and threats of using AI-enabled technology in sleep medicine: a commentary.

Over the past few years, artificial intelligence (AI) has emerged as a powerful tool used to efficiently automate several tasks across multiple domains. Sleep medicine is perfectly positioned to leverage this tool due to the wealth of physiological signals obtained through sleep studies or sleep tracking devices and abundance of accessible clinical data through electronic medical records. However, caution must be applied when utilizing AI, due to intrinsic challenges associated with novel technology. The Artificial Intelligence in Sleep Medicine committee of the American Academy of Sleep Medicine (AASM) reviews advancements in AI within the sleep medicine field. In this article, the Artificial Intelligence in Sleep Medicine committee members provide a commentary on the scope of AI technology in sleep medicine. The commentary identifies three pivotal areas in sleep medicine which can benefit from AI technologies: clinical care, lifestyle management and population health management. This article provides a detailed analysis of the strengths, weaknesses, opportunities, and threats associated with using AI enabled technologies in each pivotal area. Finally, the article broadly reviews barriers and challenges associated with using AI enabled technologies and offers possible solutions.

Endotypic traits of supine position and supine-predominant obstructive sleep apnoea in Asian patients.

BACKGROUND: Over half of all cases of obstructive sleep apnoea (OSA) are classified as supine-related OSA; however, the pathological endotype during supine position is not fully understood. This study aims to investigate the endotypic traits of supine-predominant OSA and explore the variations in endotypic traits between the supine and lateral positions. METHODS: We prospectively recruited 689 adult patients with OSA from a single sleep centre between April 2020 and December 2022. Endotypic traits, namely arousal threshold, collapsibility, loop gain and upper airway muscle compensation, were retrieved from polysomnographic signals. We identified spOSA by a supine to non-supine apnoea-hypopnoea index (AHI) ratio >2. We cross-sectionally compared demographic and endotypic traits between supine-predominant OSA and non-positional OSA and examined the associations between supine-predominant OSA and endotypic traits. Additionally, we compared the changes in endotypic traits between supine and lateral positions in patients with supine-predominant OSA and non-positional OSA. RESULTS: In our study sample, 75.8% of patients were identified as having supine-predominant OSA. Compared to non-positional OSA, supine-predominant OSA was associated with low collapsibility (ß= -3.46 %eupnoea, 95% CI -5.93- -1.00 %eupnoea) and reduced compensation (ß= -6.79 %eupnoea, 95% CI -10.60- -2.99 %eupnoea). When transitioning from the lateral to supine position, patients with supine-predominant OSA had a substantial decrease in compensation compared to those with non-positional OSA (-11.98 versus -6.28 %eupnoea). CONCLUSIONS: Supine-predominant OSA is the prevalent phenotype of OSA in Asian patients. Inadequate upper airway compensation appears to be a crucial underlying pathology in patients with supine-predominant OSA.

Review and perspective on sleep-disordered breathing research and translation to clinics.

Sleep-disordered breathing, ranging from habitual snoring to severe obstructive sleep apnea, is a prevalent public health issue. Despite rising interest in sleep and awareness of sleep disorders, sleep research and diagnostic practices still rely on outdated metrics and laborious methods reducing the diagnostic capacity and preventing timely diagnosis and treatment. Consequently, a significant portion of individuals affected by sleep-disordered breathing remain undiagnosed or are misdiagnosed. Taking advantage of state-of-the-art scientific, technological, and computational advances could be an effective way to optimize the diagnostic and treatment pathways. We discuss state-of-the-art multidisciplinary research, review the shortcomings in the current practices of SDB diagnosis and management in adult populations, and provide possible future directions. We critically review the opportunities for modern data analysis methods and machine learning to combine multimodal information, provide a perspective on the pitfalls of big data analysis, and discuss approaches for developing analysis strategies that overcome current limitations. We argue that large-scale and multidisciplinary collaborative efforts based on clinical, scientific, and technical knowledge and rigorous clinical validation and implementation of the outcomes in practice are needed to move the research of sleep-disordered breathing forward, thus increasing the quality of diagnostics and treatment.

Continuous positive airway pressure and adherence in patients with different endotypes of obstructive sleep apnea.

Determining the endotypes of obstructive sleep apnea (OSA) has potential implications for precision interventions. Here we assessed whether continuous positive airway pressure (CPAP) treatment outcomes differ across endotypic subgroups. We conducted a retrospective analysis of data obtained from 225 patients with moderate-to-severe OSA from a single sleep centre. Polysomnographic and CPAP titration study data were collected between May 2020 and January 2022. One-month CPAP treatment adherence was followed. Obstructive sleep apnea endotypes, namely arousal threshold, collapsibility, loop gain, and upper airway gain were estimated from polysomnography and dichotomised as high versus low. We examined associations between endotypic subgroups and (1) optimal CPAP titration pressure, (2) CPAP-related improvements in sleep architecture (proportions of slow-wave and rapid eye movement (REM) sleep), and (3) CPAP adherence. We observed that patients with high collapsibility required a higher CPAP pressure than those with low collapsibility (∆ = 0.4 cmH2 O, 95% confidence interval [CI] = 0.3-1.7). A larger increase in slow-wave sleep and in REM sleep proportions after CPAP treatment were observed in patients with a high arousal threshold, high collapsibility, high loop gain, or high upper airway gain than in those with low levels of endotypes. High loop gain and high collapsibility were independently associated with longer CPAP use hours per night (∆ = 0.6 h, 95% CI = 0.2-1.5 and ∆ = 0.3 h, 95% CI = 0.03-1.5, respectively). In conclusion, different endotypic subgroups of OSA exhibit a difference in outcomes of CPAP treatment. Knowledge of endotypes may help clinicians to understand which patients are expected to benefit most from CPAP therapy prior to its administration.

Relationship between Symptom Profiles and Endotypes among Patients with Obstructive Sleep Apnea: A Latent Class Analysis.

Rationale: Obstructive sleep apnea (OSA) is a heterogeneous syndrome with various endotypic traits and symptoms. A link among symptoms, endotypes, and disease prognosis has been proposed but remains unsupported by empirical data. Objectives: To link symptom profiles and endotypes by clustering endotypic traits estimated using polysomnographic signals. Methods: We recruited 509 patients with moderate to severe OSA from a single sleep center. Polysomnographic data were collected between May 2020 and January 2022. Endotypic traits, namely arousal threshold, upper airway collapsibility, loop gain, and upper airway muscle compensation, were retrieved using polysomnographic signals during non-rapid eye movement periods. We used latent class analysis to group participants into endotype clusters. Demographic and polysomnographic parameter differences were compared between clusters, and associations between endotype clusters and symptom profiles were examined using logistic regression analyses. Results: Three endotype clusters were identified, characterized by high collapsibility/loop gain, low arousal threshold, and low compensation, respectively. Patients in each cluster exhibited similar demographic characteristics, but those in the high collapsibility/loop gain cluster had the highest proportion of obesity and severe oxygen desaturation observed in polysomnographic studies. The low compensation cluster was characterized by fewer sleepy symptoms and exhibited a lower rate of diabetes mellitus. Compared with the excessively sleepy group, disturbed sleep symptoms were associated with the low arousal threshold cluster (odds ratio, 1.89; 95% confidence interval, 1.16-3.10). Excessively sleepy symptoms were associated with the high collapsibility/loop gain cluster (odds ratio, 2.16; 95% confidence interval, 1.39-3.37) compared with the minimally symptomatic group. Conclusions: Three pathological endotype clusters were identified among patients with moderate to severe OSA, each exhibiting distinct polysomnographic characteristics and clinical symptom profiles.

The effect of hyperoxia on ventilation during recovery from general anesthesia: A randomized pilot study for a parallel randomized controlled trial.

STUDY OBJECTIVE: While supplemental O2 inhalation corrects hypoxemia, its effect on post-anesthesia ventilation remains unknown. This pilot trial tested the hypothesis that hyperoxia increases the time spent with a transcutaneous PCO2 (TcPCO2) > 45 mmHg, compared with standard O2 supplementation. DESIGN: Single-blinded, parallel two-arm randomized pilot trial. SETTING: University hospital. PATIENTS: 20 patients undergoing robotic-assisted laparoscopic nephrectomy. MEASUREMENTS: After institutional approval and informed consent, patients were randomized to receive O2 titrated to arterial saturation (SpO2): 90-94% (Conservative O2, N =10), or to SpO2 > 96% (Liberal O2, N = 10) for up to 90 min after anesthesia. Continuous TcPCO2, respiratory inductance plethysmography (RIP), and SpO2, were recorded. We calculated the percentage of time at TcPCO2 > 45 mmHg for each patient and compared the two groups using analysis of covariance, adjusting for sex, age, and body mass index. We also estimated the sample size required to detect the between-group difference observed in this pilot trial. RIP signals were used to calculate apnea/hypopnea index (AHI), which was then compared between two groups. MAIN RESULTS: The mean percentage of time with a TcPCO2 > 45 mmHg was 80.6% for the Conservative O2 (N=9) and 61.2% for the Liberal O2 (N=10) group [between-group difference of 19.4% (95% CI: -18.7% to 57.6%), P = 0.140]. With an observed effect size of 0.73, we estimated that 30 participants per group are required, to demonstrate this difference with a power of 80% at a two-sided alpha of 5%. Means SpO2 were 94.5% and 99.9% for the Conservative O2 and the Liberal O2 groups, respectively. AHI was significantly higher in the Conservative O2, compared with the Liberal O2 group (median AHI: 16 vs. 3; P = 0.0014). CONCLUSIONS: Hyperoxia in the post-anesthesia period reduced the time spent at TcPCO2 > 45 mmHg and significantly decreased AHI, while mean SpO2 ranged inside the a priori defined limits. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT04723433.

The Sleep Revolution project: the concept and objectives.

Obstructive sleep apnea is linked to severe health consequences such as hypertension, daytime sleepiness, and cardiovascular disease. Nearly a billion people are estimated to have obstructive sleep apnea with a substantial economic burden. However, the current diagnostic parameter of obstructive sleep apnea, the apnea-hypopnea index, correlates poorly with related comorbidities and symptoms. Obstructive sleep apnea severity is measured by counting respiratory events, while other physiologically relevant consequences are ignored. Furthermore, as the clinical methods for analysing polysomnographic signals are outdated, laborious, and expensive, most patients with obstructive sleep apnea remain undiagnosed. Therefore, more personalised diagnostic approaches are urgently needed. The Sleep Revolution, funded by the European Union's Horizon 2020 Research and Innovation Programme, aims to tackle these shortcomings by developing machine learning tools to better estimate obstructive sleep apnea severity and phenotypes. This allows for improved personalised treatment options, including increased patient participation. Also, implementing these tools will alleviate the costs and increase the availability of sleep studies by decreasing manual scoring labour. Finally, the project aims to design a digital platform that functions as a bridge between researchers, patients, and clinicians, with an electronic sleep diary, objective cognitive tests, and questionnaires in a mobile application. These ambitious goals will be achieved through extensive collaboration between 39 centres, including expertise from sleep medicine, computer science, and industry and by utilising tens of thousands of retrospectively and prospectively collected sleep recordings. With the commitment of the European Sleep Research Society and Assembly of National Sleep Societies, the Sleep Revolution has the unique possibility to create new standardised guidelines for sleep medicine.

Methods for home-based self-applied polysomnography: the Multicenter AIDS Cohort Study.

Study Objectives: Along with multiple chronic comorbidities, sleep disorders are prevalent in people living with human immunodeficiency virus (HIV) infection. The goal of this study was to establish methods for assessing sleep quality and breathing-related disorders using self-applied home polysomnography in people with and without HIV. Methods: Self-applied polysomnography was conducted on 960 participants in the Multicenter AIDS Cohort Study (MACS) using the Nox A1 recorder to collect data on the frontal electroencephalogram (EEG), bilateral electrooculograms, and a frontalis electromyogram during sleep. Breathing patterns were characterized using respiratory inductance plethysmography bands and pulse oximetry. Continuous recordings of the electrocardiogram were also obtained. All studies were scored centrally for sleep stages and disordered breathing events. Results: Successful home polysomnography was obtained in 807 of 960 participants on the first attempt and 44 participants on the second. Thus, a successful polysomnogram was obtained in 851 (88.6%) of the participants. Reasons for an unsuccessful study included less than 3 h of data on oximetry (34.6%), EEG (28.4%), respiratory inductance plethysmography (21.0%), or two or more of these combined (16.0%). Of the successful studies (N = 851), signal quality was rated as good, very good, or excellent in 810 (95.2%). No temporal trends in study quality were noted. Independent correlates of an unsuccessful study included black race, current smoking, and cocaine use. Conclusions: Home polysomnography was successfully completed in the MACS demonstrating its feasibility in a community cohort. Given the burden of in-lab polysomnography, the methods described herein provide a cost-effective alternative for collecting sleep data in the home.

Proof of principle study: diagnostic accuracy of a novel algorithm for the estimation of sleep stages and disease severity in patients with sleep-disordered breathing based on actigraphy and respiratory inductance plethysmography.

PURPOSE: In this proof of principle study, we evaluated the diagnostic accuracy of the novel Nox BodySleepTM 1.0 algorithm (Nox Medical, Iceland) for the estimation of disease severity and sleep stages based on features extracted from actigraphy and respiratory inductance plethysmography (RIP) belts. Validation was performed against in-lab polysomnography (PSG) in patients with sleep-disordered breathing (SDB). METHODS: Patients received PSG according to AASM. Sleep stages were manually scored using the AASM criteria and the recording was evaluated by the novel algorithm. The results were analyzed by descriptive statistics methods (IBM SPSS Statistics 25.0). RESULTS: We found a strong Pearson correlation (r=0.91) with a bias of 0.2/h for AHI estimation as well as a good correlation (r=0.81) and an overestimation of 14 min for total sleep time (TST). Sleep efficiency (SE) was also valued with a good Pearson correlation (r=0.73) and an overestimation of 2.1%. Wake epochs were estimated with a sensitivity of 0.65 and a specificity of 0.59 while REM and non-REM (NREM) phases were evaluated a sensitivity of 0.72 and 0.74, respectively. Specificity was 0.74 for NREM and 0.68 for REM. Additionally, a Cohen's kappa of 0.62 was found for this 3-class classification problem. CONCLUSION: The algorithm shows a moderate diagnostic accuracy for the estimation of sleep. In addition, the algorithm determines the AHI with good agreement with the manual scoring and it shows good diagnostic accuracy in estimating wake-sleep transition. The presented algorithm seems to be an appropriate tool to increase the diagnostic accuracy of portable monitoring. The validated diagnostic algorithm promises a more appropriate and cost-effective method if integrated in out-of-center (OOC) testing of patients with suspicion for SDB.

The design of RIP belts impacts the reliability and quality of the measured respiratory signals.

PURPOSE: Evaluate the effect of respiratory inductance plethysmography (RIP) belt design on the reliability and quality of respiratory signals. A comparison of cannula flow to disposable cut-to-fit, semi-disposable folding and disposable RIP belts was performed in clinical home sleep apnea testing (HSAT) studies. METHODS: This was a retrospective study using clinical HSAT studies. The signal reliability of cannula, thorax, and abdomen RIP belts was determined by automatically identifying periods during which the signals did not represent respiratory airflow and breathing movements. Results were verified by manual scoring. RIP flow quality was determined by examining the correlation between the RIP flow and cannula flow when both signals were considered reliable. RESULTS: Of 767 clinical HSAT studies, mean signal reliability of the cut-to-fit, semi-disposable, and disposable thorax RIP belts was 83.0 ± 26.2%, 76.1 ± 24.4%, and 98.5 ± 9.3%, respectively. The signal reliability of the cannula was 92.5 ± 16.1%, 87.0 ± 23.3%, and 85.5 ± 24.5%, respectively. The automatic assessment of signal reliability for the RIP belts and cannula flow had a sensitivity of 50% and a specificity of 99% compared with manual assessment. The mean correlation of cannula flow to RIP flow from the cut-to-fit, semi-disposable, and disposable RIP belts was 0.79 ± 0.24, 0.52 ± 0.20, and 0.86 ± 0.18, respectively. CONCLUSION: The design of RIP belts affects the reliability and quality of respiratory signals. The disposable RIP belts that had integrated contacts and did not fold on top of themselves performed the best. The cut-to-fit RIP belts were most likely to be unreliable, and the semi-disposable folding belts produced the lowest-quality RIP flow signals compared to the cannula flow signal.

A scalable method of determining physiological endotypes of sleep apnea from a polysomnographic sleep study.

Sleep apnea is caused by several endophenotypic traits, namely pharyngeal collapsibility, poor muscle compensation, ventilatory instability (high loop gain), and arousability from sleep (low arousal threshold). Measures of these traits have shown promise for predicting outcomes of therapies (e.g. oral appliances, surgery, hypoglossal nerve stimulation, CPAP, and pharmaceuticals), which may become an integral part of precision sleep medicine. Currently, the methods Sands et al. developed for endotyping sleep apnea from polysomnography (PSG) are embedded in the original authors' code, which is computationally expensive and requires technological expertise to run. We present a reimplementation and validation of the integrity of the original authors' code by reproducing the endo-Phenotyping Using Polysomnography (PUP) method of Sands et al. The original MATLAB methods were reprogrammed in Python; efficient algorithms were developed to detect breaths, calculate normalized ventilation (moving time-average), and model ventilatory drive (intended ventilation). The new implementation (PUPpy) was validated by comparing the endotypes from PUPpy with the original PUP results. Both endotyping methods were applied to 38 manually scored polysomnographic studies. Results of the new implementation were strongly correlated with the original (p < 10-6 for all): ventilation at eupnea V̇ passive (ICC = 0.97), ventilation at arousal onset V̇ active (ICC = 0.97), loop gain (ICC = 0.96), and arousal threshold (ICC = 0.90). We successfully implemented the original PUP method by Sands et al. providing further evidence of its integrity. Additionally, we created a cloud-based version for scaling up sleep apnea endotyping that can be used more easily by a wider audience of researchers and clinicians.

How to measure snoring? A comparison of the microphone, cannula and piezoelectric sensor.

The objective of this study was to compare to each other the methods currently recommended by the American Academy of Sleep Medicine (AASM) to measure snoring: an acoustic sensor, a piezoelectric sensor and a nasal pressure transducer (cannula). Ten subjects reporting habitual snoring were included in the study, performed at Landspitali-University Hospital, Iceland. Snoring was assessed by listening to the air medium microphone located on a patient's chest, compared to listening to two overhead air medium microphones (stereo) and manual scoring of a piezoelectric sensor and nasal cannula vibrations. The chest audio picked up the highest number of snore events of the different snore sensors. The sensitivity and positive predictive value of scoring snore events from the different sensors was compared to the chest audio: overhead audio (0.78, 0.98), cannula (0.55, 0.67) and piezoelectric sensor (0.78, 0.92), respectively. The chest audio was capable of detecting snore events with lower volume and higher fundamental frequency than the other sensors. The 200 Hz sampling rate of the cannula and piezoelectric sensor was one of their limitations for detecting snore events. The different snore sensors do not measure snore events in the same manner. This lack of consistency will affect future research on the clinical significance of snoring. Standardization of objective snore measurements is therefore needed. Based on this paper, snore measurements should be audio-based and the use of the cannula as a snore sensor be discontinued, but the piezoelectric sensor could possibly be modified for improvement.

Non-contact assessment of COD and turbidity concentrations in water using diffuse reflectance UV-Vis spectroscopy.

Water contamination is an important environmental concern underlining the need for reliable real-time information on contaminant concentrations in natural waters. Here, a new non-contact UV-Vis spectroscopic approach for monitoring contaminants in water, and especially wastewater, is proposed. Diffuse reflectance UV-Vis spectroscopy was applied to measure simultaneously the chemical oxygen demand (COD) and turbidity (TUR) concentrations in water. The measurements were carried out in the wavelength range from 200-1100 nm. The measured spectra were analysed using partial-least-squares (PLS) regression. The correlation coefficient between the measured and the reference concentrations of COD and TUR in the water samples were R(2) = 0.85 and 0.96, respectively. These results highlight the potential of non-contact UV-Vis spectroscopy for the assessment of water contamination. A system built on the concept would be able to monitor wastewater pollution continuously, without the need for laborious sample collection and subsequent laboratory analysis. Furthermore, since no parts of the system are in contact with the wastewater stream the need for maintenance is minimised.

Cyclic conductance switching in networks of redox-active molecular junctions.

Redox-active dithiolated tetrathiafulvalene derivatives (TTFdT) were inserted in two-dimensional nanoparticle arrays to build interlinked networks of molecular junctions. Upon oxidation of the TTFdT to the dication state, we observed a conductance increase of the networks by up to 1 order of magnitude. Successive oxidation and reduction cycles demonstrated a clear switching behavior of the molecular junction conductance. These results show the potential of interlinked nanoparticle arrays as chemical sensors.

Light-controlled conductance switching of ordered metal-molecule-metal devices.

We demonstrate reversible, light-controlled conductance switching of molecular devices based on photochromic diarylethene molecules. These devices consist of ordered, two-dimensional lattices of gold nanoparticles, in which neighboring particles are bridged by switchable molecules. We independently confirm that reversible isomerization of the diarylethenes employed is at the heart of the room-temperature conductance switching. For this, we take full advantage of the possibility to use optical spectroscopy to follow molecular switching in these samples.