Validation of a novel mask-based device for monitoring of comprehensive sleep parameters and sleep disordered breathing

Sleep And Breathing, Год журнала: 2025, Номер 29(1)

Опубликована: Янв. 20, 2025

Abstract Purpose This study aimed to validate the new DormoTech Vlab device’s performance, usability, and validity as a sleep test physiological data recorder. The novel device has been designed for patient comfort, ease of use, home-based assessment disordered breathing other sleep-related measurements. Methods Forty-seven adults (mean age = 52 years, 42% female, body mass index 29.4 kg/m 2 ) underwent simultaneous testing with routine full polysomnography (PSG) using Nox A1 system (K192469, Medical). studies were manually independently scored according recommended guidelines. primary outcome measure was apnea-hypopnea (AHI) its corresponding conventional severity level (i.e., normal, mild, moderate, severe). Secondary endpoints included standard PSG parameters. Results AHI 21.7 ± 24.2 events/h deviation) versus 21.5 23.9 gold ( p 0.7). When grouped by severity, inter-test agreement high (Cohen’s kappa 0.97). between two systems largely similar in secondary endpoints, correlation systems, statistically significant < 0.05) differences only REM latency provides clinically near-identical interpretation almost all cases. Conclusion Based on these results, can be considered substantially equivalent reference terms efficacy, validity. Clinical Trial Registration name: Evaluation Usability Performance Assessment VLAB Device Home Sleep Test Identification number: NCT06224972. Date Registration: 2023-12-06.

Язык: Английский

---

Prospective Single-Arm Study of Remifentanil-Propofol Anesthesia with Manual Right Hypochondrial Compression for Painless Gastroscopy in Obese Patients

Drug Design Development and Therapy, Год журнала: 2025, Номер Volume 19, С. 877 - 890

Опубликована: Фев. 1, 2025

Purpose: The provision of comfortable and safe environment for painless gastroscopy in obese patients is an urgent clinical problem. This study aimed to determine the efficacy safety novel Li anesthetic protocol obesity (LAPO) which included remifentanil-propofol regimen, manual right hypochondrial compression (MRHC), easy-to-create mask, jaw thrust at preoperative patients. Patients Methods: prospective, single-center, single-arm trial recruited 106 participants underwent LAPO gastroscopy. primary outcome was incidence hypoxemia (peripheral oxygen saturation [SpO 2 ]: 75% ≤ SpO < 90%, > 10 s 60 s). Second outcomes severe hypoxemia, lowest (L-SpO ), duration other events. Results: 98 under LAPO, median body mass index (BMI) 39.2 kg/m 27.5%, while conventional (CAPO) reference 40.4% with BMI 31.4 . With increase class obesity, a significant rise observed, from I by 11.8%, 15.1% II, 41.7% III. Paired t test showed that L-SpO significantly higher than overnight polysomnography (Nadir ) (92% vs 76%, P< 0.001). Moreover, obstructive sleep apnea (OSA) associated 4.019-fold risk (Odds ratios [OR], 4.019; 95% confidence interval [CI], 1.184 14.610; P=0.028); diabetes 4.790-fold (OR, 4.790; CI, 1.288 23.600; P=0.030). Conclusion: Compared CAPO, reduced so, effective finding might provide some new schedules management absence advanced airway support instruments. Clinical Trial Registration: ChiCTR2300077889. Plain Language Summary: advancement comfortable-oriented medication, mostly performed sedation anesthesia. However, there are changes anatomy structure such as short neck, round chin, large tongue, makes them prone obstruction after or anesthesia, blocking passage out, will cause over time further damage body. Therefore, it necessary explore anesthesia obesity. Our team's work demonstrates (strategic remifentanil combined propofol respiratory through MRHC thrust, increased reserve via mask) lacking devices efficient. Furthermore, our findings contribute understanding values can tolerate during procedures, establishing Nadir cut-off point. We also identified OSA independent predictive factors hypoxemia. If these supported research, could aid anesthesiologists developing specific protocols managing Ultimately, this expand access medical treatment broader spectrum Keywords: compression, patient, apnea, gastroscopy, propofol,

Язык: Английский

---

Performance evaluation of an under‐mattress sleep sensor versus polysomnography in > 400 nights with healthy and unhealthy sleep

Journal of Sleep Research, Год журнала: 2025, Номер unknown

Опубликована: Фев. 28, 2025

Consumer sleep trackers provide useful insight into sleep. However, large-scale performance evaluation studies are needed to properly understand tracker accuracy. This study evaluated of an under-mattress sensor estimate and wake versus polysomnography in a large sample, including individuals with without disorders during day night opportunities, across multiple in-laboratory studies. One-hundred eighty-three participants (51%/49% male/female, mean [SD] age = 45 [18] years) attended the laboratory for research simultaneous (Withings Sleep Analyser) recordings. Epoch-by-epoch analyses determined accuracy, sensitivity specificity Withings Analyser polysomnography. Bland-Altman plots examined bias duration, efficiency, onset-latency, after onset. Overall sleep-wake classification accuracy was 83%, 95% 37%. The significantly overestimated total time (48 [81] min), efficiency (9 [15]%) sleep-onset latency (6 [26] underestimated onset (54 [78] min). Accuracy were higher daytime opportunities healthy (89% 47% 82% 26%, respectively, p < 0.05). also 97%) those (81% 91%, is comparable other consumer trackers, high but poor compared reasonably stable, more variable people disorder. Contactless, sensors show promise accurate monitoring, noting tendency over-estimate particularly where high.

Язык: Английский

---

A Comparative Study of Polysomnography‐Derived Sleep Disturbance in People Living With Multiple Sclerosis Compared to Matched Controls From the General Population

Journal of Sleep Research, Год журнала: 2025, Номер unknown

Опубликована: Апрель 28, 2025

ABSTRACT Sleep structure and sleep disorders were compared between people with multiple sclerosis (MS; n = 39) age, sex, BMI‐matched members of the general population ( using overnight polysomnography (PSG). Compared to controls, MS had a higher prevalence periodic limb movement disorder (PLMD; 59% vs. 18%, p < 0.001) PLM‐related arousals (PLMI: 21.1 0.8, 0.001); as well longer duration (402.9 [59.8] 370.4 [54.0] min, 0.014), median latency (12.4 min) reduced proportion total time in stage N1 (8.5% 14.8%, more N2 (54.4% 48.0%, 0.001). architecture appeared differ for MS, even context no recent exacerbations or relapse. Managing leg movements during may help improve quality MS.

Язык: Английский

---

Clinical Validation of Artificial Intelligence Algorithms for the Diagnosis of Adult Obstructive Sleep Apnea and Sleep Staging From Oximetry and Photoplethysmography—SleepAI

Journal of Sleep Research, Год журнала: 2025, Номер unknown

Опубликована: Май 10, 2025

ABSTRACT Home sleep apnea tests (HSATs) have emerged as alternatives to in‐laboratory polysomnography (PSG), but Type IV HSATs often show limited diagnostic performance. This study clinically validates SleepAI, a novel remote digital health system that applies AI algorithms raw oximetry data for automated staging and obstructive (OSA) diagnosis. SleepAI were trained on over 10,000 PSG recordings. The consists of wearable oximeter connected via Bluetooth mobile app transmitting cloud‐based platform AI‐driven analysis. Clinical validation was conducted in 53 subjects with suspected OSA, who used three nights at home one night centre alongside PSG. SleepAI's apnea‐hypopnea index (AHI) estimates three‐class (Wake, REM, NREM) compared references. For OSA severity classification (non‐OSA, mild, moderate, severe), achieved an overall accuracy 89%, F1‐scores 1.0, 0.9, 0.88, respectively. three‐stage Cohen's kappa 0.75. Night‐to‐night AHI variability showed 37.5% participants experienced one‐level change across home. No significant differences metrics found between the first subsequent home, indicating no disturbance by SleepAI. These findings support promising scalable alternative existing HSATs, potential address key clinical gaps improving accessibility.

Язык: Английский

---

Development and Validation of Machine Learning Models to Predict Postoperative Delirium Using Clinical Features and Polysomnography Variables

Journal of Clinical Medicine, Год журнала: 2024, Номер 13(18), С. 5485 - 5485

Опубликована: Сен. 16, 2024

Background: Delirium affects up to 50% of patients following high-risk surgeries and is associated with poor long-term prognosis. This study employed machine learning predict delirium using polysomnography (PSG) sleep-disorder questionnaire data, aimed identify key sleep-related factors for improved interventions patient outcomes. Methods: We studied 912 adults who underwent surgery under general anesthesia at a tertiary hospital (2013–2024) had PSG within 5 years surgery. was assessed via clinical diagnoses, antipsychotic prescriptions, psychiatric consultations 14 days postoperatively. Sleep-related data were collected questionnaires. Machine predictions performed postoperative delirium, focusing on model accuracy feature importance. Results: divided the into an internal training set (700) external test (212). Univariate analysis identified significant risk factors: midazolam use, prolonged duration, hypoalbuminemia. variables such as fewer rapid eye movement (REM) episodes higher daytime sleepiness also linked delirium. An extreme gradient-boosting-based classification task achieved AUC 0.81 variables, 0.60 alone, 0.84 both, demonstrating added value data. Analysis Shapley additive explanations values highlighted important predictors: age, PSG-derived oxygen saturation nadir, periodic limb index, REM episodes, relationship between sleep patterns Conclusions: The artificial intelligence integrates reliably identifies that contribute its identification. Predicting high developing closely monitoring them could reduce costs complications

Язык: Английский

---

A ballistocardiogram dataset with reference sensor signals in long-term natural sleep environments

Scientific Data, Год журнала: 2024, Номер 11(1)

Опубликована: Окт. 5, 2024

To facilitate unobtrusive and continuous sleep monitoring promote intelligent quality assessment, we present a dataset that includes multiple nights of ballistocardiogram (BCG) data collected using piezoelectric film sensors from 32 subjects in their regular environments. Besides, the referenced heart rate respiratory are also recorded by reference to validate accuracy cardiac components extracted BCG signals. The serves as foundation for research on vital sign based signals, offering support evaluation optimization quality.

Язык: Английский

---

CNN-SENet: A Convolutional Neural Network Model for Audio Snoring Detection Based on Channel Attention Mechanism

Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 24 - 35

Опубликована: Янв. 1, 2024

Язык: Английский

---

Mandibular device treatment in obstructive sleep apnea -A structured therapy adjustment considering night-to-night variability night-to-night variability in mandibular devices

Sleep And Breathing, Год журнала: 2024, Номер unknown

Опубликована: Сен. 6, 2024

Язык: Английский

---

Performance evaluation of an under-mattress sleep sensor versus polysomnography in >400 nights with healthy and unhealthy sleep

medRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Сен. 11, 2024

Abstract Consumer sleep trackers can provide useful insight into and patterns. However, large scale performance evaluation studies against direct measures are needed to comprehensively understand tracker accuracy. This study evaluated of an under-mattress sensor estimate wake versus polysomnography, during multiple in-laboratory protocols in a sample including individuals with without disorders day night opportunities. 183 participants (51% male, mean[SD] age=45[18] years) attended the laboratory for research that included simultaneous polysomnography (Withings Sleep Analyzer [WSA]) recordings. Epoch-by-epoch analyses confusion matrices were used determine accuracy, sensitivity, specificity WSA polysomnography. Bland-Altman plots examined bias duration, efficiency, onset-latency, after onset. Overall sleep-wake classification accuracy was 83%, sensitivity 95%, 37%. The significantly overestimated total time (48[81]minutes), efficiency (9[15]%), onset latency (6[26]), underestimated (54[78]), p<0.05. Accuracy higher daytime opportunities healthy (89% 47% 82% 26% respectively, p<0.05). also 97%) those (81% 91%, is comparable other consumer trackers, high but poor compared Poorer night-time likely due increased reduced efficiency. Contactless, sensors show promise accurate monitoring, noting tendency over-estimate particularly where high.

Язык: Английский

---