An expedited model for identifying potential patients with periodic leg movements.

Periodic leg movements during sleep (PLMS) may have crucial consequences in adults. This study aimed to identify baseline characteristics, symptoms, or questionnaires that could help to identify sleep-disordered breathing patients with significant PLMS. Patients aged 20-80 years who underwent polysomnography for assessing sleep disturbance were included. Various factors such as sex, age, body measurements, symptoms, apnea-hypopnea index (AHI), and sleep quality scales were analysed to determine the presence of PLMS. The study included 1480 patients with a mean age of 46.4 ± 13.4 years, among whom 110 (7.4%) had significant PLMS with a PLM index of 15 or higher. There were no significant differences observed in terms of sex or BMI between patients with and without significant PLMS. However, the odds ratios (OR) for PLMS were 4.33, 4.41, and 4.23 in patients who were aged over 50 years, had insomnia, or had an ESS score of less than 10, respectively. Notably, the OR increased up to 67.89 times in patients who presented with all three risk factors. Our analysis identified significant risk factors for PLMS: age over 50, self-reported insomnia, and lower daytime sleepiness levels. These findings aid in identifying potential PLMS patients, facilitating confirmatory examinations and managing associated comorbidities.

Combining a wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity.

STUDY OBJECTIVES: The gold standard for diagnosing obstructive sleep apnea (OSA) is polysomnography (PSG). However, PSG is a time-consuming method with clinical limitations. This study aimed to create a wireless radar framework to screen the likelihood of 2 levels of OSA severity (ie, moderate-to-severe and severe OSA) in accordance with clinical practice standards. METHODS: We conducted a prospective, simultaneous study using a wireless radar system and PSG in a Northern Taiwan sleep center, involving 196 patients. The wireless radar sleep monitor, incorporating hybrid models such as deep neural decision trees, estimated the respiratory disturbance index relative to the total sleep time established by PSG (RDIPSG_TST), by analyzing continuous-wave signals indicative of breathing patterns. Analyses were performed to examine the correlation and agreement between the RDIPSG_TST and apnea-hypopnea index, results obtained through PSG. Cut-off thresholds for RDIPSG_TST were determined using Youden's index, and multiclass classification was performed, after which the results were compared. RESULTS: A strong correlation (ρ = 0.91) and agreement (average difference of 0.59 events/h) between apnea-hypopnea index and RDIPSG_TST were identified. In terms of the agreement between the 2 devices, the average difference between PSG-based apnea-hypopnea index and radar-based RDIPSG_TST was 0.59 events/h, and 187 out of 196 cases (95.41%) fell within the 95% confidence interval of differences. A moderate-to-severe OSA model achieved an accuracy of 90.3% (cut-off threshold for RDIPSG_TST: 19.2 events/h). A severe OSA model achieved an accuracy of 92.4% (cut-off threshold for RDIPSG_TST: 28.86 events/h). The mean accuracy of multiclass classification performance using these cut-off thresholds was 83.7%. CONCLUSIONS: The wireless-radar-based sleep monitoring device, with cut-off thresholds, can provide rapid OSA screening with acceptable accuracy and also alleviate the burden on PSG capacity. However, to independently apply this framework, the function of determining the radar-based total sleep time requires further optimizations and verification in future work. CITATION: Lin S-Y, Tsai C-Y, Majumdar A, et al. Combining a wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity. J Clin Sleep Med. 2024;20(8):1267-1277.

Detection of preceding sleep apnea using ECG spectrogram during CPAP titration night: A novel machine-learning and bag-of-features framework.

Obstructive sleep apnea (OSA) has a heavy health-related burden on patients and the healthcare system. Continuous positive airway pressure (CPAP) is effective in treating OSA, but adherence to it is often inadequate. A promising solution is to detect sleep apnea events in advance, and to adjust the pressure accordingly, which could improve the long-term use of CPAP treatment. The use of CPAP titration data may reflect a similar response of patients to therapy at home. Our study aimed to develop a machine-learning algorithm using retrospective electrocardiogram (ECG) data and CPAP titration to forecast sleep apnea events before they happen. We employed a support vector machine (SVM), k-nearest neighbour (KNN), decision tree (DT), and linear discriminative analysis (LDA) to detect sleep apnea events 30-90 s in advance. Preprocessed 30 s segments were time-frequency transformed to spectrograms using continuous wavelet transform, followed by feature generation using the bag-of-features technique. Specific frequency bands of 0.5-50 Hz, 0.8-10 Hz, and 8-50 Hz were also extracted to detect the most detected band. Our results indicated that SVM outperformed KNN, LDA, and DT across frequency bands and leading time segments. The 8-50 Hz frequency band gave the best accuracy of 98.2%, and a F1-score of 0.93. Segments 60 s before sleep events seemed to exhibit better performance than other pre-OSA segments. Our findings demonstrate the feasibility of detecting sleep apnea events in advance using only a single-lead ECG signal at CPAP titration, making our proposed framework a novel and promising approach to managing obstructive sleep apnea at home.

Predicting Fatigue-Associated Aberrant Driving Behaviors Using a Dynamic Weighted Moving Average Model With a Long Short-Term Memory Network Based on Heart Rate Variability.

OBJECTIVE: This study proposed a moving average (MA) approach to dynamically process heart rate variability (HRV) and developed aberrant driving behavior (ADB) prediction models by using long short-term memory (LSTM) networks. BACKGROUND: Fatigue-associated ADBs have traffic safety implications. Numerous models to predict such acts based on physiological responses have been developed but are still in embryonic stages. METHOD: This study recorded the data of 20 commercial bus drivers during their routine tasks on four consecutive days and subsequently asked them to complete questionnaires, including subjective sleep quality, driver behavior questionnaire and the Karolinska Sleepiness Scale. Driving behaviors and corresponding HRV were determined using a navigational mobile application and a wristwatch. The dynamic-weighted MA (DWMA) and exponential-weighted MA were used to process HRV in 5-min intervals. The data were independently separated for training and testing. Models were trained with 10-fold cross-validation strategy, their accuracies were evaluated, and Shapley additive explanation (SHAP) values were used to determine feature importance. RESULTS: Significant increases in the standard deviation of NN intervals (SDNN), root mean square of successive heartbeat interval differences (RMSSD), and normalized spectrum of high frequency (nHF) were observed in the pre-event stage. The DWMA-based model exhibited the highest accuracy for both driver types (urban: 84.41%; highway: 80.56%). The SDNN, RMSSD, and nHF demonstrated relatively high SHAP values. CONCLUSION: HRV metrics can serve as indicators of mental fatigue. DWMA-based LSTM could predict the occurrence of the level of fatigue associated with ADBs. APPLICATION: The established models can be used in realistic driving scenarios.

Air pollution exacerbates mild obstructive sleep apnea by disrupting nocturnal changes in lower-limb body composition: a cross-sectional study conducted in urban northern Taiwan.

BACKGROUND: Few studies have explored the role of body composition linking air pollution to obstructive sleep apnea (OSA). OBJECTIVE: To estimate the effects of air pollution on body composition and OSA, and that of body composition on OSA. METHODS: This study included 3550 individuals. A spatiotemporal model estimated personal exposure. Nocturnal changes in body composition were assessed through bioelectric impedance analysis. OSA was diagnosed using polysomnography. A generalized linear model was used to evaluate the absolute nocturnal changes in body composition associated with an interquartile range (IQR) increase in pollutants. A generalized logistic model was used to estimate odds ratios (ORs) of mild-OSA compared to non-OSA. Association between body composition and apnea-hypopnea index (AHI) was investigated through partial least squares (PLS) regression. RESULTS: Nocturnal changes in lower-limb body composition were associated with NO2 and PM2.5 in all patients. In participants with AHI <15, both short- and long-term NO2 exposures affected body composition and mild-OSA, while PM2.5 was not associated with either outcome. In a PLS model incorporating eight NO2-associated lower-limb parameters, the variable importance projection scores (VIP) of left leg impedance (LLIMP), predicted muscle mass (LLPMM), fat-free mass (LLFFM), and right leg impedance (RLIMP) exceeded 1; the corresponding coefficients ranked in the top four for AHI prediction. The adjusted OR (mild vs. non-OSA) was 1.67 (95 % CI: 1.36-2.03) associated with an IQR increase in prediction value estimated from body compositions. Notably, the two-pollutant model investigating the effects of pollutants on body compositions revealed associations of four parameters (LLIMP, LLPMM, LLFFM, and RLIMP) with NO2 in all lags, which indicates their indispensability in the association between NO2 and AHI. CONCLUSIONS: NO2 exacerbates mild-OSA by disrupting nocturnal changes in lower-limb body composition of patients with AHI <15. PM2.5 was associated with nocturnal changes in lower-limb body composition but not with mild-OSA.

Higher Particulate Matter Deposition in Alveolar Region Could Accelerate Body Fat Accumulation in Obstructive Sleep Apnea.

We conducted a cross-sectional study to investigate associations of particulate matter (PM) of less than 2.5 µm in aerodynamic diameter (PM2.5) and PM deposition with nocturnal changes in body composition in obstructive sleep apnea (OSA) patients. A bioelectric impedance analysis was used to measure the pre- and postsleep body composition of 185 OSA patients. Annual exposure to PM2.5 was estimated by the hybrid kriging/land-use regression model. A multiple-path particle dosimetry model was employed to estimate PM deposition in lung regions. We observed that an increase in the interquartile range (IQR) (1 µg/m3) of PM2.5 was associated with a 20.1% increase in right arm fat percentage and a 0.012 kg increase in right arm fat mass in OSA (p < 0.05). We observed that a 1 µg/m3 increase in PM deposition in lung regions (i.e., total lung region, head and nasal region, tracheobronchial region, and alveolar region) was associated with increases in changes of fat percentage and fat mass of the right arm (ß coefficient) (p < 0.05). The ß coefficients decreased as follows: alveolar region > head and nasal region > tracheobronchial region > total lung region (p < 0.05). Our findings demonstrated that an increase in PM deposition in lung regions, especially in the alveolar region, could be associated with nocturnal changes in the fat percentage and fat mass of the right arm. PM deposition in the alveolar region could accelerate the body fat accumulation in OSA.

Continuous Positive Airway Pressure Reduces Plasma Neurochemical Levels in Patients with OSA: A Pilot Study.

Obstructive sleep apnea (OSA) is a risk factor for neurodegenerative diseases. This study determined whether continuous positive airway pressure (CPAP), which can alleviate OSA symptoms, can reduce neurochemical biomarker levels. Thirty patients with OSA and normal cognitive function were recruited and divided into the control (n = 10) and CPAP (n = 20) groups. Next, we examined their in-lab sleep data (polysomnography and CPAP titration), sleep-related questionnaire outcomes, and neurochemical biomarker levels at baseline and the 3-month follow-up. The paired t-test and Wilcoxon signed-rank test were used to examine changes. Analysis of covariance (ANCOVA) was performed to increase the robustness of outcomes. The Epworth Sleepiness Scale and Pittsburgh Sleep Quality Index scores were significantly decreased in the CPAP group. The mean levels of total tau (T-Tau), amyloid-beta-42 (Aß42), and the product of the two (Aß42 × T-Tau) increased considerably in the control group (ΔT-Tau: 2.31 pg/mL; ΔAß42: 0.58 pg/mL; ΔAß42 × T-Tau: 48.73 pg2/mL2), whereas the mean levels of T-Tau and the product of T-Tau and Aß42 decreased considerably in the CPAP group (ΔT-Tau: -2.22 pg/mL; ΔAß42 × T-Tau: -44.35 pg2/mL2). The results of ANCOVA with adjustment for age, sex, body mass index, baseline measurements, and apnea-hypopnea index demonstrated significant differences in neurochemical biomarker levels between the CPAP and control groups. The findings indicate that CPAP may reduce neurochemical biomarker levels by alleviating OSA symptoms.

Associations of overnight changes in body composition with positional obstructive sleep apnea.

PURPOSE: Body composition is considered to be associated with obstructive sleep apnea (OSA) severity. This cross-sectional study aimed to examine associations of overnight body composition changes with positional OSA. METHODS: The body composition of patients diagnosed with non-positional and positional OSA was measured before and after overnight polysomnography. Odds ratios (ORs) of outcome variables between the case (positional OSA) and reference (non-positional OSA) groups were examined for associations with sleep-related parameters and with changes in body composition by a logistic regression analysis. RESULTS: Among 1584 patients with OSA, we used 1056 patients with non-positional OSA as the reference group. We found that a 1-unit increase in overnight changes of total fat percentage and total fat mass were associated with 1.076-fold increased OR (95% confidence interval (CI): 1.014, 1.142) and 1.096-fold increased OR (95% CI: 1.010, 1.189) of positional OSA, respectively (all p < 0.05). Additionally, a 1-unit increase in overnight changes of lower limb fat percentage and upper limb fat mass were associated with 1.043-fold increased OR (95% CI: 1.004, 1.084) and 2.638-fold increased OR (95% CI: 1.313, 5.302) of positional OSA, respectively (all p < 0.05). We observed that a 1-unit increase in overnight changes of trunk fat percentage and trunk fat mass were associated with 1.056-fold increased OR (95% CI: 1.008, 1.106) and 1.150-fold increased OR (95% CI: 1.016, 1.301) of positional OSA, respectively (all p < 0.05). CONCLUSION: Our findings indicated that nocturnal changes in the body's composition, especially total fat mass, total fat percentage, lower limb fat percentage, upper limb fat mass, trunk fat percentage, and trunk fat mass, may be associated with increased odds ratio of positional OSA compared with non-positional OSA.

Association of air pollution exposure with low arousal threshold obstructive sleep apnea: A cross-sectional study in Taipei, Taiwan.

Emerging evidence witnesses the association of air pollution exposure with sleep disorders or the risk of obstructive sleep apnea (OSA); however, the results are not consistent. OSA patients with or without a low arousal threshold (LAT) have different pathology and therapeutic schemes. No study has evaluated the potential diverse effects of air pollution on the phenotypes of OSA. The current study aimed to evaluate the associations of short-term and long-term exposure to air pollution with sleep-disordered measures and OSA phenotypes. This cross-sectional study consisted of 4634 participants from a sleep center in Taipei from January 2015 to April 2019. The personal exposure to ambient PM2.5 and NO2 was assessed by a spatial-temporal model. Overnight polysomnography was used to measure the sleep parameters. According to a developed clinical tool, we defined the low arousal threshold (LAT) and identified the OSA patients with or without LAT. We applied a generalized linear model and multinomial logistic regression model to estimate the change of sleep measures and risk of the OSA phenotypes, respectively, associated with an interquartile range (IQR) increment of personal pollution exposure after adjusting for the essential confounders. In the single-pollutant model, we observed the associations of NO2 with sleep-disordered measures by decreasing the total sleep time, sleep efficiency, extending the time of wake after sleep onset, and the association of NO2 with the increased risk of LAT OSA by around 15%. The two-pollutant model with both long-term and short-term exposures confirmed the most robust associations of long-term NO2 exposure with sleep measures. An IQR increment of NO2 averaged over the past year (6.0 ppb) decreased 3.32 min of total sleep time and 0.85% of sleep efficiency. Mitigating exposure to air pollution may improve sleep quality and reduce the risk of LAT OSA.

Association between cyclic variation in the heart rate index and biomarkers of neurodegenerative diseases in obstructive sleep apnea syndrome: A pilot study.

PURPOSE: Obstructive sleep apnea syndrome (OSAS) has mostly been examined using in-laboratory polysomnography (Lab-PSG), which may overestimate severity. This study compared sleep parameters in different environments and investigated the association between the plasma levels of neurochemical biomarkers and sleep parameters. METHODS: Thirty Taiwanese participants underwent Lab-PSG while wearing a single-lead electrocardiogram patch. Participants' blood samples were obtained in the morning immediately after the recording. Participants wore the patch for the subsequent three nights at home. Sleep disorder indices were calculated, including the apnea-hypopnea index (AHI), chest effort index, and cyclic variation of heart rate index (CVHRI). The 23 eligible participants' derived data were divided into the normal-to-moderate (N-M) group and the severe group according to American Association of Sleep Medicine (AASM) guidelines (Lab-PSG) and the recommendations of a previous study (Rooti Rx). Spearman's correlation was used to examine the correlations between sleep parameters and neurochemical biomarker levels. RESULTS: The mean T-Tau protein level was positively correlated with the home-based CVHRI (r = 0.53, p < 0.05), whereas no significant correlation was noted between hospital-based CVHRI and the mean T-tau protein level (r = 0.25, p = 0.25). The home-based data revealed that the mean T-Tau protein level in the severe group was significantly higher than that in the N-M group (severe group: 24.75 ± 6.16 pg/mL, N-M group: 19.65 ± 3.90 pg/mL; p < 0.05). Furthermore, the mean in-hospital CVHRI was higher than the mean at-home values (12.16 ± 13.66 events/h). CONCLUSION: Severe OSAS patients classified by home-based CVHRI demonstrated the higher T-Tau protein level, and CVHRI varied in different sleep environments.

Machine learning approaches for screening the risk of obstructive sleep apnea in the Taiwan population based on body profile.

(a) Objective: Obstructive sleep apnea syndrome (OSAS) is typically diagnosed through polysomnography (PSG). However, PSG incurs high medical costs. This study developed new models for screening the risk of moderate-to-severe OSAS (apnea-hypopnea index, AHI ≥15) and severe OSAS (AHI ≥30) in various age groups and sexes by using anthropometric features in the Taiwan population.(b) Participants: Data were derived from 10,391 northern Taiwan patients who underwent PSG.(c) Methods: Patients' characteristics - namely age, sex, body mass index (BMI), neck circumference, and waist circumference - was obtained. To develop an age- and sex-independent model, various approaches - namely logistic regression, k-nearest neighbor, naive Bayes, random forest (RF), and support vector machine - were trained for four groups based on sex and age (men or women; aged <50 or ≥50 years). Dataset was separated independently (training:70%; validation: 10%; testing: 20%) and Cross-validated grid search was applied for model optimization. Models demonstrating the highest overall accuracy in validation outcomes for the four groups were used to predict the testing dataset.(d) Results: The RF models showed the highest overall accuracy. BMI was the most influential parameter in both types of OSAS severity screening models.(e) Conclusion: The established models can be applied to screen OSAS risk in the Taiwan population and those with similar craniofacial features.

Associations among sleep-disordered breathing, arousal response, and risk of mild cognitive impairment in a northern Taiwan population.

STUDY OBJECTIVES: Dementia is associated with sleep disorders. However, the relationship between dementia and sleep arousal remains unclear. This study explored the associations among sleep parameters, arousal responses, and risk of mild cognitive impairment (MCI). METHODS: Participants with the chief complaints of memory problems and sleep disorders, from the sleep center database of Taipei Medical University Shuang-Ho Hospital, were screened, and the parameters related to the Cognitive Abilities Screening Instrument, Clinical Dementia Rating, and polysomnography were determined. All examinations were conducted within 6 months and without a particular order. The participants were divided into those without cognitive impairment (Clinical Dementia Rating = 0) and those with MCI (Clinical Dementia Rating = 0.5). Mean comparison, linear regression models, and logistic regression models were employed to investigate the associations among obtained variables. RESULTS: This study included 31 participants without MCI and 37 with MCI (17 with amnestic MCI, 20 with multidomain MCI). Patients with MCI had significantly higher mean values of the spontaneous arousal index and spontaneous arousal index in the non-rapid eye movement stage than those without MCI. An increased risk of MCI was significantly associated with increased spontaneous arousal index and spontaneous arousal index in the non-rapid eye movement stage with various adjustments. Significant associations between the Cognitive Abilities Screening Instrument scores and the oximetry parameters and sleep disorder indexes were observed. CONCLUSIONS: Repetitive respiratory events with hypoxia were associated with cognitive dysfunction. Spontaneous arousal, especially in non-rapid eye movement sleep, was related to the risk of MCI. However, additional longitudinal studies are required to confirm their causality. CITATION: Tsai C-Y, Hsu W-H, Lin Y-T, et al. Associations among sleep-disordered breathing, arousal response, and risk of mild cognitive impairment in a northern Taiwan population. J Clin Sleep Med. 2022;18(4): 1003-1012.

Comparison of Hospital-Based and Home-Based Obstructive Sleep Apnoea Severity Measurements with a Single-Lead Electrocardiogram Patch.

Obstructive sleep apnoea (OSA) is a global health concern, and polysomnography (PSG) is the gold standard for assessing OSA severity. However, the sleep parameters of home-based and in-laboratory PSG vary because of environmental factors, and the magnitude of these discrepancies remains unclear. We enrolled 125 Taiwanese patients who underwent PSG while wearing a single-lead electrocardiogram patch (RootiRx). After the PSG, all participants were instructed to continue wearing the RootiRx over three subsequent nights. Scores on OSA indices-namely, the apnoea-hypopnea index, chest effort index (CEI), cyclic variation of heart rate index (CVHRI), and combined CVHRI and CEI (Rx index), were determined. The patients were divided into three groups based on PSG-determined OSA severity. The variables (various severity groups and environmental measurements) were subjected to mean comparisons, and their correlations were examined by Pearson's correlation coefficient. The hospital-based CVHRI, CEI, and Rx index differed significantly among the severity groups. All three groups exhibited a significantly lower percentage of supine sleep time in the home-based assessment, compared with the hospital-based assessment. The percentage of supine sleep time (∆Supine%) exhibited a significant but weak to moderate positive correlation with each of the OSA indices. A significant but weak-to-moderate correlation between the ∆Supine% and ∆Rx index was still observed among the patients with high sleep efficiency (≥80%), who could reduce the effect of short sleep duration, leading to underestimation of the patients' OSA severity. The high supine percentage of sleep may cause OSA indices' overestimation in the hospital-based examination. Sleep recording at home with patch-type wearable devices may aid in accurate OSA diagnosis.

Associations of ambient air pollution with overnight changes in body composition and sleep-related parameters.

This study aims to investigate the association of air pollution with overnight change in 4body composition and sleep-related parameters. Body composition of 197 subjects in New Taipei city was measured before and after sleep by bioelectric impedance analysis. Air pollutant data were collected from Taiwan Environmental Protection Administration. Sleep parameters were examined by polysomnography. We observed fine particulate matter (PM2.5) decreased arterial oxygen saturation (SaO2) and increased apnea-hypopnea index (AHI); NO2 increased arousal, AHI, and decreased mean SaO2; and O3 inmcreased mean SaO2. We observed 0.99-µg/m3 increase in PM2.5 was associated with 18.8% increase in changes of right arm fat percentage (95% confidence interval (CI): 0.004, 0.375) and 0.011-kg increase in changes of right arm fat mass (95% CI: 0.000, 0.021). 2.45-ppb increase in NO2 was associated with 0.181-kg decrease in changes of muscle mass (95% CI: -0.147, -0.001), 0.192-kg decrease in changes of fat free mass (95% CI: -0.155, -0.001), 21.1% increase in changes of right leg fat percentage (95% CI: 0.012, 0.160), and 21.3% increase in changes of left leg fat percentage (95% CI: 0.006, 0.168). 1.56-ppb increase in O3 was associated with 29.3% decrease in changes of right leg fat percentage (95% CI: -0.363, -0.013), 0.058-kg increase in changes of right leg fat free mass (95% CI: 0.008, 0.066), and 0.059-kg increase in changes of right leg muscle mass (95% CI: 0.010, 0.066). We observed AHI was associated with overnight changes in fat percentage, total fat mass, muscle mass, bone mass, fat free mass, extracellular water, basal metabolic rate, leg fat percentage, leg fat mass, and trunk fat percentage (p < 0.05). In conclusion, exposure to air pollutants was associated with overnight body composition changes and sleep-related parameters. Nocturnal changes in total muscle mass and leg fat percentage likely contribute to the relationship between air pollution and obstructive sleep apnea.

Differentiation Model for Insomnia Disorder and the Respiratory Arousal Threshold Phenotype in Obstructive Sleep Apnea in the Taiwanese Population Based on Oximetry and Anthropometric Features.

Insomnia disorder (ID) and obstructive sleep apnea (OSA) with respiratory arousal threshold (ArTH) phenotypes often coexist in patients, presenting similar symptoms. However, the typical diagnosis examinations (in-laboratory polysomnography (lab-PSG) and other alternatives methods may therefore have limited differentiation capacities. Hence, this study established novel models to assist in the classification of ID and low- and high-ArTH OSA. Participants reporting insomnia as their chief complaint were enrolled. Their sleep parameters and body profile were accessed from the lab-PSG database. Based on the definition of low-ArTH OSA and ID, patients were divided into three groups, namely, the ID, low- and high-ArTH OSA groups. Various machine learning approaches, including logistic regression, k-nearest neighbors, naive Bayes, random forest (RF), and support vector machine, were trained using two types of features (Oximetry model, trained with oximetry parameters only; Combined model, trained with oximetry and anthropometric parameters). In the training stage, RF presented the highest cross-validation accuracy in both models compared with the other approaches. In the testing stage, the RF accuracy was 77.53% and 80.06% for the oximetry and combined models, respectively. The established models can be used to differentiate ID, low- and high-ArTH OSA in the population of Taiwan and those with similar craniofacial features.

Sexual specialization in phenology in dioecious Ficus benguetensis and its consequences for the mutualism.

BACKGROUND: Timing of reproductive events has become central in ecological studies linking success in pollination and seed dispersion to optimizing the probability and periods of encounters with pollinators or dispersers. Obligate plant-insect interactions, especially Ficus-fig wasp mutualisms, offer striking examples of fine-tuned encounter optimization as biological cycles between mutualistic partners are deeply dependent on each other and intertwined over generations. Despite fig flowering phenology being crucial in maintaining Ficus-fig wasp mutualisms, until now, the forces of selection shaping the phenological evolution of dioecious fig trees have received little attention. By conducting a 2-year survey of a population of Ficus benguetensis in Northern Taiwan, we assessed whether environmental factors or other selective pressures shape the phenology of male and female fig trees. RESULTS: Constraints by mutualistic pollinating wasps and seed dispersers, rather than climatic factors, appeared to mainly shape fig phenology and allometry in F. benguetensis. We identified a new sexual specialization in dioecious fig trees: the position of fig production. We propose that the continuous male fig production on tree trunks can enhance the survival of pollinating fig wasps through faster localization of receptive figs while reducing the mutualistic conflict between the fig and its obligate pollinators. By contrast, in female trees, fig production is massive in summer, located on the twigs of the foliar crown and seem more related to seed dispersal and germination. CONCLUSIONS: Identifying variations in the allometry and phenology of dioecious figs provide valuable insights into how monoecious and dioecious species resolve mutualism conflicts and into the emergence of dioecy in fig trees.