INTRODUCTION: Obstructive sleep apnoea (OSA) is a serious but underdiagnosed condition. Demand for the gold standard diagnostic polysomnogram (PSG) far exceeds its availability. More efficient diagnostic methods are needed, even in tertiary settings. Machine learning (ML) models have strengths in disease prediction and early diagnosis. We explored the use of ML with oximetry, demographic and anthropometric data to diagnose OSA. METHODS: A total of 2,996 patients were included for modelling and divided into test and training sets. Seven commonly used supervised learning algorithms were trained with the data. Sensitivity (recall), specificity, positive predictive value (PPV) (precision), negative predictive value, area under the receiver operating characteristic curve (AUC) and F1 measure were reported for each model. RESULTS: In the best performing four-class model (neural network model predicting no, mild, moderate or severe OSA), a prediction of moderate and/or severe disease had a combined PPV of 94%; one out of 335 patients had no OSA and 19 had mild OSA. In the best performing two-class model (logistic regression model predicting no-mild vs. moderate-severe OSA), the PPV for moderate-severe OSA was 92%; two out of 350 patients had no OSA and 26 had mild OSA. CONCLUSION Our study showed that the prediction of moderate-severe OSA in a tertiary setting with an ML approach is a viable option to facilitate early identification of OSA. Prospective studies with home-based oximeters and analysis of other oximetry variables are the next steps towards formal implementation.
Humans ; Oximetry/methods* ; Sleep Apnea, Obstructive/diagnosis* ; Male ; Female ; Middle Aged ; Machine Learning ; Polysomnography ; Adult ; Anthropometry ; ROC Curve ; Aged ; Algorithms ; Predictive Value of Tests ; Sensitivity and Specificity ; Neural Networks, Computer ; Demography

INTRODUCTION: Obstructive sleep apnoea (OSA) is common in Singapore, with moderate to severe OSA affecting around 30% of residents. These consensus statements aim to provide scientifically grounded recommendations for the management of OSA, standar-dise the management of OSA in Singapore and promote multidisciplinary collaboration. METHOD: An expert panel, which was convened in 2024, identified several areas of OSA management that require guidance. The expert panel reviewed the current literature and developed consensus statements, which were later independently voted on using a 3-point Likert scale (agree, neutral or disagree). Consensus (total ratings of agree and neutral) was set a priori at ≥80% agreement. Any statement not reaching consensus was excluded. RESULTS: The final consensus included 49 statements that provide guidance on the screening, diagnosis and management of adults with OSA. Additionally, 23 statements on the screening, diagnosis and management of paediatric OSA achieved consensus. These 72 consensus statements considered not only the latest clinical evidence but also the benefits and harms, resource implications, feasibility, acceptability and equity impact of the recommendations. CONCLUSION The statements presented in this paper aim to guide clinicians based on the most updated evidence and collective expert opinion from sleep specialists in Singapore. These recommendations should augment clinical judgement rather than replace it. Management decisions should be individualised, taking into account the patient's clinical characteristics, as well as patient and caregiver concerns and preferences.
Humans ; Sleep Apnea, Obstructive/diagnosis* ; Singapore ; Consensus ; Adult

BACKGROUND: Despite the evident age differences in the risk of asthma attack due to air pollution, most studies have recruited subjects from all age groups. Although this effort might be feasible for maintaining statistical power, it biases the effect estimate towards the null among children who are more sensitive to air pollution than adults. OBJECTIVES: To estimate the risk of air pollution on children living in Seoul who have made doctor visits for asthma. METHOD: From 1992 to 1993, daily number of doctor visits due to asthma attack was tallied among children between 4 and 11 years old living in Seoul from the insurance claim forms of Korean Medical Insurance Corporation (KMIC). 24-hour mean concentrations were calculated for TSP, SO2, O3 and NO2 based on hourly concentrations measured at 20 monitoring stations. To estimate the effect of TSP or SO2 on asthma attack, a Poisson regression model was used with adjustments for long-term trend, seasonal variation, day-of-week effect, and meteorological factors such as temperature, humidity, wind velocity and duration of sunshine. To prevent autocorrelation, autoregressive error terms were tried with different lag periods. RESULTS: The cumulative effect of a current day and previous five days turned out to be stronger than that of any single day. These relationships were observed more clearly after the effects of O3 and NO2 were controlled. For 100microgram/m3 increase of cumulative concentration of TSP, the relative risk was 1.27 (95% CI: 1.08, 1.49); for 50ppb increase of cumulative concentration of SO2, the relative risk was 1.56 (95% CI: 1.29, 1.89). After the effects of O3 and NO2 were removed in a multiple regression model, it increased to 1.37 (95% CI: 1.10, 1.69) and 1.66 ((%% CI: 1.34, 2.07), respectively. CONCLUSION: The concentrations of TSP and SO2 turned out to be significantly associatedwith asthma attack among children. As expected, the risk estimates were larger than those of previous studies which recruited subjects from all age groups, or used mortality or hospitalization as their outcome.
Adult ; Air Pollution* ; Asthma* ; Bias (Epidemiology) ; Child ; Hospitalization ; Humans ; Humidity ; Insurance ; Meteorological Concepts ; Mortality ; Seasons ; Seoul* ; Sunlight ; Wind