AIMS: To develop a model to detect the respiratory phases present in the pulmonary acoustic signals and to evaluate the performance of the model in detecting the respiratory phases.
METHODS: Normalised averaged power spectral density for each frame and change in normalised averaged power spectral density between the adjacent frames were fuzzified and fuzzy rules were formulated. The fuzzy inference system (FIS) was developed with both Mamdani and Sugeno methods. To evaluate the performance of both Mamdani and Sugeno methods, correlation coefficient and root mean square error (RMSE) were calculated.
RESULTS: In the correlation coefficient analysis in evaluating the fuzzy model using Mamdani and Sugeno method, the strength of the correlation was found to be r = 0.9892 and r = 0.9964, respectively. The RMSE for Mamdani and Sugeno methods are RMSE = 0.0853 and RMSE = 0.0817, respectively.
CONCLUSION: The correlation coefficient and the RMSE of the proposed fuzzy models in detecting the respiratory phases reveals that Sugeno method performs better compared with the Mamdani method.
OBJECTIVES: This study aimed to segment the breath cycles from pulmonary acoustic signals using the newly developed adaptive neuro-fuzzy inference system (ANFIS) based on breath phase detection and to subsequently evaluate the performance of the system.
METHODS: The normalised averaged power spectral density for each segment was fuzzified, and a set of fuzzy rules was formulated. The ANFIS was developed to detect the breath phases and subsequently perform breath cycle segmentation. To evaluate the performance of the proposed method, the root mean square error (RMSE) and correlation coefficient values were calculated and analysed, and the proposed method was then validated using data collected at KIMS Hospital and the RALE standard dataset.
RESULTS: The analysis of the correlation coefficient of the neuro-fuzzy model, which was performed to evaluate its performance, revealed a correlation strength of r = 0.9925, and the RMSE for the neuro-fuzzy model was found to equal 0.0069.
CONCLUSION: The proposed neuro-fuzzy model performs better than the fuzzy inference system (FIS) in detecting the breath phases and segmenting the breath cycles and requires less rules than FIS.
METHODS: This was a prospective, observational cohort study replacing the undivided nasal cannula with a divided nasal cannula during routine polysomnography (n = 28).
RESULTS: Integration of the divided nasal cannula pressure transducer system into routine polysomnography was easy and affordable. Most patients (89%) demonstrated nasal cycle changes during the test. Nasal cycle changes tended to occur during body position changes (62%) and transitions from non-rapid eye movement sleep to rapid eye movement sleep (41%). The mean nasal cycle duration was 2.5 ± 2.1 hours. Other sleep study metrics did not reveal statistically significant findings in relation to the nasal cycle.
CONCLUSIONS: Replacing an undivided nasal cannula with a divided nasal cannula is easy to implement, adding another physiologic measure to polysomnography. Although the divided nasal cannula did not significantly affect traditional polysomnographic metrics such as the apnea-hypopnea index or periodic limb movement index based on this small pilot study, we were able to replicate past nasal cycle findings that may be of interest to sleep clinicians and researchers. Given the ease with which the divided nasal cannula can be integrated, we encourage other sleep researchers to investigate the utility of using a divided nasal cannula during polysomnography.
METHODS: This was a planned post-hoc analysis of multicenter prospective cohort study involving 1,218 at-risk surgical patients without prior diagnosis of sleep apnea. All patients underwent home sleep apnea testing (ApneaLink Plus, ResMed) simultaneously with pulse oximetry (PULSOX-300i, Konica Minolta Sensing, Inc). The predictive performance of the 4% oxygen desaturation index (ODI) versus apnea-hypopnea index (AHI) were determined.
RESULTS: Of 1,218 patients, the mean age was 67.2 ± 9.2 years and body mass index (BMI) was 27.0 ± 5.3 kg/m2. The optimal cut-off for predicting moderate-to-severe and severe OSA was ODI ≥15 events/hour. For predicting moderate-to-severe OSA (AHI ≥15), the sensitivity and specificity of ODI ≥ 15 events per hour were 88.4% (95% confidence interval [CI], 85.7-90.6) and 95.4% (95% CI, 94.2-96.4). For severe OSA (AHI ≥30), the sensitivity and specificity were 97.2% (95% CI, 92.7-99.1) and 78.8% (95% CI, 78.2-79.0). The area under the curve (AUC) for moderate-to-severe and severe OSA was 0.983 (95% CI, 0.977-0.988) and 0.979 (95% CI, 0.97-0.909) respectively.
DISCUSSION: ODI from oximetry is sensitive and specific in predicting moderate-to-severe or severe OSA in at-risk surgical population. It provides an easy, accurate, and accessible tool for at-risk surgical patients with suspected OSA.