Krongthong Tawaranurak
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Krongthong Tawaranurak has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
Nattarin Nilrat
Department of Otolaryngology Head and Neck Surgery, Faculty of Medicine, Prince of Songkla University, Songkhla, Hat Yai, 90110, Thailand
2 shared publicationsChuanchom Bumrungsena
Nursing Department, Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla, Thailand
1 shared publication- Hnin Thiri Chaw
Department of Computer Engineering, Prince of Songkla University, Hat Yai, Thailand
1 shared publication Methini Werathammo
Department of Otolaryngology Head and Neck Surgery, Faculty of Medicine, Prince of Songkla University, Hat Yai, Thailand
1 shared publicationNavaponn Phuttasuatta
Department of Otolaryngology Head and Neck Surgery, Faculty of Medicine, Prince of Songkla University, Songkhla, Hat Yai, 90110, Thailand
1 shared publicationNiti Tawaranurak
Division of Plastic and Reconstructive Surgery, Department of Surgery, Faculty of Medicine Prince of Songkla University Songkhla Thailand
1 shared publication
Latest Publications
- Open Access
AP Model: A Simple Tool to Predict Poor Long‐Term PAP Adherence in Obstructive Sleep Apnea
Objective: To identify clinical and polysomnographic factors predicting poor long‐term positive airway pressure (PAP) compliance in patients with obstructive sleep apnea (OSA) in a Southeast Asian population and to develop a simple risk model applicable in routine practice. Study Design: Retrospective cohort study. Setting: Songklanagarind Hospital, Thailand, from January 2012 to December 2022. Methods: Adult OSA patients aged 18 to 65 years prescribed PAP therapy were included. Adherence was objectively recorded from device downloads. Good adherence was defined as ≥4 hours per night on ≥70% of nights at 12 months. Logistic regression identified predictors of poor long‐term adherence. Results: A total of 343 patients were enrolled; 253 had follow‐up data at 12 months. Good adherence was observed in 47.8% of patients. In multivariate analysis, age < 50 years (odds ratio [OR] 1.92; 95% CI 1.01‐3.67; P = .046) and poor short‐term adherence (OR 4.47; 95% CI 2.33‐8.72; P < .001) independently predicted poor long‐term adherence. The resulting “AP” model (Age and Poor early adherence) achieved an area under the curve of 0.73 (95% CI 0.66‐0.79), with a high specificity of 94%. Conclusion: Although predictors of PAP adherence have been described in Western populations, this study provides the first large data set from Thailand. The AP model, while simple, is pragmatic and easily applied in resource‐limited settings. Prospective, multicenter validation across Southeast Asia is warranted to enhance its generalizability and incorporate modifiable predictors.
Open AccessAP Model: A Simple Tool to Predict Poor Long‐Term PAP Adherence in Obstructive Sleep Apnea
Objective: To identify clinical and polysomnographic factors predicting poor long‐term positive airway pressure (PAP) compliance in patients with obstructive sleep apnea (OSA) in a Southeast Asian population and to develop a simple risk model applicable in routine practice. Study Design: Retrospective cohort study. Setting: Songklanagarind Hospital, Thailand, from January 2012 to December 2022. Methods: Adult OSA patients aged 18 to 65 years prescribed PAP therapy were included. Adherence was objectively recorded from device downloads. Good adherence was defined as ≥4 hours per night on ≥70% of nights at 12 months. Logistic regression identified predictors of poor long‐term adherence. Results: A total of 343 patients were enrolled; 253 had follow‐up data at 12 months. Good adherence was observed in 47.8% of patients. In multivariate analysis, age < 50 years (odds ratio [OR] 1.92; 95% CI 1.01‐3.67; P = .046) and poor short‐term adherence (OR 4.47; 95% CI 2.33‐8.72; P < .001) independently predicted poor long‐term adherence. The resulting “AP” model (Age and Poor early adherence) achieved an area under the curve of 0.73 (95% CI 0.66‐0.79), with a high specificity of 94%. Conclusion: Although predictors of PAP adherence have been described in Western populations, this study provides the first large data set from Thailand. The AP model, while simple, is pragmatic and easily applied in resource‐limited settings. Prospective, multicenter validation across Southeast Asia is warranted to enhance its generalizability and incorporate modifiable predictors.
- Open Access
- 19 September 2023
- Published by Wiley in Applied Computational Intelligence and Soft Computing
- Vol. 2023 (1) , 1-12
ZleepNet: A Deep Convolutional Neural Network Model for Predicting Sleep Apnea Using SpO2 Signal
Sleep apnea is one of the most common sleep disorders in the world. It is a common problem for patients to suffer from sleep disturbances. In this paper, we propose a deep convolutional neural network (CNN) model based on the oxygen saturation (SpO2) signal from a smart sensor. This is the reason why we called ZleepNet a network for sleep apnea detection. The proposed model includes three convolutional layers, which include ReLu activation function, 2 dense layers, and one dropout layer for predicting sleep apnea. In this proposed model, the use of signals for detecting the sleep apnea can be reduced from 25 sensors to 1 sensor. We conducted experiments to evaluate the performance of the proposed CNN using real patient data and compared them with traditional machine learning methods such as least discriminant analysis (LDA) and support vector machine (SVM), baggy representation tree, and artificial neural network (ANN) on publicly available sleep datasets using the same parameter setting. The results show that the proposed model outperformed the other methods with the accuracy of 91.30 with the split rate of 0.2 in which the training data are 20 and testing data are 80. The accuracy of the proposed CNN is 90.33 when compared with the LDA which achieved 86.5 accuracy with the split rate of 0.5 in which training data are 50 and testing data are 50. It achieved 91.56 accuracy when compared with the support vector machine (SVM) in which training data are 70 and testing data are 30. The achieved accuracy of the proposed CNN is 91.89 when compared with bagging representation tree in which training data are 90 and testing data are 10. The accuracy of the proposed CNN is 91.30 in which training data are 83 and testing data are 17 when compared with artificial neural networks (ANN).
- 19 September 2023
- Published by Wiley in Applied Computational Intelligence and Soft Computing
- Vol. 2023 (1) , 1-12
Open AccessZleepNet: A Deep Convolutional Neural Network Model for Predicting Sleep Apnea Using SpO2 Signal
Sleep apnea is one of the most common sleep disorders in the world. It is a common problem for patients to suffer from sleep disturbances. In this paper, we propose a deep convolutional neural network (CNN) model based on the oxygen saturation (SpO2) signal from a smart sensor. This is the reason why we called ZleepNet a network for sleep apnea detection. The proposed model includes three convolutional layers, which include ReLu activation function, 2 dense layers, and one dropout layer for predicting sleep apnea. In this proposed model, the use of signals for detecting the sleep apnea can be reduced from 25 sensors to 1 sensor. We conducted experiments to evaluate the performance of the proposed CNN using real patient data and compared them with traditional machine learning methods such as least discriminant analysis (LDA) and support vector machine (SVM), baggy representation tree, and artificial neural network (ANN) on publicly available sleep datasets using the same parameter setting. The results show that the proposed model outperformed the other methods with the accuracy of 91.30 with the split rate of 0.2 in which the training data are 20 and testing data are 80. The accuracy of the proposed CNN is 90.33 when compared with the LDA which achieved 86.5 accuracy with the split rate of 0.5 in which training data are 50 and testing data are 50. It achieved 91.56 accuracy when compared with the support vector machine (SVM) in which training data are 70 and testing data are 30. The achieved accuracy of the proposed CNN is 91.89 when compared with bagging representation tree in which training data are 90 and testing data are 10. The accuracy of the proposed CNN is 91.30 in which training data are 83 and testing data are 17 when compared with artificial neural networks (ANN).
- Open Access
- 27 June 2023
- Published by Springer Nature in BMC Research Notes
- Vol. 16 (1) , 1-6
Effect of positive airway pressure compliance on laryngopharyngeal reflux in obstructive sleep apnea patients
Objectives: To determine the effect of continuous positive airway pressure (CPAP) compliance on symptoms and signs of laryngopharyngeal reflux (LPR) in patients with obstructive sleep apnea (OSA). Results: Thirty patients were included. The participants were divided into good compliance (n = 21) and poor compliance (n = 9) groups. After 6-month CPAP treatment, the reflux symptom index score significantly decreased in both the good compliance group (20 (17,24) vs. 14 (10,18), p < 0.001) and the poor compliance group (21 (18,25) vs. 10 (5,16), p < 0.05). Reflux finding score was significantly reduced in both the good compliance group (8 (6,9) vs. 4 (3,5), p < 0.001) and the poor compliance group (6 (4,8) vs. 3 (2,4), p < 0.05). However, there were no statistically significant differences between the groups. CPAP treatment reduces the symptoms and signs of LPR. However, CPAP compliance does not correlate with improvement in LPR in patients with OSA.
- 27 June 2023
- Published by Springer Nature in BMC Research Notes
- Vol. 16 (1) , 1-6
Open AccessEffect of positive airway pressure compliance on laryngopharyngeal reflux in obstructive sleep apnea patients
Objectives: To determine the effect of continuous positive airway pressure (CPAP) compliance on symptoms and signs of laryngopharyngeal reflux (LPR) in patients with obstructive sleep apnea (OSA). Results: Thirty patients were included. The participants were divided into good compliance (n = 21) and poor compliance (n = 9) groups. After 6-month CPAP treatment, the reflux symptom index score significantly decreased in both the good compliance group (20 (17,24) vs. 14 (10,18), p < 0.001) and the poor compliance group (21 (18,25) vs. 10 (5,16), p < 0.05). Reflux finding score was significantly reduced in both the good compliance group (8 (6,9) vs. 4 (3,5), p < 0.001) and the poor compliance group (6 (4,8) vs. 3 (2,4), p < 0.05). However, there were no statistically significant differences between the groups. CPAP treatment reduces the symptoms and signs of LPR. However, CPAP compliance does not correlate with improvement in LPR in patients with OSA.
- Open Access
- 14 February 2022
- Published by Wiley in International Wound Journal
- Vol. 19 (7) , 1729-1735
Combination of fractional carbon dioxide laser and topical triamcinolone vs intralesional triamcinolone for keloid treatment: A randomised clinical trial
To compare the therapeutic effect of fractional carbon dioxide (CO2) laser + topical triamcinolone (TA) with intralesional TA on keloids. Twenty‐two participants were randomised into two groups: group A, treated with fractional CO2 laser + topical TA, and group B, treated with intralesional TA. The interventions were performed at every 4‐week interval until the keloids were resolved or at the completion of 1 year. At each session, the scar volume, Vancouver Scar Scale (VSS) were assessed. Recurrence was observed for 1 year. The mean scar volumes and VSS scores were not significantly different between the two groups. After 1 year, the scar volume change in group B was greater than group A (86.5% vs 59.1%, P‐value = .016). The mean VSS scores were significantly decreased in group A (8.0 ± 1.5 to 4.8 ± 1.6, P‐value <.001) and group B (8.4 ± 0.8 to 4.8 ± 1.6, P‐value <.001). The keloids were completely resolved in 63.6% and 72.7% of the patients, and recurrence was observed in 9.1% and 18.2% of the patients in groups A and B, respectively. The combination of fractional CO2 laser with topical TA was an alternative option for the treatment of keloids.
- 14 February 2022
- Published by Wiley in International Wound Journal
- Vol. 19 (7) , 1729-1735
Open AccessCombination of fractional carbon dioxide laser and topical triamcinolone vs intralesional triamcinolone for keloid treatment: A randomised clinical trial
To compare the therapeutic effect of fractional carbon dioxide (CO2) laser + topical triamcinolone (TA) with intralesional TA on keloids. Twenty‐two participants were randomised into two groups: group A, treated with fractional CO2 laser + topical TA, and group B, treated with intralesional TA. The interventions were performed at every 4‐week interval until the keloids were resolved or at the completion of 1 year. At each session, the scar volume, Vancouver Scar Scale (VSS) were assessed. Recurrence was observed for 1 year. The mean scar volumes and VSS scores were not significantly different between the two groups. After 1 year, the scar volume change in group B was greater than group A (86.5% vs 59.1%, P‐value = .016). The mean VSS scores were significantly decreased in group A (8.0 ± 1.5 to 4.8 ± 1.6, P‐value <.001) and group B (8.4 ± 0.8 to 4.8 ± 1.6, P‐value <.001). The keloids were completely resolved in 63.6% and 72.7% of the patients, and recurrence was observed in 9.1% and 18.2% of the patients in groups A and B, respectively. The combination of fractional CO2 laser with topical TA was an alternative option for the treatment of keloids.
- Open Access
- 4 January 2021
- Published by SAGE Publications in Ear, Nose & Throat Journal
- Vol. 102 (1) , 52-57
Validity of a New Prediction Model to Identify Patients at Risk for Obstructive Sleep Apnea Hypopnea Syndrome
Objectives: To develop and validate a new clinical prediction model for screening patients at risk for obstructive sleep apnea–hypopnea syndrome (OSAHS). Methods: This study used 2 data sets to develop and validate the model. To build the model, the first data set comprised 892 patients who had diagnostic polysomnography (PSG); data were assessed by multivariate logistic regression analysis. To validate the new model, the second data set comprised 374 patients who were enrolled to undergo overnight PSG. Receiver operating characteristic analysis and all predictive parameters were validated. Results: In the model development phase, univariate analysis showed 6 parameters were significant for prediction apnea–hypopnea index ≥15 events/hour: male sex, choking or apnea, high blood pressure, neck circumference >16 inches (female) or 17 inches (male), waist circumference ≥80 (female) or 90 cm (male), and body mass index >25 kg/m2. Estimated coefficients showed an area under the curve of 0.753. In the model validation phase, the sensitivity and specificity were approximately 93% and 26%, respectively, for identifying OSAHS. Comparison with the Epworth Sleepiness Scale score of ≥10 and STOP-Bang score ≥3 showed sensitivity of 42.26% and 56.23%, respectively, for detecting patients at risk. Conclusions: This new prediction model gives a better result on identifying patients at risk for OSAHS than Epworth Sleepiness Scale and STOP-Bang in terms of sensitivity. Moreover, this model may play a role in clinical decision-making for a comprehensive sleep evaluation to prioritize patients for PSG.
- 4 January 2021
- Published by SAGE Publications in Ear, Nose & Throat Journal
- Vol. 102 (1) , 52-57
Open AccessValidity of a New Prediction Model to Identify Patients at Risk for Obstructive Sleep Apnea Hypopnea Syndrome
Objectives: To develop and validate a new clinical prediction model for screening patients at risk for obstructive sleep apnea–hypopnea syndrome (OSAHS). Methods: This study used 2 data sets to develop and validate the model. To build the model, the first data set comprised 892 patients who had diagnostic polysomnography (PSG); data were assessed by multivariate logistic regression analysis. To validate the new model, the second data set comprised 374 patients who were enrolled to undergo overnight PSG. Receiver operating characteristic analysis and all predictive parameters were validated. Results: In the model development phase, univariate analysis showed 6 parameters were significant for prediction apnea–hypopnea index ≥15 events/hour: male sex, choking or apnea, high blood pressure, neck circumference >16 inches (female) or 17 inches (male), waist circumference ≥80 (female) or 90 cm (male), and body mass index >25 kg/m2. Estimated coefficients showed an area under the curve of 0.753. In the model validation phase, the sensitivity and specificity were approximately 93% and 26%, respectively, for identifying OSAHS. Comparison with the Epworth Sleepiness Scale score of ≥10 and STOP-Bang score ≥3 showed sensitivity of 42.26% and 56.23%, respectively, for detecting patients at risk. Conclusions: This new prediction model gives a better result on identifying patients at risk for OSAHS than Epworth Sleepiness Scale and STOP-Bang in terms of sensitivity. Moreover, this model may play a role in clinical decision-making for a comprehensive sleep evaluation to prioritize patients for PSG.
- 1 September 2013
- Vol. 96 (9) , 1159-63
Comparison of the clinical outcomes between unattended home APAP and polysomnography manual titration in obstructive sleep apnea patients.
To compare the clinical outcomes and determine the difference in therapeutic pressure between Automatic positive airway pressure (APAP) and polysomnography manual titration. Fifty patients of obstructive sleep apnea (OSA), moderate to severe cases, were randomized into two groups of intervention: 95-percentile pressure derived from APAP titration and an optimal pressure derived from manual titration. Clinical outcomes were assessed before and after four weeks. The average 95-percentile pressure derived from APAP titration was 11.7 +/- 0.3 cmH2O with median mask leak 1.3 L/min. The average optimal pressure derived from manual titration was 8.2 +/- 0.3 cmH2O. Pearson correlation analysis showed weak positive correlation (r = 0.336, p = 0.017). The Epworth Sleepiness Score (ESS), Quality of life tests: PSQI (Pittsburg Sleep Quality Index), and SF-36 (Medical Outcomes Study 36-Item Short-Form Health Survey) were improved significantly in both groups, but there were no statistical significant differences between groups. An APAP titration is an effective method of pressure determination for conventional CPAP therapy and shows no difference in clinical outcomes comparing the standard titration.
- 1 September 2013
- Vol. 96 (9) , 1159-63
Comparison of the clinical outcomes between unattended home APAP and polysomnography manual titration in obstructive sleep apnea patients.
To compare the clinical outcomes and determine the difference in therapeutic pressure between Automatic positive airway pressure (APAP) and polysomnography manual titration. Fifty patients of obstructive sleep apnea (OSA), moderate to severe cases, were randomized into two groups of intervention: 95-percentile pressure derived from APAP titration and an optimal pressure derived from manual titration. Clinical outcomes were assessed before and after four weeks. The average 95-percentile pressure derived from APAP titration was 11.7 +/- 0.3 cmH2O with median mask leak 1.3 L/min. The average optimal pressure derived from manual titration was 8.2 +/- 0.3 cmH2O. Pearson correlation analysis showed weak positive correlation (r = 0.336, p = 0.017). The Epworth Sleepiness Score (ESS), Quality of life tests: PSQI (Pittsburg Sleep Quality Index), and SF-36 (Medical Outcomes Study 36-Item Short-Form Health Survey) were improved significantly in both groups, but there were no statistical significant differences between groups. An APAP titration is an effective method of pressure determination for conventional CPAP therapy and shows no difference in clinical outcomes comparing the standard titration.