透過您的圖書館登入
IP:54.166.96.191
  • 學位論文

應用類神經網路提升內科加護病患脫離呼吸器的預測能力

The prediction of ventilator weaning outcome improved by artificial neural network in medical intensive care unit

指導教授 : 邱泓文
共同指導教授 : 邊苗瑛

摘要


臨床上常以淺快呼吸指標(rapid shallow breathing index, RSBI)與自發性呼吸測試(spontaneous breathing trail, SBT)的結果作為病患能否脫離之參考。然而,淺快呼吸指標的閾值容易受不同病患族群與測量方法而有所不同、25 ~ 40%的病患在通過自發性呼吸測試後仍無法脫離呼吸器,目前研究顯示沒有一個合適且方便的脫離指標或測量方式足以做協助臨床人員預測病患脫離呼吸器預後之依據。類神經網路(artificial neural networks, ANNs)可藉由外在或內在輸入變數進行學習,進而更改架構中的權重與預測結果,為機器學習的模型之一,常被用於醫療決策支持系統(medical decision support systems)的建構。本研究期望設計、建立類神經網路預測模型,以改善臨床預測病患移除人工氣道預後的準確率。本研究收集63位與32位入住加護病房、插管使用呼吸器的內科病患,隨機以80%作為訓練組(n = 76)、20%為測試組(n = 19)建立並測試類神經網路模型。選用病患的年齡、呼吸器使用原因、呼吸器使用時間、急性生理和長期健康評估表與於30分鐘壓力支持型通氣(pressure support ventilation, PSV) 5 cmH2O/吐氣末陽壓(positive end-expiratory pressure, PEEP) 5 cmH2O下進行自發性呼吸測試中所測得之平均吸氣時間、平均吐氣時間、平均呼吸次數及平均潮氣容積等8項指標作為類神經網路模型的輸入參數,並以接受者操作特性曲線與分類模糊矩陣(confusion matrix)比較類神經網路與臨床常規一分鐘淺快呼吸指標的預測能力。研究結果發現:藉由類神經網路訓練模型的訓練,相較於淺快呼吸指標的閾值設為105 breaths/min/L時,其接受者操作特性曲線下面積(0.95 vs. 0.51, p < 0.05)、敏感度(91.7% vs. 75%)、特異度(85.7% vs. 14.3%)與準確率(89.5% vs. 52.6%),顯示能顯著提升預測呼吸器脫離結果的能力。本研究利用8項參數所建立的類神經網路預測模型可改善臨床預測病患是否可以移除人工氣道的能力,期望藉由其臨床應用,可協助醫療人員選擇移除人工氣道的適當時機,降低呼吸器不必要的延長使用及過早移除人工氣道的機會,進而降低其併發症的發生與醫療費用的支出。

並列摘要


The rapid shallow breathing index (RSBI) and the result of spontaneous breathing trail (SBT) are commonly used clinically for predicting the outcome of weaning from mechanical ventilation. However, there are existed different thresholds and sensitivities of RSBI among different populations and measurement conditions, between 25 and 40% patients passed SBT develop failure signs after weaning from mechanical ventilation. There is no single appropriate and convenient predictor or method can be used to help the clinicians to predict the weaning outcome. The artificial neural networks (ANNs) are the machine-learning models which can change their structures and outputs based on the external or internal information during the learning phase. They had been applied in modeling the medical decision support systems. The purpose of this study was to design an ANN model for predicting the weaning outcome of mechanically ventilated patients. Sixty-three and thirty-two ready for weaning patients living in medical intensive care unit were recruited and randomly divided into training (80%, n = 76) and testing groups (20%, n = 19). Eight key features including age, reasons for intubation, duration of using mechanical ventilator, APACHE II score, the mean of inspiratory time, the mean of expiratory time, the mean of respiratory rate and the mean of tidal volume in thirty minutes spontaneous breathing trail under PSV 5 cmH2O/PEEP 5 cmH2O ventilator setting were selected as the ANN input variables. The performance of ANN model was compared with 1-minute RSBI measurement method by using confusion matrix and the receiver operating characteristic curve. The area under receiver operating characteristic curve of ANN model was 0.95 and that for the RSBI method was 0.51 when the threshold was set to 105 breaths/min/L. Predictions by the testing group of ANN model had a sensitivity of 91.7%, a specificity of 85.7%, and an accuracy rate of 89.5%, compared with 75%, 14.3% and 52.6%, respectively, for the RSBI method. In summary, the ANN model improved the accuracy for prediction of weaning outcome. By applying this ANN model clinically, the clinicians could select the appropriate weaning time as early as possible, which could decrease the chance of unnecessary prolonged ventilatory support and premature weaning. Therefore, the incidence of patients’ complication rate and medical cost related to ventilator support will decrease.

參考文獻


[1] Epstein, S. (2003). Weaning from ventilatory support. In J. Crapo, Glassroth, J, Karlinsky, J, King, TE (Ed.), Textbook of Pulmonary Diseases (7 ed., pp. 1089): Philadelphia : Lippincott Williams & Wilkins.
[4] MacIntyre, N. R. (2001). Evidence-based guidelines for weaning and discontinuing ventilatory support: a collective task force facilitated by the American College of Chest Physicians; the American Association for Respiratory Care; and the American College of Critical Care Medicine. Chest, 120(90060), 375S-396S.
[5] Seymour, C. W., Martinez, A., Christie, J. D., & Fuchs, B. D. (2004). The outcome of extubation failure in a community hospital intensive care unit: a cohort study. Crit Care, 8(5), R322-R327.
[7] Girard, T. D., & Ely, E. W. (2008). Protocol-driven ventilator weaning: reviewing the evidence. Clin. Chest Med., 29(2), 241-252.
[8] Conti, G., Montini, L., Pennisi, M. A., Cavaliere, F., Arcangeli, A., Bocci, M. G., et al. (2004). A prospective, blinded evaluation of indexes proposed to predict weaning from mechanical ventilation. Intensive Care Med, 30(5), 830-836.

延伸閱讀