Improvement in the Prediction of Ventilator Weaning Outcomes by an Artificial Neural Network in a Medical ICU

Kuo, Hung Ju; Chiu, Hung Wen; Lee, Chun Nin; Chen, Tzu Tao; Chang, Chih Cheng; Bien, Mauo Ying

doi:10.4187/respcare.03648

Cited by 60 publications

(41 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…94 For example, neural networks have been used successfully for breathing-pattern recognition in critical care, weaning from mechanical ventilation, and ICU outcomes prediction. [95][96][97] Likewise, neural networks may be able to recognize other types of respiratory patterns during periods of poor patient-ventilator interaction. A neural network may be fed with respiratory waveforms (flow, pressure, or both) and trained to recognize normal and asynchronous breaths.…”

Section: Monitoring Asynchronies In the Era Of Precision Medicinementioning

confidence: 99%

Minimizing Asynchronies in Mechanical Ventilation: Current and Future Trends

et al. 2018

View full text Add to dashboard Cite

Patient-ventilator asynchrony exists when the phases of breath delivered by the ventilator do not match those of the patient. Asynchronies occur throughout mechanical ventilation and negatively affect patient comfort, duration of mechanical ventilation, length of ICU stays, and mortality. Identifying asynchronies requires careful attention to patients and their ventilator waveforms. This review discusses the different types of asynchronies, how they are generated, and their impact on patient comfort and outcome. Moreover, it discusses practical approaches for detecting, correcting, and preventing asynchronies. Current evidence suggests that the best approach to managing asynchronies is by adjusting ventilator settings. Proportional modes improve patient-ventilator coupling, resulting in greater comfort and less dyspnea, but not in improved outcomes with respect to the duration of mechanical ventilation, delirium, or cognitive impairment. Advanced computational technologies will allow smart alerts, and models based on time series of asynchronies will be able to predict and prevent asynchronies, making it possible to tailor mechanical ventilation to meet each patient's needs throughout the course of mechanical ventilation.

show abstract

Section: Monitoring Asynchronies In the Era Of Precision Medicinementioning

confidence: 99%

Minimizing Asynchronies in Mechanical Ventilation: Current and Future Trends

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Mueller et al used neural networks, support vector machines, Bayesian classifiers, decision trees, and logistic regression to find the classifier best suited for predicting the outcome of weaning an infant off MV [12]. Kuo et al used a similar neural network approach to predict successful extubation, the removal of the inserted tube, in medical ICU patients who were mechanically intubated [13]. Prasad et al built on that research and used reinforcement learning to come up with a strategy for weaning patients off of MV in the ICU [14].…”

Section: Related Workmentioning

confidence: 99%

Predicting and Understanding Unexpected Respiratory Decompensation in Critical Care Using Sparse and Heterogeneous Clinical Data

Ren

Johnson

Lehman

et al. 2018

2018 IEEE International Conference on Healthcare Informatics (ICHI)

View full text Add to dashboard Cite

Hospital intensive care units (ICUs) care for severely ill patients, many of whom require some form of organ support. Clinicians in ICUs are often challenged with integrating large volumes of continuously recorded physiological and clinical data in order to diagnose and treat patients. In this work, we focus on developing interpretable models for predicting unexpected respiratory decompensation requiring intubation in ICU patients. Predicting need for intubation could have important implications for the patient and medical staff and potentially enable timely interventions for improved patient outcome. Using data from adult ICU patients from the Medical Information Mart for Intensive Care (MIMIC)-III database, we developed gradient boosting models for predicting intubation onset. In a cohort of 12,470 patients, of whom 1,067 were intubated (8.55%), we achieved an area under the receiver operating characteristic curve (AUROC) of 0.89, with 95% confidence interval (CI) 0.87-0.91, when predicting intubation 3 hours ahead of time, a significant increase (p<0.001) over the AUROC achieved using several baselines, including logistic regression (0.81, 95% CI 0.78-0.84) and neural networks (0.80, 95% CI 0.77-0.83]). Finally, we conducted feature importance analysis using gradient boosting and derived useful insights in understanding the relative importance of clinical vs. biological variables in predicting impending respiratory decompensation in ICUs.

show abstract

“…They have also been successfully used to predict mortality in trauma patients [ 10 ]. Recently, ANNs have been introduced to predict extubation outcomes, but findings vary by study [ 11 , 12 ]. The main reasons for poor outcome predictions might be because of differences in clinical input data.…”

Section: Introductionmentioning

confidence: 99%

An Artificial Neural Network Model for Predicting Successful Extubation in Intensive Care Units

Hsieh

Chen

et al. 2018

JCM

View full text Add to dashboard Cite

Background: Successful weaning from mechanical ventilation is important for patients in intensive care units (ICUs). The aim was to construct neural networks to predict successful extubation in ventilated patients in ICUs. Methods: Data from 1 December 2009 through 31 December 2011 of 3602 patients with planned extubation in Chi-Mei Medical Center’s ICUs was used to train and test an artificial neural network (ANN). The input was 37 clinical risk factors, and the output was a failed extubation prediction. Results: One hundred eighty-five patients (5.1%) had a failed extubation. Multivariate analyses revealed that failure was positively associated with therapeutic intervention scoring system (TISS) scores (odds ratio [OR]: 1.814; 95% Confidence Interval [CI]: 1.283–2.563), chronic hemodialysis (OR: 12.264; 95% CI: 8.556–17.580), rapid shallow breathing (RSI) (OR: 2.003; 95% CI: 1.378–2.910), and pre-extubation heart rate (OR: 1.705; 95% CI: 1.173–2.480), but negatively associated with pre-extubation PaO2/FiO2 (OR: 0.529; 95%: 0.370–0.750) and maximum expiratory pressure (MEP) (OR: 0.610; 95% CI: 0.413–0.899). A multilayer perceptron ANN model with 19 neurons in a hidden layer was developed. The overall performance of this model was F1: 0.867, precision: 0.939, and recall: 0.822. The area under the receiver operating characteristic curve (AUC) was 0.85, which is better than any one of the following predictors: TISS: 0.58 (95% CI: 0.54–0.62; p < 0.001); 0.58 (95% CI: 0.53–0.62; p < 0.001); and RSI: 0.54 (95% CI: 0.49–0.58; p = 0.097). Conclusions: The ANN performed well when predicting failed extubation, and it will help predict successful planned extubation.

show abstract

Improvement in the Prediction of Ventilator Weaning Outcomes by an Artificial Neural Network in a Medical ICU

Cited by 60 publications

References 44 publications

Minimizing Asynchronies in Mechanical Ventilation: Current and Future Trends

Minimizing Asynchronies in Mechanical Ventilation: Current and Future Trends

Predicting and Understanding Unexpected Respiratory Decompensation in Critical Care Using Sparse and Heterogeneous Clinical Data

An Artificial Neural Network Model for Predicting Successful Extubation in Intensive Care Units

Contact Info

Product

Resources

About