Abstract
Intensive Care Units (ICUs) are equipped with many sophisticated sensors and monitoring devices to provide the highest quality of care for critically ill patients. However, these devices might generate false alarms that reduce standard of care and result in desensitization of caregivers to alarms. Therefore, reducing the number of false alarms is of great importance. Many approaches such as signal processing and machine learning, and designing more accurate sensors have been developed for this purpose. However, the significant intrinsic correlation among the extracted features from different sensors has been mostly overlooked. A majority of current data mining techniques fail to capture such correlation among the collected signals from different sensors that limits their alarm recognition capabilities. Here, we propose a novel information-theoretic predictive modeling technique based on the idea of coalition game theory to enhance the accuracy of false alarm detection in ICUs by accounting for the synergistic power of signal attributes in the feature selection stage. This approach brings together techniques from information theory and game theory to account for inter-features mutual information in determining the most correlated predictors with respect to false alarm by calculating Banzhaf power of each feature. The numerical results show that the proposed method can enhance classification accuracy and improve the area under the ROC (receiver operating characteristic) curve compared to other feature selection techniques, when integrated in classifiers such as Bayes-Net that consider inter-features dependencies.
Highlights
As there is no single sensor/device capable of complying with all clinical requirements, multiple therapeutic and monitoring devices are often deployed in the Intensive Care Units (ICUs) to collect real-time data for diagnosis, prognosis, treatment and more generally, patient monitoring.These devices generate visual and acoustic alarms to inform nurses and physicians about changes in a patient’s condition or a failure in device functionality [1]
To suppress the false alarm in ICUs, here we develop a new coalition game-theoretical model based on Banzhaf power index that accounts for interdependency among the extracted features and their relevancy to the target class
In [29], we studied the problem of false alarm reduction in ICUs, where three main signals; electrocardiogram (ECG), plethysmogram (PLETH), arterial blood pressure (ABP), were used to classify alarms to false and true
Summary
As there is no single sensor/device capable of complying with all clinical requirements, multiple therapeutic and monitoring devices are often deployed in the Intensive Care Units (ICUs) to collect real-time data for diagnosis, prognosis, treatment and more generally, patient monitoring. Antink et al [13] applied band-pass filtering, peak detection, fast Fourier transform (FFT), principle component analysis (PCA), and some statistical analyses and extracted a number of features to train machine learning methods They applied four classifiers: random forest, SVM binary classification decision tree and regularized linear discriminant analysis for classifying alarms. Considering smaller coalitions of features resulted in reducing the accuracy of the alarm detection model To address these challenges, in this paper, we propose a new game-theoretic feature selection method based on utilizing Banzhaf power to declare salient features with comparable accuracy but much less complexity. In this paper, we propose a new game-theoretic feature selection method based on utilizing Banzhaf power to declare salient features with comparable accuracy but much less complexity This metric is proportional to the number of times that a feature is a critical player for a coalition.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
