Classification Patient-Ventilator Asynchrony with Dual-Input Convolutional Neural Network

Chong, Thern Chang; Loo, Nien Loong; Chiew, Yeong Shiong; Nor, Mohd Basri Mat; Ralib, Azrina

doi:10.1016/j.ifacol.2021.10.276

Cited by 8 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…What we find from Table 12 is that our semi-supervised AD approach, when combined with oversampling, is competitive with the existing supervised techniques, even though we are training the AD only on normal data. In addition, while Chong et al [69] build their model directly from the flow and pressure data streams, the other two extract additional metadata from it, in the form of derived quantities (tidal volume, etc. [39]) or shape features extracted from the waveforms [56].…”

Section: Pva Detectionmentioning

confidence: 99%

Deep-Shallow Metaclassifier with Synthetic Minority Oversampling for Anomaly Detection in a Time Series

Reshadi,

Li,

et al. 2024

Algorithms

View full text Add to dashboard Cite

Anomaly detection in data streams (and particularly time series) is today a vitally important task. Machine learning algorithms are a common design for achieving this goal. In particular, deep learning has, in the last decade, proven to be substantially more accurate than shallow learning in a wide variety of machine learning problems, and deep anomaly detection is very effective for point anomalies. However, deep semi-supervised contextual anomaly detection (in which anomalies within a time series are rare and none at all occur in the algorithm’s training data) is a more difficult problem. Hybrid anomaly detectors (a “normal model” followed by a comparator) are one approach to these problems, but the separate loss functions for the two components can lead to inferior performance. We investigate a novel synthetic-example oversampling technique to harmonize the two components of a hybrid system, thus improving the anomaly detector’s performance. We evaluate our algorithm on two distinct problems: identifying pipeline leaks and patient-ventilator asynchrony.

show abstract

Section: Pva Detectionmentioning

confidence: 99%

Deep-Shallow Metaclassifier with Synthetic Minority Oversampling for Anomaly Detection in a Time Series

Reshadi,

Li,

et al. 2024

Algorithms

View full text Add to dashboard Cite

show abstract

“…In particular, this data can be further processed to provide respiratory mechanics and other ventilatory information not available on today’s ventilators, but useful to personalise and guide MV treatment [12] , [13] , [14] , [15] , [16] . CAREDAQ thus provides a platform for future development and integration of software modules, including machine learning models [17] , [18] , [19] or model-based algorithms [14] , [16] , [20] , [21] which could potentially provide real-time, proactive, and patient-specific MV decision support and care, improving care and outcomes.…”

Section: Hardware In Contextmentioning

confidence: 99%