Robustness of Support Vector Machine-based Classification of Heart Rate Signals

Kampouraki, A.; Nikou, Christophoros; Manis, George

doi:10.1109/iembs.2006.4397866

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Support Vector Machine (SVM) is a machine learning methodology of widespread use in classification, regression and ranking [ 27 , 28 ]. A state of art classification method with high accuracy and flexibility, SVM is used in bioinformatics and other disciplines to model data of varying source and meaning [ 28 ]; applicability in HRV measures processing has been documented [ 29 ]. In this study, SVM was used to develop a model able to predict target data values (presence or absence of a visual pursuit response) to which specific attributes (the HRV descriptors) could be related at the different steps of the training set for classification.…”

Section: Methodsmentioning

confidence: 99%

Visual pursuit response in the severe disorder of consciousness: modulation by the central autonomic system and a predictive model

et al. 2013

View full text Add to dashboard Cite

BackgroundA visual pursuit response is reportedly observed in ~20-30% of subjects in vegetative state (VS/UWS) and predicts better outcome; it is a key marker of evolution into the minimally conscious state (MCS). The probability of observing a positive response, however, has proven variable during the day, with comparable timing of the minima and maxima in VS/UWS and MCS. We verified if measures of sympathetic/parasympathetic balance are possible independent variables on which the occurrence of a pursuit response could depend and be predicted.MethodsFourteen subjects in VS/UWS and sixteen in MCS for more than one year were studied. A mirror was used to test the pursuit response for a total 231 useful trials. Non-invasive measures of the sympathetic/parasympathetic functional state (Heart rate variability descriptors nuLF and peakLF) used in the study of responsiveness in VS/UWS and MCS subjects were recorded and processed by descriptive statistics and advanced Support Vector Machine (SVM).ResultsA pursuit response was observed in 33% and 78.2% of subjects in VS or MCS, respectively. Incidence was higher at HRV nuLF values in the 20–60 range and peakLF values at 0.06-0.12 Hz (76.6%) and at nuLF values in the 10–60 range and peakLF values at 0.05-0.10 Hz (80.7%) in the VS and MCS, respectively. The SVM generated model confirmed the results in the training leave one out and 10 fold cross validation tests (81% and 81.4%).ConclusionThe pursuit response incidence depends to a relevant extent on the sympathetic/parasympathetic balance and autonomic functional state. Extensive monitoring appears advisable.

show abstract

Section: Methodsmentioning

confidence: 99%

Visual pursuit response in the severe disorder of consciousness: modulation by the central autonomic system and a predictive model

et al. 2013

View full text Add to dashboard Cite

show abstract

“…Support Vector Machine (SVM) was also applied in a research by Alty and his colleagues [33] to detect cardiovascular disease which confirmed the high accuracy as long as the noise is minimized. SVM is also applied in [34] for heart diseases with the same conclusion. Moreover, SVM also proved to be efficient in [35] in biochemical data but with computational complexity.…”

Section: Related Workmentioning

confidence: 99%

A Collaborative Evaluation Metrics Approach for Classification Algorithms

Idrees¹,

Alsheref²

2020

Journal of Southwest Jiaotong University

View full text Add to dashboard Cite

Evaluating Algorithms is one of the critical steps which should be strongly considered as this is the pillar of most of the decisions. This research proposes a novel method for accurate algorithms’ evaluation according to the metrics’ relationships and weight. The weight of the evaluation metrics is determined according to their invariance level. The proposed method validity is confirmed by applying and evaluating the most famous and well-populated classification techniques. The results have been considered according to the calculated weight of the evaluation measures to reveal the final algorithm evaluation. As a case study, the most suitable classification technique for Tinnitus data are explored. This research considered the Tinnitus data as Tinnitus symptoms are not clearly recognized which highlights the difficulty of the patients to have a direct and fast diagnosis which highlighted the motivation in investigating intelligent methods for fast Tinnitus diagnosing. The research applied the experiment on a real dataset that is gathered in Egypt and the results highlighted that the Support Vector Machine classification algorithm is the most suitable technique for Tinnitus data classification with an accuracy equal to 90.1%.

show abstract

Robustness of Support Vector Machine-based Classification of Heart Rate Signals

Cited by 2 publications

References 0 publications

Visual pursuit response in the severe disorder of consciousness: modulation by the central autonomic system and a predictive model

Visual pursuit response in the severe disorder of consciousness: modulation by the central autonomic system and a predictive model

A Collaborative Evaluation Metrics Approach for Classification Algorithms

Contact Info

Product

Resources

About