A comparison of the ECG classification performance of different feature sets
Metadata:Show full item record
Citation:O'Dwyer, M., deChazal, P., Reilly, R.B., A comparison of the ECG classification performance of different feature sets: proceedings. of the Computer in Cardiology 2000 Conference, Boston: IEEE, 2000, pp327-330
This study investigates the automatic classification of the Frank lead ECG into different disease categories. A comparison of the performance of a number of different feature sets is presented. The feature sets considered include wavelet-based features, standard cardiology features, and features taken directly from time-domain samples of the EGG. The classification performance of each feature set was optimised using automatic feature selection and choosing the best classifier model from linear, quadratic and logistic discriminants. The ECG database used contains 500 cases classed into seven categories with 100% confidence. Using multiple runs of ten-fold cross-validation, the overall seven-way accuracy of different feature sets and classifier model combinations ranged between 60% and 75%. The best performing classifier used linear discriminants processing selected time-domain features. This is also found to be the simplest and fastest classifier to implement
Other Titles:Computer in Cardiology 2000 Conference
Keywords:ECG classification performance ECG database Frank lead ECG automatic classification automatic feature selection classifier model disease categories feature sets linear discriminants logistic discriminants multiple runs quadratic discriminants selected time-domain features seven-way accuracy standard cardiology features ten-fold cross-validation time-domain samples wavelet-based features
Showing items related by title, author, creator and subject.
KOKARAM, ANIL CHRISTOPHER; RING, DANIEL (IEEE Computer Society, 2009)Accurately segmenting objects in video is a difficult and time consuming process in modern post-production houses. Automatic systems may work for a small number of frames, but will typically fail over longer video shots. ...
[PEER Project Non-TCD]<italic>DICER1</italic> syndrome ? clarifying the diagnosis, clinical features and management implications of a pleiotropic tumor predisposition syndrome Rahman, Nazneen; Slade, Ingrid; Bacchelli, Chiara; Davies, Helen; Murray, Anne; Abbaszadeh, Fatemeh; Hanks, Sandra; Barfoot, Rita; Burke, Amos; Chisholm, Julia; Hewitt, Martin; Jenkinson, Helen; King, Derek; Morland, Bruce; Pizer, Barry; Prescott, Katrina; Saggar, Anand; Side, Lucy; Traunecker, Heidi; Vaidya, Sucheta; Ward, Paul; Futreal, P. Andrew; Vujanic, Gordan; Nicholson, Andrew; Sebire, Neil; Turnbull, Clare; Priest, John; Pritchard-Jones, Kathryn; Houlston, Richard; Stiller, Charles; Stratton, Michael; Douglas, Jenny (BMJ Publishing Group Ltd.BMA House, Tavistock Square, London UK, 2011-01-25)Abstract Background: Constitutional DICER1 mutations were recently reported to cause familial pleuropulmonary blastoma (PPB). We sought to investigate the contribution and phenotypic spectrum of constitutional ...
[PEER Project Non-TCD]An eXtended Manufacturing Integrated System for feature based manufacturing with STEP-NC Laguionie, Raphael; Rauch, Matthieu; Hascoet, Jean-Yves (Taylor & Francis, 2011-07-29)Abstract CNC feature-based programming with STEP-NC standard extends the collaborative model of manufacturing data exchange all along the numerical data chain. This paper considers the mutations related to this ...