Decision Support Systems for Cardio Vascular Disease using Deep Learning Techniques

Abstract

On thisgraduationproject,weinvestigatetheuseofDSSforcardiovasculardiseasesdiag- nostic, focusingonmultipletypesofarrhythmia.wealsoexploretheuseofLSTMasaclas- sification modelusinglongertermECGsignal(10s),aswellasbuildingaplatformtodisplay and acquirenewsignalsusingIoTdevice(MySignals). The useofLSTMprovestobeperformingverywellonthetaskofclassifying17categories of rhythm(15arrhythmia+normalsinusrhythm+pacemakerrhythm)comparedtootherstate- of-the-art methods,yieldinganaccuracyof93%,sensitivity95%,andaspecificityof99.46%. As wellasareal-timeresponsiveplatformallowingthevisualizationofnewlyobtainedsignals and outputofthecomputer-aideddiagnostic. This projectdepictsthatthetopicofstudystillfacesmanyobstacles,challengesandlim- itations suchas(clinicalimplication,healthydata...etc)andothers.Despiteallthelater,the LSTM anddeeplearningmechanismsdemonstratedthatitiscapableofrecognizinglong-term patterns withinourtime-seriesdataalongwithgeneralizingthediagnosticprocess.