IJBET 2017 Volume 6 Issue 1
International Journal of BioEngineering and Technology (IJBET) ISSN: 0976 - 2965
An Open Access Journal -- NO Fees -- NO Processing Charges -- 100% Non Profit Initiatives
Heart Patient Pathology Assessment using ANFIS with ECG & ECHO Information. S.Ananthi, V.Vignesh and K.Padmanabhan. IJBET (2017), 6(1):1-12
FULL TEXT SWF:
Heart Patient Pathology Assessment using ANFIS with ECG & ECHO Information
Authors & Affiliation:
S.Ananthi1, V.Vignesh2 and K.Padmanabhan3
1 Associate Professor, Department of Network Systems and Information Technology, University of Madras, India.
2 V.Vignesh, Research Scholar, Department of Network Systems and Information Technology, University of Madras, India.
3 K.Padmanabhan, Emeritus Professor, A.C College of Technology, Anna University, Chennai, India.
There are many diagnostic procedures for cardiac pathology assessment today. ECG, Echo cardiography, Doppler Blood flow parameters, heart valve conditions, hypertrophy, X-ray image, angiogram etc. are some of these. In a continued assessment of a heart patient under long term observation with medication, some of the tests are repeated periodically to determine the prognosis, mainly the ECG and Echo. From these, it is required to assess the risk factor of the patient and determine if there is improvement or otherwise, so that the medication can be altered. Presently, this is based merely on the judgment of the cardiologist physician. Herein, a rather precise soft computing technique for correct assessment of the risk factor from these diagnostic tests is described. The use of Fuzzy logic based inferences are proven techniques even in many control and automation fields. Therefore, such a technique will be helpful in a precise determination of the health condition of the heart. The ANFIS is a combined Neural network cum fuzzy inference technique. It can deal with data from the ECG and its wave segments and the Echo Doppler information for this assessment.
Key words: Electrocardiography (ECG), Echocardiography, Medical Diagnostic Imaging, Fuzzy logic, Fuzzy Neural Networks, Takagi-Sugeno Model.