Local Robust Gradient Patterns for Recognition of Cardiomyopathy
P. Megana Santhoshi1, Mythili Thirugnanam2

1P.Megana Santhoshi,* PhD Scholar, School of Computer Science Engineering, (SCOPE), Vellore Institute of Technology, Vellore, (Tamil Nadu), India.
2Mythili Thirugnanam, Associate Professor Senior, School of Computer Science Engineering, (SCOPE), Vellore Institute of Technology, Vellore, (Tamil Nadu), India.
Manuscript received on November 20, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP: 1416-1422 | Volume-9 Issue-2, December, 2019. | Retrieval Number:  B3331129219/2020©BEIESP | DOI: 10.35940/ijeat.B3331.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Cardiomyopathy is one of the heart diseases that cause chamber damages. The impact of heart disease ends up in unforeseen fall with light-headedness. IoT plays an important role in human healthcare systems. Through IoT, it’s terribly simple to watch the health condition of the heart disease patient by detection the abnormality within the electrocardiogram signal generated by IoT sensors. The varied ECG signals represent the severity of the heart disease and every graphical record signal has distinctive patterns. This paper describes the recognition of cardiomyopathy disease based on local robust gradient patterns technique LBP operator is one of the foremost powerful techniques to recognize the patterns within the ECG graph signals. But it’s highly sensitive to noise and little fluctuations. To beat these limitations LTP and its derivatives are applied. LTP operator removes the noise by dividing the signals into 3 regions. It doesn’t provide fruitful results if the signal has an additional range of peaks and valleys. Merely it replaces peaks by the valley and vice-versa. RLTP technique is appropriate to beat this limitation by finding the minimum value of LTP and its complement value. However, it fails for little fluctuation in the signals. To enhance the recognition rate of little fluctuation graphical record signals the discriminant robust local ternary pattern technique is proposed by multiplying the edge gradient values with RLTP techniques. This method is applied to PTB information and therefore the Experimental results are created within the variety of tables and graphs. The proposed technique has high results on the LTP and its derivative methods and is useful for detecting cardiomyopathy with 85% accuracy.
Keywords: LBP, LTP, RLTP, DRLTP, Cardiomyopathy, PTB, DCM, HCM.