ABSTRACT

Multiple organs in human body make up human organ system among which heart plays an important role for correct functioning of human body. Usually heart disease is recognized when the situation is severe or when the person is attacked by heart attack. The cost to cure the disease at this stage is high. Hence, it is important to implement a system for early stage detection of heart abnormality with low cost. This paper proposes a hybrid deep learning model using convolution neural network and support vector machine for detection of heart Dysfunctionality through iris. Considering the performance of the proposed model which is evaluated using different evaluation metrics such as precision, recall, f-score & accuracy, it is found that the accuracy of the proposed hybrid model is outperforming than exiting.

CONCLUSION

In this paper, a hybrid model using convolution neural network and support vector machine is presented to establish a relationship between iris of human eye and the detection of heart abnormality (Normal/Abnormal). The proposed identifies the abnormality in heart in an early stage with minimum cost. The results stated that using the hybrid CNN and SVM model the process of feature extraction is made automatic which was hand engineered in traditional and machine learning models and the proposed hybrid model obtained an accuracy of 80%. The data available on this methodology of detecting diseases is minimum and different deep learning architectures can be considered for feature extraction and classification processes which are considered as the future tasks in development of the proposed hybrid convolution neural network and support vector machine.

Authors:

Kadamati Dileep Kumar1
Asstant Professor
Dept of Computer Science and Engineering,
Sri Sivani College of engineering, Srikakulam
KamalakaraRao Vankala2
Dept of Computer Science and Engineering,
Sri Sivani College of Engineering, Srikakulam
Dr. T.PanduRanga Vital3
Associate Professor
Dept of Computer Science and Engineering,
Aditya Institute of Technology and Management Tekkali, Srikakulam

Download full abstract: An_Efficient_Method_for_Detection_of_Dys