ORIGINAL PAPER Heart disease classification using hybridized Ruzzo-Tompa memetic based deep trained Neocognitron neural network J. Vijayashree 1 & H. Parveen Sultana 1 Received: 5 October 2018 /Accepted: 27 December 2018 # IUPESM and Springer-Verlag GmbH Germany, part of Springer Nature 2019 Abstract According to the survey 17.5 million deaths are happened due to the cardiovascular disease that leads to create heart attack, chest pain and stroke. Based on the survey it clearly concludes that most of the people affected by heart problem that need to be identified in the earlier stage for eliminating the future risk in patient health. The importance of the heart disease detection process helps to create the earlier detection system for identifying heart problem by using machine learning and optimized techniques but the developed forecasting systems are difficult to predict the heart problems in an accurate manner with minimum time. So, hybridized Ruzzo–Tompa memetic based deep trained Neocognitron neural network is introduced to analyze the heart disease related features and predict the heart disease in earlier stage. First, heart disease data has been collected from UCI repository, dimensionality of the data is minimized by hybridized Ruzzo–Tompa memetic approach. After reducing the number of features, that are trained by deep learning approach which analyze the features using maximum number of hidden layers that used to predict heart disease features successfully while making the Neocognitron neural network classification. Further efficiency of the system is evaluated using MATLAB based simulation results. Keywords Heart disease . Hybridized Ruzzo–Tompa memetic based deep trained Neocognitron neural network . Heart disease data set-UCI repository 1 Introduction Heart disease [1] also named as the congenital heart disease or acquired heart disease which means human heart cannot func- tion normally. The abnormal function of heart is identified by several symptoms such as feeling weak, swelling legs, breath- ing trouble, chest pain, cyanosis and palpitations. These symp- toms are varied person by person also sometimes these symp- toms are difficult to recognize in the beginning stage [2]. Due to the low level of symptoms most of the people died, for example 4 out of 10 people were died due to the heart problem [3] especially in United States, Wales, Canada and England. Especially, in United States, for every one minute one person is died because of heart problems and 630,000 peoples are died for every year. According to the survey of world health organization, heart disease creates serious risk factors [4] in upcoming years due to the smoking habits, physical activities, blood pressure, diet, junk food, obesity, family history, over- weigh and blood cholesterol. Not only these habits, around 80 to 90% of the people suffered by heart disease due to the genetic factors which may related to several disease such as myocardial infarction, angina, cardiovascular disease, con- genital heart problem and coronary disease. Even though the heart problem creates serious issues, it could be resolved [5] in earlier stage by two ways such as medication and surgery. The medication process includes in aspirin, beta blocker, stains and so on. The surgery process also has replacement of wrong heart valves, angioplasty, bypass surgery and pacemaker. But these two treatment process also need to handle the difficulties which means, detection of heart disease might be very slow that also leads to death and take longer time to recover the heart problem. For avoiding these issues, heart problem might be recognized in the beginning stage without making any fault. So, several automatic system [6] has been created in medical field by utilizing data mining, machine learning This article is part of the Topical Collection on Internet of Medical Things in E-Health Hassan Fouad Mohamed- El-Sayed and M. Hemalatha * J. Vijayashree vijayashree.j@vit.ac.in 1 School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, India Health and Technology https://doi.org/10.1007/s12553-018-00292-2