In recent days, health problems are becoming more prevalent because of changes in lifestyles and inherited factors. Heart disease (HD), in particular, has become increasingly widespread in recent years, putting people’s lives in jeopardy. Blood pressure, cholesterol, and pulse rate are all varied for each person. Normal blood pressure should be 120/90, cholesterol should be 100–129[Formula: see text]mg/dL, pulse rate should be 72, fasting blood sugar level should be 100[Formula: see text]mg/dL, heart rate should be 60–100[Formula: see text]bpm, and ECG should be normal, according to medically established data. The aorta is 25[Formula: see text]mm (1 inch) wide, whereas capillaries are only 8[Formula: see text][Formula: see text]m wide, which signifies HD. This paper is based on a public health dataset that also contains a cardiac dataset. In our previous work, a new conventional neural network (CNN) architecture was used to extract and categorize histopathological images using the [Formula: see text]-means consensus clustering. We achieved good results with the cardiac dataset compared to the existing results. The outcome of the proposed work achieves a precision rate of 97%. In this paper, a novel conventional neural network (CNN) architecture was utilized to identify and characterize histopathology pictures with the help of using the consensus [Formula: see text]-nearest neighbor algorithm (CKNN). The usage of a deep neural network, as well as the selection and recovery of functionality, is an essential step until the dataset is classified. As a result, the training process has pre-formed the dataset. The system evaluates the person’s symptoms as input and provides the disease’s possibility as an output. The suggested approach is linked to temporal data modeling and makes use of a prior HD CNN prediction. In comparison to previous outcomes, we had good outcomes with the current cardiac dataset. The proposed model’s conclusion has brought out accuracy of 99.1%.
