The applications of IoT have been employed in diverse domains like industries, clinical care, and farming, and so forth. Nowadays, the constitution of this technology is more prevalent in clinical observation, where the wearable devices have stimulated the development of the Internet of Medical Things (IoMT). In the process of reducing the death rate, it is necessary to detect the disease at an earlier stage. The cardiac disease prediction is a major defect in the examination of the dataset in clinics. The research proposed aims to recognize the important cardiac complaint prediction characteristics by utilizing machine‐learning methodologies. Numerous projects have been established regarding the diagnosis of cardiac complaints, which results in low accuracy rate. Thus, for improving the accuracy of prediction and for cardiac complaint investigation this article utilized a fuzzy c‐means neural network (FNN) and a deep convolution neural network for feature extraction. From the clinical dataset, data were obtained for the risk prediction of cardiac complaints that includes blood pressure (BP), age, sex, chest pain, cholesterol, blood sugar, and so forth. The hearts condition is recognized by categorizing the sensor data received by FNN. The evaluation performances were carried out and the results revealed that FNN is good in predicting the cardiac complaints. In addition to this, the proposed model achieves better accuracy than the other approaches through the demonstration of simulation results. The proposed approach attains the accuracy rate of 86.4% and F1‐score of 97%, precision 76.2%, and 64.6% of FPR.