Arrhythmias are anomalies in the heartbeat rhythm that occur occasionally in people’s lives. These arrhythmias can lead to potentially deadly consequences, putting your life in jeopardy. As a result, arrhythmia identification and classification are an important aspect of cardiac diagnostics. An electrocardiogram (ECG), a recording collecting the heart’s pumping activity, is regarded the guideline for catching these abnormal episodes. Nevertheless, because the ECG contains so much data, extracting the crucial data from imagery evaluation becomes extremely difficult. As a result, it is vital to create an effective system for analyzing ECG’s massive amount of data. The ECG image from ECG signal is processed by some image processing techniques. To optimize the identification and categorization process, this research presents a hybrid deep learning-based technique. This paper contributes in two ways. Automating noise reduction and extraction of features, 1D ECG data are first converted into 2D pictures. Then, based on experimental evidence, a hybrid model called CNNLSTM is presented, which combines CNN and LSTM models. We conducted a comprehensive research using the broadly used MIT_BIH arrhythmia dataset to assess the efficacy of the proposed CNN-LSTM technique. The results reveal that the proposed method has a 99.10 percent accuracy rate. Furthermore, the proposed model has an average sensitivity of 98.35 percent and a specificity of 98.38 percent. These outcomes are superior to those produced using other procedures, and they will significantly reduce the amount of involvement necessary by physicians.

No abstract

No abstract

In recent years, health applications based on the internet of medical things have exploded in popularity in smart cities (IoMT). Many real-time systems help both patients and professionals by allowing remote data access and appropriate responses. The major research problems include making timely medical judgments and efficiently managing massive data utilising IoT-based resources. Furthermore, in many proposed solutions, the dispersed nature of data processing openly raises the risk of information leakage and compromises network integrity. Medical sensors are burdened by such solutions, which reduce the stability of real-time transmission systems. As a result, this study provides a machine-learning approach with SDN-enabled security to forecast network resource usage and enhance sensor data delivery. With a low administration cost, the software define network (SDN) design allows the network to resist dangers among installed sensors. It provides an unsupervised machine learning approach that reduces IoT network communication overheads and uses dynamic measurements to anticipate link status and refines its tactics utilising SDN architecture. Finally, the SDN controller employs a security mechanism to efficiently regulate the consumption of IoT nodes while also protecting them against unidentified events. When the number of nodes and data production rate varies, the suggested approach enhances network speed. As a result, to organise the nodes in a cluster, the suggested model uses an iterative centroid technique. By balancing network demand via durable connections, the multihop transmission technique for routing IoT data improves speed while simultaneously lowering the energy hole problem.

One of the important aspects of human-computer interaction is the detection of emotions using facial expressions. Emotion recognition has problems such as facial expressions, variations of posture, non-uniform illuminations, and so on. Deep learning techniques becomes important to solve these classification problems. In this chapter, VGG19, Inception V3, and Resnet50 pre-trained networks are used for the transfer learning approach to predict human emotions. Finally, the study achieved 98.32% of accuracy for emotion recognition and classification using the CK+ dataset.