A survey on various machine learning approaches for human electrocardiograms identification

Human identification is a critical function that can aid in data security protection. Developing deep learning models for human identification from electrocardiogram (ECG) data is one of the most promising strategies. It has a number of specific advantages, including the identification of liveness, insensitivity, ease of collecting, and greater security. On the other hand, present classifier-based methods can only identify closed sets, whilst existing matching-based methods are computationally intensive. Additionally, virtually all algorithms analyze only one-shot identification, which is subject to noise. In light of the fact that the electrocardiogram (ECG) is the most often used diagnostic tool for monitoring electrical activity in the heart, it is critical to use it to find early detection and diagnosis signals. The rapid growth and adoption of electronic health records, which include a systematized collection of various types of digitalized medical data, along with the development of new methods for quickly evaluating this massive amount of data, has resurrected interest in the fields of machine learning and deep learning in recent decades. The purpose of this article is to provide an overview of the EKG's significance in terms of learning approaches, as well as a comparison of the most well-known research and technical phrases relating to the electrocardiogram.