Artificial Intelligence in Cardiology—A Narrative Review of Current Status

Abstract

Artificial intelligence (AI) is an integral part of clinical decision support systems (CDSS), offering methods to approximate human reasoning and computationally infer decisions. Such methods are generally based on medical knowledge, either directly encoded with rules or automatically extracted from medical data using machine learning (ML). ML techniques, such as Artificial Neural Networks (ANNs) and support vector machines (SVMs), are based on mathematical models with parameters that can be optimally tuned using appropriate algorithms. The ever-increasing computational capacity of today’s computer systems enables more complex ML systems with millions of parameters, bringing AI closer to human intelligence. With this objective, the term deep learning (DL) has been introduced to characterize ML based on deep ANN (DNN) architectures with multiple layers of artificial neurons. Despite all of these promises, the impact of AI in current clinical practice is still limited. However, this could change shortly, as the significantly increased papers in AI, machine learning and deep learning in cardiology show. We highlight the significant achievements of recent years in nearly all areas of cardiology and underscore the mounting evidence suggesting how AI will take a central stage in the field.

Construction of diagnostic and prognostic models based on gene signatures of nasopharyngeal carcinoma by machine learning methods

“…ANN differs from traditional regression analysis in that neural networks can analyze nonlinear data due to their data processing capabilities. With the selection of appropriate input and output layers, functional relationships with infinitely close correlations between the input and output layers can be discovered by learning and debugging large amounts of clinical data through network models ( 39 ). The network model is trained by providing the neural network with input and output layers and the connection weights can be adjusted during iterations to match the output with the actual output until the desired result is obtained ( 40 ).…”

Sixty years of the Bruce protocol: reappraising the contemporary role of exercise stress testing with electrocardiographic monitoring

“…23,110,111 Indeed, as digital tools become increasingly intertwined with several diagnostic modalities, so too may ExECG be affected by these applications. 11,[110][111][112] When considering ExECG in contemporary settings, an integrated assessment should be undertaken focusing not only on the ST segment but also on all the different and interlinked components that take part in the cardiovascular response to exercise and can be dynamically influenced by pathologic processes. 17,30,31,52 In this way, data derived from this complex examination can be streamlined as to improve its applicability and relevance while harnessing its full potential.…”

Early Diagnosis of Cardiovascular Diseases in the Era of Artificial Intelligence: An In-Depth Review

“…This area aims to solve human problems. A more cohesive, trustworthy, and efficient method of providing high-quality healthcare has been encouraged by the advent of artificial intelligence (AI), which provides methods for computers to mimic human cognitive functions such as learning and reasoning [9][10][11]. Research into the early detection and prevention of cardiovascular disorders is now underway, building on the well-established practice of using AI in cardiovascular sciences.…”