Statistics from Altmetric.com
We live in an era with unprecedented availability of clinical and biological data that include electronic health records, wearable sensors, biomedical imaging and multiomics. The scale, complexity and rate at which such data are collected require innovative approaches to statistics and computer science that draw on the rapid advances in artificial intelligence (AI) for efficiently identifying actionable insights into disease processes. A basic understanding of AI’s strengths, applications and limitations is now essential for researchers and clinical cardiologists.
In this context, AI refers to a collection of computational concepts that can be summarised as a machine’s ability to generalise learning in order to efficiently achieve complex tasks autonomously. Machine learning (ML) achieves this by using algorithms to improve task performance without needing to be explicitly programmed and can be broadly divided into supervised and unsupervised approaches. In supervised learning, the mapping between paired input and output variables is iteratively optimised for use in regression and classification tasks. In unsupervised learning, only input data are available and algorithms are used to find inherent clusters or associations. In recent years, ML has become dominated by deep learning (DL), which is a methodology using multilayer neural networks to progressively obtain more abstract representations of complex data. Figure 1 provides a high-level schematic of the field of AI.
A DL algorithm consists of three types of layer: an input layer, hidden layers …
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.