The rapidly growing field of Artificial Intelligence (AI)—a field of computer science that seeks to perform and directly mimic tasks that generally require human intelligence—has developed a series of techniques that have been applied in cardiovascular medicine. Designed to enhance patient care, improve cost-effectiveness, and reduce readmission and mortality rates, these machine-learning techniques have been increasingly used for cardiovascular disease diagnosis and prediction.
Researchers believe that AI, in the near future, will ultimately result in ‘a paradigm shift toward precision cardiovascular medicine,’ as its applications in clinical care have tremendous potential in facilitating improvements in care: including machine learning, deep learning, and cognitive computing. The term ‘big data’ connotes extraordinarily large sets of data, which cannot be neither analyzed nor interpreted through traditional methods of data-processing. This type of data includes statistics from mobile phone apps, wearable devices, social media, and ‘omic’ data (i.e. genomics and proteomics), in addition to data from standardized electronic health records.
Clinical care in cardiovascular medicine currently faces several challenges, most of which relate to high costs in prevention and treatment, low cost-effectiveness and inadequacy in patient care, and overutilization. Moreover, because cardiovascular diseases are inherently complex—due to the multiple genetic, environmental, and behavioral factors that cause them—deep learning in AI, through the use of big data, can be used in pattern recognition to heterogeneous syndromes and image recognition in CV imaging; deep learning also uses multiple layers and transformations through several algorithms.
The concept of deep-learning, a new machine-learning technique that plays a critical role in areas including image recognition, has been applied through ideas and inventions including Facebook’s facial recognition system, speech-recognition, self-driving cars, mobile apps, machine vision camera software, IBM Watson, and robots. Because AI can effectively ‘classify new genotypes or phenotypes of heart failure with preserved ejection fraction,’ AI can greatly improve the accuracy of cardiac imaging methods.
Moreover, using big data can automatically generate new hypotheses, allowing physicians to make improved clinical decisions and diagnoses. The role of AI applications, such as machine learning, deep learning, and cognitive computing, can enable precision cardiovascular medicine, and move beyond traditional statistical tools—which will improve the estimated CVD risk scores to automate prediction.