LAUSR

J Card Surg. 2022;37:3846–3847. 3846 | wileyonlinelibrary.com/journal/jocs Computers changed our lives for good and for bad quite a while ago. People in the kind of seniority side may agree that the Commodore 64, also known as the C64 or the CBM 64, an 8‐bit home computer introduced in January 1982 by Commodore International, was the home computer that allowed many of us to be introduced in regular daily computing. At some point, the Commodore 64 was the best‐ selling home computer. Commodore International, an American computer and electronics manufacturer, was instrumental in the development and global acceptance of the personal computer industry in the 1970s and 1980s. It was operational between 1958 and 1994, when it ceased operations after the company announced voluntary bankruptcy. Other computer platforms replaced it. Today we are almost at the mercy of computers as our daily life is almost controlled by them in different ways. Artificial intelligence (AI) is the human‐like behavior displayed by a machine or system. For the dummy, like those of us who signed this brief commentary, computers are aiming to mimic human behavior using as many data as possible from other examples that behave similarly. The goal is that machines should be able to work on the extreme side of efficiency by analyzing large, vast amounts of data in the shortest period of time. Historically, one may say that the seminal contribution of the late Alan Turing in 1936 paved the way for the development of what computing is today. This seems to be his most important contribution ever that was, of course, highlighted by many with the occasion of his centenary. With the understanding that history is useful for us to understand what happened in the past that we did not witness, AI is an old and very powerful tool in any field where large data are continuously generated. Medicine is a good example. Machine learning (ML), a form of AI, is rapidly making inroads and is going to be a part of the future in Medicine. It is currently used in different applications in healthcare. ML shows some potential ability to predict specific complications such as pancreatic fistula after pancreaticoduodenectomy. Deep learning software as a subset of ML seems to confirm around 80% sensitivity and specificity in detecting intracranial aneurysms on digital subtraction angiography as per a recent feasibility study. In the field of imaging in cardiology, studies suggest that ML may play a role in predicting follow‐up outcomes and, generally speaking improving patient care. Finally, AI seems to produce subjective image quality in the evaluation of pulmonary nodules with the proviso that there may be eventual false‐positive findings. Pooling all together, AI will likely and hopefully positively impact healthcare. Having said that, in this issue of the Journal, Penny‐Dimri et al. present our readership with an interesting systematic review and meta‐analysis on how ML may be a helpful tool for the prediction of outcomes in cardiac surgery. Cardiac surgery is a specialty that has produced substantial amount of data resulting in different tools to predict outcomes before an operation is performed. The European System for Cardiac Operative Risk Evaluation (EuroSCORE additive, logistic and II) and the Society of Thoracic Surgeons predicted risk of mortality (STS‐PROM) have been in use for a couple of decades