The artificial intelligence revolution in medicine: challenging boundaries and transforming medical care.

Abstract

In May 2020, OpenAI launched its first artificial intelligence (AI) model called Chat GPT-3, which stands for Generative Pre-Trained Chat. On November 30, 2022 it was available for the public to interact with this tool, progressively plugins were added to it that allowed its access to the internet and integration with browsers, WhatsApp and other applications with the purpose of making it more accessible. From this moment on, the world's leading software companies rushed to include AI in their products, thus it is already integrated into Microsoft Office 365 products, Google's browser, Bing and others.

One of the greatest achievements of AI in medicine has been its ability to improve the diagnostic process. In a study published in the journal Nature Medicine by Li et al. it was shown that a deep learning model outperformed human pathologists in identifying lung cancers in CT images. The speed and accuracy with which AI analyzes large medical imaging datasets suggests a future where early and accurate diagnosis will be the norm, thus improving survival rates and reducing the emotional burden on patients.

At the time of publication of this article, there is no evidence of AI applications in the daily practice of CPB. And this fact is probably due to the complexity of decision making during perfusion conduction and even more importantly to the lack of data with which to train an AI. For an AI to learn to make coherent decisions in constantly changing environments, it needs to be trained with a massive amount of data.

One question that arises from this possibility: would it be possible in the future for people with less training than current perfusion professionals to safely conduct AI-assisted procedures? Will AI contribute to closing the demand gap for dedicated ECC professionals by allowing shorter training time and shorter academic requirements?