Artificial intelligence in medicine and research – the good, the bad, and the ugly

Abstract

Artificial intelligence (AI) broadly refers to machines that simulate intelligent human behavior, and research into this field is exponential and worldwide, with global players such as Microsoft battling with Google for supremacy and market share. This paper reviews the “good” aspects of AI in medicine for individuals who embrace the 4P model of medicine (Predictive, Preventive, Personalized, and Participatory) to medical assistants in diagnostics, surgery, and research. The “bad” aspects relate to the potential for errors, culpability, ethics, data loss and data breaches, and so on. The “ugly” aspects are deliberate personal malfeasances and outright scientific misconduct including the ease of plagiarism and fabrication, with particular reference to the novel ChatGPT as well as AI software that can also fabricate graphs and images. The issues pertaining to the potential dangers of creating rogue, super-intelligent AI systems that lead to a technological singularity and the ensuing perceived existential threat to mankind by leading AI researchers are also briefly discussed.

Artificial-Intelligence-Based Clinical Decision Support Systems in Primary Care: A Scoping Review of Current Clinical Implementations

“…By incorporating AI into CDSSs, they become more capable of clinical reasoning as they can handle more information and approach it more holistically. With ML, AI algorithms can identify patterns, trends, and correlations in EHRs that may not be apparent to physicians [ 15 , 16 , 19 , 40 ]. Likewise, they can learn from historical patient data to make predictions and recommendations for current patients [ 23 , 27 ].…”

“…With DL, AI arms CDSSs with the possibility of offering personalized treatment recommendations based on a patient’s unique medical history, genetics, and treatment responses [ 15 , 16 , 17 , 19 , 23 , 28 , 30 , 32 , 41 , 42 ]. Similarly, it can report abnormal tests or clinical results in real time and suggest alternative treatment options [ 23 , 29 , 31 , 32 ].…”

“…Although the concept of Artificial Intelligence (AI) was first introduced seven decades ago, its evolution began in 2010 with the enhancement of graphic processing units [ 12 , 13 , 14 , 15 ]. AI can mimic human reasoning and behavior [ 12 , 13 , 15 , 16 ] and handle the increasing volume of medical data within healthcare systems [ 17 ]. Machine learning (ML) is the most common AI technique used, and it can be categorized into three types: supervised, unsupervised, and reinforcement algorithms [ 12 , 18 ].…”

“…Machine learning (ML) is the most common AI technique used, and it can be categorized into three types: supervised, unsupervised, and reinforcement algorithms [ 12 , 18 ]. Massive training datasets are used as input to train ML algorithms to make accurate predictions, allowing computers to learn without explicit programming [ 12 , 15 , 16 , 19 ]. With deep learning (DL), a subset of ML, programs can learn and modify themselves by feedback from multiple layers, achieving the most stable prediction outcome [ 12 , 14 , 16 , 19 ].…”

How Artificial Intelligence Can Enhance the Diagnosis of Cardiac Amyloidosis: A Review of Recent Advances and Challenges

“…ML can be supervised, unsupervised, or reinforcement learning depending on the type and availability of data and feedback. Deep learning (DL) is a subset of machine learning that uses artificial neural networks to automatically identify and extract features from raw data such as images and text to make predictions or decisions [12,13]. Therefore, although the terms AI, ML, and DL may be used interchangeably, they are in fact hierarchical (Figure 2).…”