AI and Medicine
Today, AI is the subject of much research in anticipation of its wide range of applications in modern medicine. From diagnosis and treatment planning to drug discovery (Gupta, 2021) and precision medicine, which is targeted and personalized. It
( Figure 3 - The results per 100,000 citations for artificial intelligence or machine learning-Day, 2022 )
could be used to improve the patient experience by predicting patient outcomes, thus allowing for a more thorough and meaningful consultation and consent (Bhinder, 2021). Furthermore, it can vastly improve the efficiency within healthcare systems, which in the face of current global population concerns can only be of benefit. As a result of the huge potential of ML in medicine over the past decade, we have seen exponential growth in the amount of research done on AI, to the point where now >2% of medical research papers mention artificial intelligence or machine learning (Day, 2020). Further evidence of this is the fact that there are now numerous journals dedicated specifically to this field, which is graphically represented in Fig. 3. This shows the results per 100,000 for AI and machine learning in the past 20 years. As recently as November 2022 we have seen significant advancements being made in the field of foetal cardiology. The Prometheus Trial, a study in which foetal echocardiograms were undertaken with the assistance of AI, began at this time. This software was designed to identify different anatomical features of the foetal heart. This included taking measurements of the ventricles and atria as well as any septal defects that were identified. With the use of the Prometheus software, there was a significant reduction in the overall length of the scans by 34.7%. In addition to this, the scans produced clearer views of the relevant anatomy. In the field of paediatric and foetal sonography, it is important to note that there is a large geographical disparity in the availability of expertise. This suggests that the use of AI of this kind could indeed make healthcare systems more efficient as well as serving as a tool for reducing healthcare inequality. (Matthew, 2021) Within the current generation of medical AI, the predominant focus appears to be on imaging and diagnosis, using DL technology and computer vision to increase the speed and efficiency of diagnosis. There appears to be a great deal of scope for practice in this specific field. However, in fields such as pharmacology and drug manufacturing, there is also promise (Gupta 2021). Given the vast array of potential applications, it is important to establish criteria for what constitutes effective medical AI.
34
Made with FlippingBook - PDF hosting