Enhancing Cardiac Arrest Education: exploring the potential use of MidJourney.

Artificial Intelligence (AI) is rapidly reshaping the landscape of healthcare by transforming various aspects of patient care, medical research, and operational efficiency. In healthcare settings, AI systems are enabling precise diagnosis, accurate prediction of disease progression and outcomes, personalizing patient treatments, automating routine tasks, managing vast volumes of patient data, and facilitating remote patient monitoring, thereby propelling healthcare towards an era of unprecedented accuracy and efficiency.1Yoon J.H. Pinsky M.R. Clermont G. Artificial intelligence in critical care medicine.Crit Care. 2022; 26https://doi.org/10.1186/s13054-022-03915-3Crossref PubMed Scopus (16) Google Scholar In the light of the expanding influence of AI, we embarked on an exploration of the potential of MidJourney (MJ), a San Francisco-based generative AI program developed by an independent research lab. MJ, similar to AI models like OpenAI's DALL-E and Stable Diffusion, is capable of generating images from textual prompts, known as natural language descriptions. This unique capability makes MJ an invaluable tool in medical education. The use of high-quality, custom-designed images can enhance teaching materials and enrich the overall learning experience. The efficacy of visual learning and visual media in medical education is well-documented.2Ferrara V. De S.S. Manicone F. Martinino A. Consorti F. The visual art as a learning tool in medical education.Senses Sci. 2020; 7: 1028-1040https://doi.org/10.14616/sands-Crossref Google Scholar Leveraging MJ's capabilities, educators can craft interactive, real-life simulations that foster an engaging learning environment for medical trainees, particularly in areas as critical as cardiac arrest (CA) education (Fig. 1). Moreover, MJ's capacity to construct interactive scenarios extends its applicability to layperson training. It enables the creation of simplified scenarios that can be integrated with virtual reality tools for educating children about health issues. This potential makes MJ a powerful tool for crafting engaging dissemination materials to heighten public awareness about pressing health matters like CA. However, the application of MJ in the domain of medical art is not without limitations. MJ employs a list of banned terms, which includes certain medical terminologies, to prevent the generation of violent or graphic content. In addition, the platform's algorithm may generate varied results due to the limited pool of quality medical images, algorithmic biases, and inherent challenges associated with interpreting specialized medical terminology3Cammer M. Too bad to be fraud, Midjourney has yet to embark in science.bioRxiv. 2023; https://doi.org/10.1101/2023.01.28.526052Crossref PubMed Scopus (0) Google Scholar (Supplemental Materials). To address these limitations, future enhancements could involve creating certified accounts for recognized public institutions or scientific societies.4Kuslich S. Is Midjourney AI Smart enough to make Medical Art? https://www.ghostproductions.com/news/is-midjourney-ai-smart-enough-to-make-medical-art. Accessed June 13, 2023.Google Scholar These certified users could work without certain terminology restrictions, taking on the responsibility for content creation. Additionally, the establishment of a closed database featuring accurate scientific illustrations could help train MJ's system further. Strictly limiting the use of such images to scientific applications would ensure compliance with copyright and privacy regulations. In conclusion, AI-generated art platforms like MJ hold immense potential in the field of medical education, particularly in areas like CA training.5Meskó B. The real era of the art of medicine begins with artificial intelligence.J Med Internet Res. 2019; 21e16295https://doi.org/10.2196/16295Crossref Scopus (16) Google Scholar These platforms can significantly contribute to the development of educational materials for professionals and community education in cardiopulmonary resuscitation. However, the current limitations in the production of medical art using AI necessitate continuous improvements. These include involving professionals in the training process, expanding the image resources pool, and developing more robust algorithms capable of effectively handling complex medical terminology. By addressing these challenges, we can enhance the accuracy and reliability of AI platforms like MJ, bringing them more in line with user expectations, and unlocking their full potential in healthcare education. Carlo Alberto Mazzoli: Conceptualization, Data curation, Methodology, Supervision, Writing - original draft, Writing - review & editing. Federico Semeraro: Conceptualization, Data curation, Methodology, Supervision, Writing – original draft, Writing – review & editing. Lorenzo Gamberini: Conceptualization, Data curation, Methodology, Supervision, Writing – original draft, Writing – review & editing. The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: No relationship exists between any of the authors and any commercial entity or product mentioned in this manuscript that might represent a conflict of interest. No inducements have been made by any commercial entity to submit the manuscript for publication. All within 3 years of beginning the work submitted. FS is the Chair-Elect of the European Resuscitation Council, Chair of the ILCOR Social Media Working Group and ILCOR BLS Working Group members. LG are Scientific Committee members of the Italian Resuscitation Council. CAM has no conflicts of interest. The following are the Supplementary data to this article: Download .pdf (.54 MB) Help with pdf files Supplementary Data 1