Department of Health Sciences (DISSAL), University of Genoa, Genoa, Italy.
Clinica Malattie Infettive, IRCCS Ospedale Policlinico San Martino, Italy.
Ann Med. 2023;55(2):2286336. doi: 10.1080/07853890.2023.2286336. Epub 2023 Nov 27.
Artificial intelligence (AI) and machine learning (ML) are revolutionizing human activities in various fields, with medicine and infectious diseases being not exempt from their rapid and exponential growth. Furthermore, the field of explainable AI and ML has gained particular relevance and is attracting increasing interest. Infectious diseases have already started to benefit from explainable AI/ML models. For example, they have been employed or proposed to better understand complex models aimed at improving the diagnosis and management of coronavirus disease 2019, in the field of antimicrobial resistance prediction and in quantum vaccine algorithms. Although some issues concerning the dichotomy between explainability and interpretability still require careful attention, an in-depth understanding of how complex AI/ML models arrive at their predictions or recommendations is becoming increasingly essential to properly face the growing challenges of infectious diseases in the present century.
人工智能(AI)和机器学习(ML)正在彻底改变各个领域的人类活动,医学和传染病也不例外,它们正在迅速而指数级地发展。此外,可解释 AI 和 ML 领域已经变得尤为重要,并吸引了越来越多的关注。传染病已经开始从可解释 AI/ML 模型中受益。例如,已经有人提议或使用它们来更好地理解旨在改善 2019 年冠状病毒病诊断和管理的复杂模型,以及在预测抗菌素耐药性和量子疫苗算法方面。尽管解释性和可解释性之间的二分法仍然存在一些问题需要谨慎关注,但深入了解复杂的 AI/ML 模型如何得出其预测或建议,对于正确应对本世纪传染病日益增长的挑战变得越来越重要。