Department of Transplantation Surgery, Hippokration General Hospital, School of Medicine, Aristotle University of Thessaloniki, Thessaloniki, Greece.
School of Medicine, Aristotle University of Thessaloniki, Thessaloniki, Greece.
Microrna. 2022;11(3):175-184. doi: 10.2174/2211536611666220818145553.
MicroRNAs constitute small non-coding RNAs that play a pivotal role in regulating the translation and degradation of mRNA and have been associated with many diseases. Artificial Intelligence (AI) is an evolving cluster of interrelated fields, with machine learning (ML) standing out as one of the most prominent AI fields, with a plethora of applications in almost every aspect of human life. ML could be defined as computer algorithms that learn from past data to predict future data. This review comprehensively reviews the current applications of microRNA-based ML models in healthcare. The majority of the identified studies investigated the role of microRNA-based ML models in the management of cancer and specifically gastric cancer (maximum diagnostic accuracy (Accmax): 94%), pancreatic cancer (Accmax: 93%), colorectal cancer (Accmax: 100%), breast cancer (Accmax: 97%), ovarian cancer, neck squamous cell carcinoma, liver cancer, lung cancer (Accmax: 100%), and melanoma. Except for cancer, microRNA-based ML models have been applied for a plethora of other diseases, including ulcerative colitis (Accmax: 92.8%), endometriosis, gestational diabetes mellitus (Accmax: 86%), hearing loss, ischemic stroke, coronary heart disease (Accmax: 96%), tuberculosis, pulmonary arterial hypertension (Accmax: 83%), dementia (Accmax: 82.9%), major cardiovascular events in end-stage renal disease patients, and alcohol dependence (Accmax: 79.1%). Our findings suggest that the development of microRNA-based ML models could be used to enhance the diagnostic accuracy of a plethora of diseases while at the same time substituting or minimizing the use of more invasive diagnostic means (such as endoscopy). Even not as fast as anticipated, AI will eventually infiltrate the entire healthcare industry. AI is the key to a clinical practice where medicine's inherent complexity is embraced. Therefore, AI will become a reality that physicians should conform with to avoid becoming obsolete.
微小 RNA 构成了小的非编码 RNA，在调节 mRNA 的翻译和降解方面发挥着关键作用，并且与许多疾病有关。人工智能 (AI) 是一个不断发展的相互关联的领域，机器学习 (ML) 是其中最突出的领域之一，在人类生活的几乎各个方面都有大量的应用。ML 可以定义为从过去的数据中学习以预测未来数据的计算机算法。本综述全面回顾了基于 microRNA 的 ML 模型在医疗保健中的当前应用。大多数已确定的研究都探讨了基于 microRNA 的 ML 模型在癌症管理中的作用，特别是胃癌（最大诊断准确性 (Accmax)：94%）、胰腺癌（Accmax：93%）、结直肠癌（Accmax：100%）、乳腺癌（Accmax：97%）、卵巢癌、颈部鳞状细胞癌、肝癌、肺癌（Accmax：100%）和黑色素瘤。除了癌症，基于 microRNA 的 ML 模型还被应用于许多其他疾病，包括溃疡性结肠炎（Accmax：92.8%）、子宫内膜异位症、妊娠糖尿病（Accmax：86%）、听力损失、缺血性中风、冠心病（Accmax：96%）、结核病、肺动脉高压（Accmax：83%）、痴呆症（Accmax：82.9%）、终末期肾病患者的主要心血管事件和酒精依赖（Accmax：79.1%）。我们的研究结果表明，基于 microRNA 的 ML 模型的开发可以用于提高许多疾病的诊断准确性，同时替代或最小化使用更具侵入性的诊断手段（如内窥镜检查）。即使没有像预期的那样快，人工智能最终也将渗透到整个医疗保健行业。AI 是接受医学固有复杂性的临床实践的关键。因此，AI 将成为医生必须顺应的现实，以避免过时。
