Annese Vito, Annese Monica
Department Gastroenterology IRCCS San Donato Policlinic, Vita-Salute San Raffaele University, 20100 Milan, Italy.
Department Gastroenterology, IRCCS Hospital Casa Sollievo della Sofferenza, 71013 San Giovanni Rotondo, Italy.
Diagnostics (Basel). 2023 Aug 29;13(17):2797. doi: 10.3390/diagnostics13172797.
Ulcerative colitis and Crohn's disease are traditionally defined as the two main subtypes of inflammatory bowel disease. However, a more recent view considers IBD as a spectrum of heterogeneous phenotypes with consistent differences in clinical presentation and behaviors, likely explained by differences in underlying pathogenetic mechanisms. The etiology is still elusive, and the suggested pathogenesis is a complex interplay among genetic predisposition and abnormal immune response at the mucosal intestinal level, activated by only partially identified environmental triggers leading to altered intestinal permeability and impaired handling of gut microbiota. The undeniable continuous progress of medical therapy with more frequent shifts from traditional to more advanced modalities also underlines the actual unmet needs. We are using medications with completely different mechanisms of action, with a lack of predictive factors of outcomes and response and still an unsatisfactory rate of success. In addition, we are missing still valuable and accurate markers to predict disease progression and severity in order to avoid under- or over-treatment. In such a complex scenario, it is undoubtful that the application of artificial intelligence and machine learning algorithms may improve the management and pave the way for precision and eventually personalized medicine in these patients; however, there are still several challenges that will be the focus of this review.
溃疡性结肠炎和克罗恩病传统上被定义为炎症性肠病的两种主要亚型。然而，最近的一种观点认为炎症性肠病是一系列异质性表型，在临床表现和行为上存在持续差异，这可能是由潜在致病机制的差异所解释的。其病因仍然不明，推测的发病机制是遗传易感性与肠道黏膜水平异常免疫反应之间的复杂相互作用，这种相互作用由仅部分确定的环境触发因素激活，导致肠道通透性改变和肠道微生物群处理受损。随着医学治疗从传统模式向更先进模式的更频繁转变，其取得的不可否认的持续进展也凸显了实际未满足的需求。我们正在使用作用机制完全不同的药物，缺乏结果和反应的预测因素，成功率仍然不尽人意。此外，我们仍然缺少预测疾病进展和严重程度的有价值且准确的标志物，以避免治疗不足或过度治疗。在这样一个复杂的情况下，毫无疑问，人工智能和机器学习算法的应用可能会改善这些患者的管理，并为精准乃至个性化医疗铺平道路；然而，仍有几个挑战将成为本综述的重点。
