Chung Chen-Chih, Wu I-Chieh, Bamodu Oluwaseun Adebayo, Hong Chien-Tai, Chiu Hou-Chang
Department of Neurology, Taipei Medical University Shuang Ho Hospital, New Taipei City 235, Taiwan.
Department of Neurology, School of Medicine, College of Medicine, Taipei Medical University, Taipei 110, Taiwan.
Diagnostics (Basel). 2025 Aug 14;15(16):2044. doi: 10.3390/diagnostics15162044.
: Myasthenia gravis (MG), a chronic autoimmune disorder with variable disease trajectories, presents considerable challenges for clinical stratification and acute care management. This systematic review evaluated machine learning models developed for prognostic assessment in patients with MG. : Following PRISMA guidelines, we systematically searched PubMed, Embase, and Scopus for relevant articles published from January 2010 to May 2025. Studies using machine learning techniques to predict MG-related outcomes based on structured or semi-structured clinical variables were included. We extracted data on model targets, algorithmic strategies, input features, validation design, performance metrics, and interpretability methods. The risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool. : Eleven studies were included, targeting ICU admission ( = 2), myasthenic crisis ( = 1), treatment response ( = 2), prolonged mechanical ventilation ( = 1), hospitalization duration ( = 1), symptom subtype clustering ( = 1), and artificial intelligence (AI)-assisted examination scoring ( = 3). Commonly used algorithms included extreme gradient boosting, random forests, decision trees, multivariate adaptive regression splines, and logistic regression. Reported AUC values ranged from 0.765 to 0.944. Only two studies employed external validation using independent cohorts; others relied on internal cross-validation or repeated holdout. Of the seven prognostic modeling studies, four were rated as having high risk of bias, primarily due to participant selection, predictor handling, and analytical design issues. The remaining four studies focused on unsupervised symptom clustering or AI-assisted examination scoring without predictive modeling components. : Despite promising performance metrics, constraints in generalizability, validation rigor, and measurement consistency limited their clinical application. Future research should prioritize prospective multicenter studies, dynamic data sharing strategies, standardized outcome definitions, and real-time clinical workflow integration to advance machine learning-based prognostic tools for MG and support improved patient care in acute settings.
重症肌无力（MG）是一种具有可变疾病轨迹的慢性自身免疫性疾病，给临床分层和急性护理管理带来了巨大挑战。本系统评价评估了为MG患者预后评估开发的机器学习模型。：遵循PRISMA指南，我们系统检索了PubMed、Embase和Scopus中2010年1月至2025年5月发表的相关文章。纳入了使用机器学习技术基于结构化或半结构化临床变量预测MG相关结局的研究。我们提取了关于模型目标、算法策略、输入特征、验证设计、性能指标和可解释性方法的数据。使用预测模型偏倚风险评估工具评估偏倚风险。：纳入了11项研究，目标包括重症监护病房（ICU）入院（n = 2）、肌无力危象（n = 1）、治疗反应（n = 2）、机械通气延长（n = 1）、住院时间（n = 1）、症状亚型聚类（n = 1）以及人工智能（AI）辅助检查评分（n = 3）。常用算法包括极端梯度提升、随机森林、决策树、多元自适应回归样条和逻辑回归。报告的AUC值范围为0.765至0.944。只有两项研究使用独立队列进行外部验证；其他研究依赖内部交叉验证或重复留出法。在七项预后建模研究中，四项被评为具有高偏倚风险，主要是由于参与者选择、预测变量处理和分析设计问题。其余四项研究专注于无监督症状聚类或AI辅助检查评分，没有预测建模成分。：尽管性能指标很有前景，但在可推广性、验证严谨性和测量一致性方面的限制限制了它们的临床应用。未来的研究应优先进行前瞻性多中心研究、动态数据共享策略、标准化结局定义以及实时临床工作流程整合，以推进基于机器学习的MG预后工具，并支持改善急性情况下的患者护理。
