应对慢性粒细胞白血病 (CML) 诊断和管理的复杂性提出了重大挑战，包括需要准确预测疾病进展和治疗反应。人工智能 (AI) 提供了一种变革性方法，可以开发复杂的预测模型和个性化治疗策略，从而增强早期检测并改善治疗干预，从而获得更好的患者结果。我们进行了广泛的搜索，从 PubMed、Scopus 和 Web 中检索相关文章截至 2023 年 4 月 24 日的科学数据库。使用标准化提取表格收集数据，结果以表格和图表形式呈现，显示频率和百分比。作者遵守 PRISMA-ScR 检查表，以确保研究报告透明。在最初确定的 176 篇文章中，在删除重复项并应用纳入和排除标准后，选择了 12 篇进行我们的研究。 AI在管理CML方面的主要应用包括肿瘤诊断/分类（n = 9，75%）、预测/预后（n = 2，17%）和治疗（n = 1，8%）。对于肿瘤诊断，人工智能分为基于血涂片图像的方法（n = 5）、基于临床参数的方法（n = 2）和基于基因分析的方法（n = 2）。最常用的人工智能模型包括支持向量机（SVM）（n = 5）、极限梯度提升（XGBoost）（n = 4）以及各种神经网络方法，例如人工神经网络（ANN）（n = 3） 。此外，具有交互式自学学校的混合卷积神经网络 (HCNN-IAS) 在组织白血病数据类型方面实现了 100% 的准确度和灵敏度，而 MayGAN 在从血涂片图像诊断 CML 方面实现了 99.8% 的准确度和高性能。人工智能提供了突破性的见解和工具用于增强慢性粒细胞白血病的预测、预后和个性化治疗。集成的人工智能系统使医疗保健从业者能够进行高级分析，优化患者护理并改善 CML 管理的临床结果。© 2024。作者。

Navigating the complexity of chronic myeloid leukemia (CML) diagnosis and management poses significant challenges, including the need for accurate prediction of disease progression and response to treatment. Artificial intelligence (AI) presents a transformative approach that enables the development of sophisticated predictive models and personalized treatment strategies that enhance early detection and improve therapeutic interventions for better patient outcomes.An extensive search was conducted to retrieve relevant articles from PubMed, Scopus, and Web of Science databases up to April 24, 2023. Data were collected using a standardized extraction form, and the results are presented in tables and graphs, showing frequencies and percentages. The authors adhered to the PRISMA-ScR checklist to ensure transparent reporting of the study.Of the 176 articles initially identified, 12 were selected for our study after removing duplicates and applying the inclusion and exclusion criteria. AI's primary applications of AI in managing CML included tumor diagnosis/classification (n = 9, 75%), prediction/prognosis (n = 2, 17%), and treatment (n = 1, 8%). For tumor diagnosis, AI is categorized into blood smear image-based (n = 5), clinical parameter-based (n = 2), and gene profiling-based (n = 2) approaches. The most commonly employed AI models include Support Vector Machine (SVM) (n = 5), eXtreme Gradient Boosting (XGBoost) (n = 4), and various neural network methods, such as Artificial Neural Network (ANN) (n = 3). Furthermore, Hybrid Convolutional Neural Network with Interactive Autodidactic School (HCNN-IAS) achieved 100% accuracy and sensitivity in organizing leukemia data types, whereas MayGAN attained 99.8% accuracy and high performance in diagnosing CML from blood smear images.AI offers groundbreaking insights and tools for enhancing prediction, prognosis, and personalized treatment in chronic myeloid leukemia. Integrated AI systems empower healthcare practitioners with advanced analytics, optimizing patient care and improving clinical outcomes in CML management.© 2024. The Author(s).