Ye Xu-Xing, Qu Hui-Heng, Yang Chao, Teng Wei-Jun, Chen Yan-Ping, Lin Jun-Mei, Wang Xiao-Bo
Department of Traditional Chinese Medicine, Jinhua Municipal Central Hospital, Jinhua 321000, Zhejiang Province, China.
Department of General Surgery, Wuxi No. 2 people's Hospital, Wuxi 214002, Jiangsu Province, China.
World J Gastrointest Oncol. 2025 Apr 15;17(4):102469. doi: 10.4251/wjgo.v17.i4.102469.
Rectal cancer is a major global health concern, and metachronous liver metastasis (MLM) significantly worsens patient prognosis. Advances in imaging and machine learning have led to the development of radiomics models, particularly those utilizing multiparametric magnetic resonance imaging, which are highly valuable in predicting MLM. These models analyze imaging features to provide insights that can aid clinical decision-making and potentially improve treatment outcomes and survival rates. However, realizing the full potential of radiomics models faces challenges in terms of accuracy, generalizability, and data dependency. This editorial comments on a study regarding radiomics prediction models for rectal cancer MLM published recently in the , discusses the progress, challenges, and strategies for diagnostic models of MLM in rectal cancer, and proposes directions for future research.
直肠癌是一个重大的全球健康问题，异时性肝转移（MLM）会显著恶化患者预后。成像技术和机器学习的进展促使了影像组学模型的发展，特别是那些利用多参数磁共振成像的模型，这些模型在预测MLM方面具有很高的价值。这些模型分析成像特征以提供有助于临床决策的见解，并有可能改善治疗效果和生存率。然而，要充分发挥影像组学模型的潜力，在准确性、可推广性和数据依赖性方面面临挑战。这篇社论对最近发表在《》上的一项关于直肠癌MLM影像组学预测模型的研究进行了评论，讨论了直肠癌MLM诊断模型的进展、挑战和策略，并提出了未来研究的方向。
