MRI机器学习在垂体神经内分泌肿瘤预后评估中的研究进展  

Research progress of MRI machine learning in predicting the prognosis of pituitary neuroendocrine tumors

作　　者：陈春晖 雒攀 董文洁 韩涛 孙嘉晨 周俊林[1,2,3,4] CHEN Chunhui;LUO Pan;DONG Wenjie;HAN Tao;SUN Jiachen;ZHOU Junlin(Department of Radiology,the Second Hospital of Lanzhou University,Lanzhou 730000,China;Second Clinical School of Lanzhou University,Lanzhou 730000,China;Key Laboratory of Medical Imaging of Gansu Province,Lanzhou 730000,China;Gansu International Scientific and Technological Cooperation Base of Medical Imaging Artificial Intelligence,Lanzhou 730000,China)

机构地区：[1]兰州大学第二医院放射科,兰州730000 [2]兰州大学第二临床医学院,兰州730000 [3]甘肃省医学影像重点实验室,兰州730000 [4]医学影像人工智能甘肃省国际科技合作基地,兰州730000

出　　处：《磁共振成像》2025年第2期154-158,171,共6页Chinese Journal of Magnetic Resonance Imaging

基　　金：国家自然科学基金项目(编号:82371914)。

摘　　要：垂体神经内分泌肿瘤(pituitary neuroendocrine tumor,PitNETs)虽然大多数表现为良性肿瘤,但垂体功能障碍、肿瘤侵袭性及不同并发症的出现可显著影响PitNETs患者的生存质量,因此对肿瘤预后进行无创评估在临床决策中具有重要意义。PitNETs的诊断与治疗高度依赖MRI,而机器学习作为人工智能的一个分支领域,近年来已成为医学领域中的热门议题。MRI机器学习在PitNETs的预后评估方面发挥了重要作用。本文就MRI机器学习在预测PitNETs化疗预后、术后复发/缓解、术后并发症及术后放疗预后方面的研究进展进行综述,以期为其个体化预后评估提供临床指导意义,为进一步研究指明方向。Pituitary neuroendocrine tumors(PitNETs)are mostly benign tumors,but pituitary dysfunction,tumor invasiveness,and the occurrence of various complications can significantly affect the quality of life of PitNETs patients,therefore,the non-invasive assessment of tumor prognosis is of great significance in clinical decision-making.MRI is the most commonly used examination method for PitNETs,and MRI machine learning have played an important role in the prognosis assessment of PitNETs.This review summarizes the research progress of MRI machine learning in predicting the chemotherapy prognosis,postoperative recurrence/remission,postoperative complications,and radiotherapy prognosis of PitNETs,with the aim of providing clinical guidance for individualized prognosis assessment and guiding future research.

关 键 词：垂体神经内分泌肿瘤 磁共振成像 机器学习 深度学习 影像组学 预后评估 

分 类 号：R445.2[医药卫生—影像医学与核医学] R739.41[医药卫生—诊断学]