en
×

分享给微信好友或者朋友圈

使用微信“扫一扫”功能。
通讯作者:

洪汛宁,E-mail: hongxunning@sina.com

中图分类号:R445.2

文献标识码:A

文章编号:1007-4368(2024)12-1729-06

DOI:10.7655/NYDXBNSN240381

参考文献 1
LOPES M B S.The 2017 World Health Organization clas-sification of tumors of the pituitary gland:a summary[J].Acta Neuropathol,2017,134(4):521-535
参考文献 2
MOLITCH M E.Diagnosis and treatment of pituitary ade-nomas:a review[J].JAMA,2017,317(5):516-524
参考文献 3
ČERNÝ M,SEDLÁK V,LESÁKOVÁ V,et al.Methods of preoperative prediction of pituitary adenoma consistency:a systematic review[J].Neurosurg Rev,2022,46(1):11
参考文献 4
CHEN X Y,DING C Y,YOU H H,et al.Relationship be-tween pituitary adenoma consistency and extent of resec-tion based on tumor/cerebellar peduncle T2-weighted im-aging intensity(TCTI)ratio of the point on preoperative magnetic resonance imaging(MRI)corresponding to the residual point on postoperative MRI[J].Med Sci Monit,2020,26:e919565
参考文献 5
SU C Q,ZHANG X,PAN T,et al.Texture analysis of high b-value diffusion-weighted imaging for evaluating consis-tency of pituitary macroadenomas[J].J Magn Reson Im-aging,2020,51(5):1507-1513
参考文献 6
DING W,HUANG Z,ZHOU G,et al.Diffusion-weighted imaging for predicting tumor consistency and extent of re-section in patients with pituitary adenoma[J].Neurosurg Rev,2021,44(5):2933-2941
参考文献 7
PIERALLINI A,CARAMIA F,FALCONE C,et al.Pitu-itary macroadenomas:preoperative evaluation of consis-tency with diffusion-weighted MR imaging--initial experi-ence[J].Radiology,2006,239(1):223-231
参考文献 8
MOHAMED F F,ABOUHASHEM S.Diagnostic value of apparent diffusion coefficient(ADC)in assessment of pi-tuitary macroadenoma consistency[J].Egypt J Radiol Nu-cl Med,2013,44(3):617-624
参考文献 9
MASTORAKOS P,MEHTA G U,CHATRATH A,et al.Tumor to cerebellar peduncle T2-weighted imaging inten-sity ratio fails to predict pituitary adenoma consistency[J].J Neurol Surg B Skull Base,2019,80(3):252-257
参考文献 10
BARBOSA M A,PEREIRA E G R,DA MATA PEREIRA P J,et al.Diffusion-weighted imaging does not seem to be a predictor of consistency in pituitary adenomas[J].Pitu-itary,2024,27:187-196
参考文献 11
LIN D,LIU J,KE C,et al.Radiomics analysis of quantita-tive maps from synthetic MRI for predicting grades and molecular subtypes of diffuse gliomas[J].Clin Neuroradi-ol,2024,DOI:10.1007/s00062-024-01421-3
参考文献 12
黄晓星,汪泽燕,肖学红,等.术前MRI强化特征预测胶质母细胞瘤患者术后复发风险的价值[J].临床放射学杂志,2022,41(2):217-223
参考文献 13
RUTKOWSKI M J,CHANG K E,CARDINAL T,et al.De-velopment and clinical validation of a grading system for pituitary adenoma consistency[J].J Neurosurg,2020,134(6):1800-1807
参考文献 14
SONG Y,ZHANG J,ZHANG Y D,et al.FeAture Explorer(FAE):a tool for developing and comparing radiomics models[J].PLoS One,2020,15(8):e0237587
参考文献 15
LI P,ZHANG D,MA S,et al.Consistency of pituitary ade-nomas:amounts of collagen types Ⅰ and Ⅲ and the pre-dictive value of T2WI MRI[J].Exp Ther Med,2021,22(5):1255
参考文献 16
SU C Q,WANG B B,TANG W T,et al.Diffusion-relax-ation correlation spectrum imaging for predicting tumor consistency and gross total resection in patients with pitu-itary adenomas:a preliminary study[J].Eur Radiol,2023,33(10):6993-7002
参考文献 17
COHEN-COHEN S,HELAL A,YIN Z,et al.Predicting pituitary adenoma consistency with preoperative magnetic resonance elastography[J].J Neurosurg.2021,136(5):1356-1363
参考文献 18
林洁,苏春秋,唐文天,等.基于T1WI增强影像组学在鉴别高级别胶质瘤复发与假性进展中的应用价值[J].南京医科大学学报(自然科学版),2023,43(9):1279-1284
参考文献 19
ZEYNALOVA A,KOCAK B,DURMAZ E S,et al.Preop-erative evaluation of tumour consistency in pituitary mac-roadenomas:a machine learning-based histogram analysis on conventional T2-weighted MRI[J].Neuroradiology,2019,61(7):767-774
参考文献 20
CUOCOLO R,UGGA L,SOLARI D,et al.Prediction of pituitary adenoma surgical consistency:radiomic data mining and machine learning on T2-weighted MRI[J].Neuroradiology,2020,62(12):1649-1656
目录contents

    摘要

    目的:探讨基于T2WI和DWI影像组学在术前无创预测垂体腺瘤质地中的应用价值。方法:回顾性分析经病理证实为垂体腺瘤的108例患者临床及术前MRI资料,术中2名神经外科医生评估肿瘤质地,将其分为质软组和质硬组。按7∶3随机分为训练组和验证组,在T2WI和DWI 图像上手动勾画肿瘤实质区的体积作为感兴趣区容积(volume of interest,VOI),用 FeAture Explorer软件提取特征,采用无监督特征选择(unsupervised feature selection,UFS)进行特征筛选,采用支持向量机(sup- port vector machine,SVM)构建影像组学模型。通过受试者工作特征曲线下面积(area under curve,AUC)及校准曲线评估模型的效能。结果:在联合T2WI和DWI影像组学模型中,训练组预测垂体腺瘤质地的AUC为0.89,验证组的AUC为0.80。校准曲线显示模型预测值与实际值一致性较好。结论:联合T2WI和DWI影像组学模型具有较好的诊断效能,有助于术前预测垂体腺瘤的质地。

    Abstract

    Objective:To explore the application value of radiomics based on T2-weighted imaging(T2WI)and diffusion-weighted imaging(DWI)in non-invasive preoperative prediction of pituitary adenoma consistency. Methods:The clinical and preoperative MRI data of 108 patients with pathologically confirmed pituitary adenoma were retrospectively analyzed. Two neurosurgeons evaluated tumor consistency intraoperatively and categorized them into soft and hard groups. Patients were randomly divided into a training set and a validation set in a 7∶3 ratio. Volume of interest(VOI)representing the tumor solid component were manually delineated on T2WI and DWI images. Radiomics features were extracted by FeAture Explorer software. Unsupervised feature selection(UFS)was applied for feature selection. Support vector machine(SVM)was used to conduct the radiomics models. Area under curve(AUC)and calibration curve were used to assess the performance of the models. Results:In the combined T2WI and DWI radiomics model,the AUC for predicting the consistency of pituitary adenoma was 0.89 in the training set and 0.80 in the validation set. The calibration curve showed a good consistency between predicted and actual values. Conclusion:The combined T2WI and DWI radiomics model demonstrates good diagnostic performance and aids in preoperative prediction of the consistency of pituitary adenoma.

  • 垂体腺瘤是鞍区最常见的病变,在脑肿瘤中排名第3 [1]。内镜下经鼻蝶窦手术已被公认为首选的手术方法[2]。绝大多数的腺瘤质地柔软,易于完整切除,但有10%~15%的腺瘤质地坚硬,常规经蝶入路术式切除困难,可能需要二次或经颅手术。因此,术前无创性评估垂体腺瘤的质地对于制定手术方案及患者早期危险分层具有重要的临床意义。

  • 既往研究基于常规 T2 加权成像(T2-weighted imaging,T2WI)和扩散加权成像(diffusion-weighted imaging,DWI)对垂体腺瘤的质地进行研究,但结果存在争议[3]。T2WI信号强度较低和/或表观扩散系数(apparent diffusion coefficient,ADC)较低在一些研究中被认为是硬腺瘤的指标[4-6],而在另一些研究中提出了相反的研究结果[7-8],甚至有研究提出两者差异无统计学意义[9-10]。但垂体腺瘤病理成分复杂,其内部的出血、钙化或囊变组织内蛋白含量高的液体均可引起T2WI及DWI信号改变,通过T2WI信号比值及单纯的细胞密度角度,无法反映整个肿瘤的异质性,这可能是既往研究存在局限性的原因。

  • 影像组学将图像转化为可定量的影像特征数据,利用机器学习算法设计并建立相关的预测模型,为临床决策提供支持。研究表明,影像组学模型在预测胶质瘤分级、分子分型及预后有较大的临床价值[11-12],提高了患者诊断准确性及预后评估。然而,影像组学分析在垂体腺瘤质地预测中的相关研究较少。因此,本研究的目的在于评估基于T2WI和DWI 影像组学在术前预测垂体腺瘤质地中的应用价值。

  • 1 对象和方法

  • 1.1 对象

  • 回顾性分析了2016年11月—2023年4月在南京医科大学第一附属医院神经外科经手术切除的垂体腺瘤患者共115例。纳入标准如下:①所有患者临床资料完整,均经手术病理证实为垂体腺瘤;②所有患者均为初发首治的患者,均未经过药物及其他治疗; ③所有患者术前均在南京医科大学第一附属医院进行垂体MRI检查,包括 T1WI、T2WI、DWI及T1WI增强序列扫描等。排除标准如下:①MRI运动伪影大,图像质量差;②垂体腺瘤病变直径<10 mm;③明显出血影响评估。最终入组108例,其中,男61例,女 47例,年龄13~81岁,平均年龄(51.49±14.26)岁。

  • 1.2 方法

  • 1.2.1 质地评估标准

  • 记录人口统计学和临床特征,包括年龄、性别、内分泌功能、Knosp 分级、肿瘤最大直径和肿瘤体积。所有患者均由2名神经外科医生行经蝶窦手术切除。由2位神经外科医生(分别具有10年及18年神经外科手术经验)术中根据Rutkowski等[13] 报道的 5分分级量表共同评估肿瘤质地。评估标准如下: 1 级和2级的垂体腺瘤可以自由抽吸;3级肿瘤可以通过抽吸和刮除相结合的方法切除;4级肿瘤质地坚硬,无法抽吸,需要刮除或机械减瘤;5级肿瘤质地非常坚硬,无法刮除,需要广泛的利器、机械或整体切除。本研究将每个肿瘤的质地分为质软(1~3级)和质硬(4~5 级)。根据以上标准,本研究中 108 例患者,分为质软组78例,质硬组30例。

  • 1.2.2 MRI检查

  • 采用3.0T磁共振成像扫描仪,使用头颅24通道正交线圈,常规 MRI 方案包括以下序列:冠状面 T1WI,TR 500 ms,TE 9 ms,层厚 2 mm,视野(FOV) 200 mm×200 mm,矩阵 320×240;矢状面 T1WI,TR 200 ms,TE 2.59 ms,层厚2 mm,FOV 200 mm×200 mm,矩阵320×240;冠状面T2WI,TR 3 500 ms,TE 89 ms,层厚 2 mm,FOV 200 mm×200 mm,矩阵 384×268。此外,注射 0.1 mmol/kg 造影剂钆喷酸葡胺(Gd-DTPA)后,获得相同方向的增强T1WI序列;获得冠状面和矢状面T1WI。

  • 注射造影剂前冠状面DWI图像在标准b值(b = 1 000 s/mm2)下获得,使用以下参数:TR 3 000 ms, TE 75 ms,层厚 3 mm,相交间隙 0.9 mm,矩阵 200× 200,体素 1.5 mm×1.5 mm×3.0 mm,翻转角度 180°。 ADC1000图由 b=0 和 1 000 s/mm2 的 DWI 图逐体素计算得到。

  • 1.2.3 影像组学分析

  • 1.2.3.1 图像分割与特征提取

  • 使用 ITK-SNAP 软件在 T2WI 和 DWI 序列图像上逐层手工勾画肿瘤体积作为最终分析的三维感兴趣区容积(volume of interest,VOI),避开囊变坏死及出血区域。随机选取20例患者,由2名神经影像诊断医生(分别具有2年及8年神经影像诊断经验) 分别对 T2WI 和 DWI 图像进行勾画,用以计算组内相关系数(intraclass correlation coefficient,ICC), ICC >0.8 为一致性好。其余病例由医师 1 独立勾画。所有图像由 1 名具有 20 年神经影像诊断经验的神经放射学专家评估和验证。

  • 采用 FeAture Explorer 软件[14] (https://github. com/salan668/FAE)对T2WI和DWI图像的VOI分别提取 1 316 个特征,包括:①形状特征;②一阶直方图特征;③二阶纹理特征:包括灰度共生矩阵、灰度游程矩阵、灰度大小区域矩阵、邻域灰度差矩阵及灰度依赖矩阵;④高阶特征:包括对以上特征进行小波变换等高阶变换。将所有病例以7∶3的比例随机分为训练组(75例)和验证组(33例)。影像组学流程图如图1所示。

  • 1.2.3.2 特征处理

  • 首先使用 Z-score 法将影像组学特征标准化到 0~1范围内,剔除ICC<0.8的特征。由于质地为硬的垂体腺瘤相对较少,考虑到两组间的样本量不平衡会影响分类器的效能,因此采用合成少数过采样技术(synthetic minority over-sampling technique, SMOTE)来避免数据偏倚。使用无监督特征选择 (unsupervised feature selection,UFS)进行特征筛选。

  • 1.2.3.3 模型建立与效能评价

  • 采用支持向量机(support vector machine,SVM) 作为分类器建立组学模型,分别构建临床模型、基于 T2WI 影像组学模型、基于 DWI 影像组学模型及联合 T2WI 和 DWI 影像组学模型。在建模过程使用网格搜索和 5 倍交叉验证法以提高预测模型的精度和泛化性。并绘制模型训练组和验证组的受试者工作特征(receiver operating characteristic,ROC) 曲线,计算曲线下面积(area under curve,AUC)、准确率、灵敏度、特异度。绘制校准曲线验证模型效能。

  • 图1 影像组学流程图

  • Figure1 The flow chart of radiomics

  • 1.3 统计学方法

  • 使用SPSS23.0软件进行统计分析。首先对计量资料包括年龄、肿瘤最大直径、肿瘤体积进行 Kol-mogorov-Smirnov 正态性检验和 Levene 方差齐性检验,符合正态分布的计量资料用均数±标准差(x-±s) 表示,采用独立样本t检验;不符合正态分布者以中位数(四分位数)[MP25P75)]表示,采用非参数Mann-Whitney U检验进行比较。分类资料包括性别、Knosp 分级、内分泌功能用频数(百分比)表示,采用卡方检验进行比较。P <0.05为差异有统计学意义。

  • 2 结果

  • 2.1 临床影像资料

  • 患者临床影像资料见表1,质软组与质硬组间的性别差异有统计学意义(P <0.05)。年龄、Knosp 分级、肿瘤体积、肿瘤最大直径及内分泌功能在质软组和质硬组间差异无统计学意义(P均>0.05)。

  • 2.2 影像组学模型及效能评价

  • 影像组学模型最终均纳入9个影像组学特征,利用ROC曲线来评估组学模型的诊断效能(图2)。校准曲线分析模型校准效能(图3)。在临床模型中,训练组预测垂体腺瘤质地的 AUC 为 0.82,验证组的AUC为0.63;在基于T2WI影像组学模型中,训练组预测垂体腺瘤质地的 AUC 为 0.85,验证组的 AUC 为 0.52;在基于 DWI 影像组学模型中,训练组预测垂体腺瘤质地的AUC为0.85,验证组的AUC为 0.75;在联合T2WI和DWI影像组学模型中模型效能最优,训练组预测垂体腺瘤质地的 AUC 为 0.89,验证组的 AUC 为 0.80。各模型详细预测效能见表2。校准曲线结果显示联合T2WI和DWI影像组学模型一致性较好。

  • 表1 患者临床影像信息

  • Table1 Clinical imaging information of patients

  • 图2 各种模型预测效能的ROC曲线

  • Figure2 ROC curves of the predictive efficacy of various models

  • 3 讨论

  • 术前评估垂体腺瘤的质地对神经外科医生制定适当的手术策略至关重要。本研究表明,与临床模型和基于单序列影像组学模型相比,联合 T2WI 和DWI影像组学模型具有更好的诊断效能。

  • 研究表明垂体腺瘤质地与病理中胶原纤维含量密切相关,胶原成分越高,其质地越坚硬[15]。肿瘤内纤维含量会缩短弛豫时间,因此有研究表明 T2WI 信号越低,肿瘤质地越硬[4],此外既往基于 DWI 通过评估肿瘤的细胞密度角度来预测垂体腺瘤质地的研究结果也存在争议[5-68-10]。垂体腺瘤病理成分复杂,其内部的出血、钙化或囊变组织内蛋白含量高的液体均可引起 T2WI 及 DWI 信号改变。而且既往多采用热点法勾画感兴趣区,通过 T2WI 信号比值及单纯的细胞密度角度,无法反映整个肿瘤的异质性,这可能是既往研究存在局限性的原因。现有一些研究通过功能 MRI 来预测垂体腺瘤的质地[16-17],虽然这些研究取得了良好的结果,但功能MRI未包括在常规扫描方案中,且扫描失败率及诊断不确定性依然存在。

  • 图3 联合T2WI和DWI影像组学模型校准曲线

  • Figure3 Calibration curves of the combined T2WI and DWI radiomics model

  • 表2 临床及影像组学模型在训练组和验证组的预测效能

  • Table2 Predictive performance of clinical and radiomics models in the training and validation sets

  • 本研究基于影像组学方法,高通量挖掘图像的定量特征,使许多肉眼无法识别的信息被充分挖掘,从不同层面揭示了肿瘤的异质性。本研究提取了丰富的影像组学特征,并利用SVM及网格搜索进行模型建立。在机器学习中,SVM 是用来解决二分类问题的有效监督学习算法。其从本质上避开了从归纳到演绎的传统过程,更为高效地完成从训练样本到验证样本的“工作流程”。网格搜索又称超参数搜索,其核心是遍历这个参数网格中的所有可能的参数组合。对于每种组合,模型都会被训练并评估其性能,最终找到在验证集上精度最高的参数,显著提高了模型的准确性。SVM和网格搜索对小样本数据分类有很好的预测效果,目前广泛用于其他领域的影像组学研究[18]。本研究的联合模型预测效能优于Zeynalova等[19] 基于T2WI图像直方图分析(AUC=0.710),这可能是由于直方图提供的定量特征有限,难以反映图像的细微差别和更深层次的信息。Cuocolo等[20] 基于T2WI 图像构建的预测模型效能较高(AUC=0.99),但其样本量较小,AUC 值过高,模型再现性及过拟合可能存在问题。

  • 本研究存在以下局限性。①手动勾画VOI可能存在差异。但本研究VOI包含整个肿瘤的实质区可以减小由于选择单一层面带来的误差,能够更加全面地反映肿瘤内部的异质性特征。②本研究为回顾性研究,样本量相对较小。本研究通过采用合成少数过采样技术来避免两组间的样本量不平衡带来的数据偏倚,下一步将联合其他研究中心做更广泛、更深入的研究。③由于缺乏对肿瘤胶原蛋白的定量分析,本研究未对病理学和影像组学结果进行分析比较。

  • 综上所述,基于 T2WI 和 DWI 影像组学有助于预测垂体腺瘤的质地,为临床决策提供支持。

  • 参考文献

    • [1] LOPES M B S.The 2017 World Health Organization clas-sification of tumors of the pituitary gland:a summary[J].Acta Neuropathol,2017,134(4):521-535

    • [2] MOLITCH M E.Diagnosis and treatment of pituitary ade-nomas:a review[J].JAMA,2017,317(5):516-524

    • [3] ČERNÝ M,SEDLÁK V,LESÁKOVÁ V,et al.Methods of preoperative prediction of pituitary adenoma consistency:a systematic review[J].Neurosurg Rev,2022,46(1):11

    • [4] CHEN X Y,DING C Y,YOU H H,et al.Relationship be-tween pituitary adenoma consistency and extent of resec-tion based on tumor/cerebellar peduncle T2-weighted im-aging intensity(TCTI)ratio of the point on preoperative magnetic resonance imaging(MRI)corresponding to the residual point on postoperative MRI[J].Med Sci Monit,2020,26:e919565

    • [5] SU C Q,ZHANG X,PAN T,et al.Texture analysis of high b-value diffusion-weighted imaging for evaluating consis-tency of pituitary macroadenomas[J].J Magn Reson Im-aging,2020,51(5):1507-1513

    • [6] DING W,HUANG Z,ZHOU G,et al.Diffusion-weighted imaging for predicting tumor consistency and extent of re-section in patients with pituitary adenoma[J].Neurosurg Rev,2021,44(5):2933-2941

    • [7] PIERALLINI A,CARAMIA F,FALCONE C,et al.Pitu-itary macroadenomas:preoperative evaluation of consis-tency with diffusion-weighted MR imaging--initial experi-ence[J].Radiology,2006,239(1):223-231

    • [8] MOHAMED F F,ABOUHASHEM S.Diagnostic value of apparent diffusion coefficient(ADC)in assessment of pi-tuitary macroadenoma consistency[J].Egypt J Radiol Nu-cl Med,2013,44(3):617-624

    • [9] MASTORAKOS P,MEHTA G U,CHATRATH A,et al.Tumor to cerebellar peduncle T2-weighted imaging inten-sity ratio fails to predict pituitary adenoma consistency[J].J Neurol Surg B Skull Base,2019,80(3):252-257

    • [10] BARBOSA M A,PEREIRA E G R,DA MATA PEREIRA P J,et al.Diffusion-weighted imaging does not seem to be a predictor of consistency in pituitary adenomas[J].Pitu-itary,2024,27:187-196

    • [11] LIN D,LIU J,KE C,et al.Radiomics analysis of quantita-tive maps from synthetic MRI for predicting grades and molecular subtypes of diffuse gliomas[J].Clin Neuroradi-ol,2024,DOI:10.1007/s00062-024-01421-3

    • [12] 黄晓星,汪泽燕,肖学红,等.术前MRI强化特征预测胶质母细胞瘤患者术后复发风险的价值[J].临床放射学杂志,2022,41(2):217-223

    • [13] RUTKOWSKI M J,CHANG K E,CARDINAL T,et al.De-velopment and clinical validation of a grading system for pituitary adenoma consistency[J].J Neurosurg,2020,134(6):1800-1807

    • [14] SONG Y,ZHANG J,ZHANG Y D,et al.FeAture Explorer(FAE):a tool for developing and comparing radiomics models[J].PLoS One,2020,15(8):e0237587

    • [15] LI P,ZHANG D,MA S,et al.Consistency of pituitary ade-nomas:amounts of collagen types Ⅰ and Ⅲ and the pre-dictive value of T2WI MRI[J].Exp Ther Med,2021,22(5):1255

    • [16] SU C Q,WANG B B,TANG W T,et al.Diffusion-relax-ation correlation spectrum imaging for predicting tumor consistency and gross total resection in patients with pitu-itary adenomas:a preliminary study[J].Eur Radiol,2023,33(10):6993-7002

    • [17] COHEN-COHEN S,HELAL A,YIN Z,et al.Predicting pituitary adenoma consistency with preoperative magnetic resonance elastography[J].J Neurosurg.2021,136(5):1356-1363

    • [18] 林洁,苏春秋,唐文天,等.基于T1WI增强影像组学在鉴别高级别胶质瘤复发与假性进展中的应用价值[J].南京医科大学学报(自然科学版),2023,43(9):1279-1284

    • [19] ZEYNALOVA A,KOCAK B,DURMAZ E S,et al.Preop-erative evaluation of tumour consistency in pituitary mac-roadenomas:a machine learning-based histogram analysis on conventional T2-weighted MRI[J].Neuroradiology,2019,61(7):767-774

    • [20] CUOCOLO R,UGGA L,SOLARI D,et al.Prediction of pituitary adenoma surgical consistency:radiomic data mining and machine learning on T2-weighted MRI[J].Neuroradiology,2020,62(12):1649-1656

  • 参考文献

    • [1] LOPES M B S.The 2017 World Health Organization clas-sification of tumors of the pituitary gland:a summary[J].Acta Neuropathol,2017,134(4):521-535

    • [2] MOLITCH M E.Diagnosis and treatment of pituitary ade-nomas:a review[J].JAMA,2017,317(5):516-524

    • [3] ČERNÝ M,SEDLÁK V,LESÁKOVÁ V,et al.Methods of preoperative prediction of pituitary adenoma consistency:a systematic review[J].Neurosurg Rev,2022,46(1):11

    • [4] CHEN X Y,DING C Y,YOU H H,et al.Relationship be-tween pituitary adenoma consistency and extent of resec-tion based on tumor/cerebellar peduncle T2-weighted im-aging intensity(TCTI)ratio of the point on preoperative magnetic resonance imaging(MRI)corresponding to the residual point on postoperative MRI[J].Med Sci Monit,2020,26:e919565

    • [5] SU C Q,ZHANG X,PAN T,et al.Texture analysis of high b-value diffusion-weighted imaging for evaluating consis-tency of pituitary macroadenomas[J].J Magn Reson Im-aging,2020,51(5):1507-1513

    • [6] DING W,HUANG Z,ZHOU G,et al.Diffusion-weighted imaging for predicting tumor consistency and extent of re-section in patients with pituitary adenoma[J].Neurosurg Rev,2021,44(5):2933-2941

    • [7] PIERALLINI A,CARAMIA F,FALCONE C,et al.Pitu-itary macroadenomas:preoperative evaluation of consis-tency with diffusion-weighted MR imaging--initial experi-ence[J].Radiology,2006,239(1):223-231

    • [8] MOHAMED F F,ABOUHASHEM S.Diagnostic value of apparent diffusion coefficient(ADC)in assessment of pi-tuitary macroadenoma consistency[J].Egypt J Radiol Nu-cl Med,2013,44(3):617-624

    • [9] MASTORAKOS P,MEHTA G U,CHATRATH A,et al.Tumor to cerebellar peduncle T2-weighted imaging inten-sity ratio fails to predict pituitary adenoma consistency[J].J Neurol Surg B Skull Base,2019,80(3):252-257

    • [10] BARBOSA M A,PEREIRA E G R,DA MATA PEREIRA P J,et al.Diffusion-weighted imaging does not seem to be a predictor of consistency in pituitary adenomas[J].Pitu-itary,2024,27:187-196

    • [11] LIN D,LIU J,KE C,et al.Radiomics analysis of quantita-tive maps from synthetic MRI for predicting grades and molecular subtypes of diffuse gliomas[J].Clin Neuroradi-ol,2024,DOI:10.1007/s00062-024-01421-3

    • [12] 黄晓星,汪泽燕,肖学红,等.术前MRI强化特征预测胶质母细胞瘤患者术后复发风险的价值[J].临床放射学杂志,2022,41(2):217-223

    • [13] RUTKOWSKI M J,CHANG K E,CARDINAL T,et al.De-velopment and clinical validation of a grading system for pituitary adenoma consistency[J].J Neurosurg,2020,134(6):1800-1807

    • [14] SONG Y,ZHANG J,ZHANG Y D,et al.FeAture Explorer(FAE):a tool for developing and comparing radiomics models[J].PLoS One,2020,15(8):e0237587

    • [15] LI P,ZHANG D,MA S,et al.Consistency of pituitary ade-nomas:amounts of collagen types Ⅰ and Ⅲ and the pre-dictive value of T2WI MRI[J].Exp Ther Med,2021,22(5):1255

    • [16] SU C Q,WANG B B,TANG W T,et al.Diffusion-relax-ation correlation spectrum imaging for predicting tumor consistency and gross total resection in patients with pitu-itary adenomas:a preliminary study[J].Eur Radiol,2023,33(10):6993-7002

    • [17] COHEN-COHEN S,HELAL A,YIN Z,et al.Predicting pituitary adenoma consistency with preoperative magnetic resonance elastography[J].J Neurosurg.2021,136(5):1356-1363

    • [18] 林洁,苏春秋,唐文天,等.基于T1WI增强影像组学在鉴别高级别胶质瘤复发与假性进展中的应用价值[J].南京医科大学学报(自然科学版),2023,43(9):1279-1284

    • [19] ZEYNALOVA A,KOCAK B,DURMAZ E S,et al.Preop-erative evaluation of tumour consistency in pituitary mac-roadenomas:a machine learning-based histogram analysis on conventional T2-weighted MRI[J].Neuroradiology,2019,61(7):767-774

    • [20] CUOCOLO R,UGGA L,SOLARI D,et al.Prediction of pituitary adenoma surgical consistency:radiomic data mining and machine learning on T2-weighted MRI[J].Neuroradiology,2020,62(12):1649-1656