综述
影像组学在胰腺导管内乳头状黏液性肿瘤中的应用及研究进展
磁共振成像, 2022,13(11) : 154-156,168. DOI: 10.12015/issn.1674-8034.2022.11.031
摘要

胰腺导管内乳头状黏液性肿瘤(intraductal papillary mucinous neoplasm of the pancreas, IPMN)是一种具有广泛疾病谱的潜在恶性肿瘤,目前普遍认为胰腺IPMN是胰腺癌的癌前病变,术前对其进行恶性程度评估判断具有重要意义。目前常用的传统影像学方法包括CT、MRI、超声内镜(endoscopic ultrasonography, EUS)及正电子发射计算机断层显像(positron emission tomography-computed tomography, PET-CT)。影像组学通过高通量提取和定量分析影像特征为肿瘤定性提供了新方法,可对胰腺IPMN恶变潜能进行有效评估,现已逐步应用于胰腺IPMN恶性程度分级、预后预测及疗效评估等方面。本文就影像组学在胰腺IPMN恶性风险分层领域的发展及应用现状作一综述,并对未来发展做出展望。

引用本文: 赵钰莹, 许万博. 影像组学在胰腺导管内乳头状黏液性肿瘤中的应用及研究进展 [J] . 磁共振成像, 2022, 13(11) : 154-156,168. DOI: 10.12015/issn.1674-8034.2022.11.031.
参考文献导出:   Endnote    NoteExpress    RefWorks    NoteFirst    医学文献王
扫  描  看  全  文

正文
作者信息
基金 0  关键词  0
English Abstract
评论
阅读 0  评论  0
相关资源
引用 | 论文 | 视频

本刊刊出的所有论文不代表本刊编委会的观点,除非特别声明

胰腺导管内乳头状黏液性肿瘤(intraductal papillary mucinous neoplasm of the pancreas, IPMN)是一种起源于胰腺导管上皮细胞并可产生黏液的胰腺囊性或囊实性肿瘤[1, 2, 3]。根据导管受累程度将胰腺IPMN分为三种类型:主胰管型(main duct IPMN, MD-IPMN)、分支胰管型(branch duct IPMN, BD-IPMN)和混合型(mixed type IPMN, MT-IPMN);根据细胞形态学和免疫组化特征将其分为四种亚型:胃型、肠型、胰胆管型和嗜酸细胞型[4, 5];根据WHO 2019年的病理分期将其分为低级别异型增生(low-grade dysplasia, LGD)、高级别异型增生(high-grade dysplasia, HGD)和浸润性导管内乳头状黏液性癌(intraductal papillary mucinous carcinoma, IPMC)[6]。由于胰腺IPMN具有恶变潜能,最初建议采用积极的手术切除方法来预防其发展为胰腺癌[7],但有研究发现随访病例在随访期间恶性肿瘤的发生率较低,只有2.8%的患者发展为恶性肿瘤[8],并且手术切除存在术后并发症和死亡率的风险[9]。因此,目前对于胰腺IPMN的管理需要在潜在的恶变风险和胰腺切除术风险之间保持平衡,所以提升诊断准确率对于胰腺IPMN患者的诊疗方案选择及预后至关重要[10, 11]。传统影像学检查无法对胰腺IPMN风险分层做出精准判断,影像组学的日渐成熟为实现对胰腺IPMN高精度诊断提供了新机会[12]。目前影像组学在胰腺IPMN恶性程度分级、疗效评估及预后预测等领域应用日渐广泛,并取得了显著成果,但针对其在胰腺IPMN诊断价值的应用还缺乏综述性研究。因此,本文对其进行综述,以期为后续相关研究提供参考。

1 传统影像学方法在胰腺IPMN中的应用

目前最常用于胰腺IPMN诊断的传统影像学方法主要包括CT、MRI、超声内镜(endoscopic ultrasonography, EUS)以及正电子发射计算机断层显像(positron emission tomography-computed tomography, PET-CT)。各种检查方法各有利弊且互为补充。

1.1 CT在胰腺IPMN中的应用

CT扫描是诊断胰腺IPMN最常用的无创性影像学检查方法之一[13]。经CT扫描可准确显示胰管的扩张、壁结节、囊性成分等,观察病灶与周围组织器官之间的关系,为临床诊断提供可靠的影像学依据[14]

Lee等[15]通过研究发现,增强CT用于鉴别胰腺IPMN良恶性的敏感度为86%、特异度为74%。其研究结果显示,CT扫描对胰腺IPMN术前诊断及其良恶性鉴别均具有较高的能力,尤其是结合增强扫描其诊断价值更大。

1.2 MRI和MR胰胆管造影在胰腺IPMN中的应用

MRI具有较好的软组织分辨力[16],腹部MRI和MR胰胆管造影(MR cholangiopancreatography, MRCP)被认为是诊断胰腺IPMN首选的无创检查手段。MRI在识别壁结节、病灶分隔和肿瘤多灶性方面具有高敏感性。MRCP能清楚显示病变大小、部位、胰管扩张程度以及肿瘤与周围组织器官之间的关系,特别是在显示BD-IPMN囊腔与主胰管(main pancreatic duct, MPD)的相通情况中具有明显优势。囊腔与MPD相通是诊断胰腺IPMN的可靠征象,MRCP对胰腺IPMN与胰管相通的显示率高达93.7%,明显高于CT的79.2%(P<0.05)[17]。动态对比增强MRI(dynamic contrast-enhanced MRI, DCE-MRI)、扩散加权成像(diffusion-weighted imaging, DWI)与表观扩散系数(apparent diffusion coefficient, ADC)结合是有效诊断恶性胰腺IPMN的影像学方法[18],其对于准确识别直径≥5 mm壁结节具有较高的特异性及敏感性,且与胰腺IPMN恶性程度相关[19]

Min等[20]回顾性分析了接受胰腺IPMN手术患者的术前CT和MRI图像,发现所有3个高风险的特征(MPD≥10 mm,壁结节≥5 mm,梗阻性黄疸)和2个令人担忧的特征[MPD 5~9 mm,糖类抗原(carbohydrate antigen, CA)19-9水平升高]在CT和MRI上与恶性IPMN相关(P<0.05),提示CT和MRI在恶性胰腺IPMN诊断方面具有可比性(73.7% vs.75.4%;P=0.505)。

对于长期随访的患者,没有辐射损伤且无创的MRI是目前首选的检查手段。但其检查时间较长、禁忌证较多,使其应用范围有所受限。

1.3 EUS在胰腺IPMN中的应用

EUS可在内镜引导下将探头最大程度接近病灶,使其对胰腺IPMN的诊断能力明显提高,是目前临床上非常有价值的一种胰腺IPMN影像检查方法。

壁结节是预测恶性胰腺IPMN最重要的征象之一[21],而EUS最大的优势就是更好地发现胰腺IPMN的壁结节,对于胰腺IPMN早期诊断至关重要。对比增强EUS通过检测壁结节的血流信号,可以准确区分黏蛋白球和壁结节(敏感度60.0%,特异度92.9%,精确度75.9%)[22]。Lim等[23]通过研究发现,使用对比增强CT [观察者1:曲线下面积(area under the curve, AUC)=0.792;观察者2:AUC=0.830]、DCE-MRI(观察者1:AUC=0.742;观察者2:AUC=0.776)和EUS(AUC=0.733)准确识别HGD和IPMC的能力方面无明显差异。

1.4 PET-CT在胰腺IPMN中的应用

PET-CT基于肿瘤细胞具有更高的葡萄糖代谢活性,以18-氟-脱氧葡萄糖(18F-fluorodeoxyglucose, 18F-FDG)作为载体可以区分正常组织和肿瘤组织[24, 25],主要用于胰腺IPMN良恶性的鉴别诊断。Yamashita等[26]、Ohta等[27]通过回顾性分析发现对18F-FDG摄取是一种鉴别胰腺IPMN良恶性的有效标志物。但是由于PET-CT价格比较昂贵,所以在实际临床应用上不如CT、MRI及EUS等广泛。

2 影像组学在胰腺IPMN中的应用
2.1 影像组学的概念及发展

影像组学(radiomics)一词最早在2010年由美国学者Gillies等[28]首次提出,指的是将影像内包含的信息提取出来进行综合系统化分析;2012年荷兰学者Lambin等[29]进一步将影像组学的概念完善为“从放射图像中高通量地提取影像特征并创建高维数据集”。影像组学的工作流程主要包括四个部分:图像获取、图像分割、特征提取和筛选(包括形态学特征、一阶特征、二阶特征、高阶特征等)、数据的处理与分析并构建预测和分类模型[30, 31]

目前影像组学在肿瘤影像领域已经被广泛地应用,主要是在肿瘤诊断、自动结构化报告、治疗后疗效评估等方面进行了较为广泛的临床研究和科研试验[32, 33, 34, 35]。在实际临床工作中,其在肺结节检出、肺癌筛查、前列腺癌及直肠癌诊断等领域应用较为广泛[36, 37, 38]

2.2 影像组学在胰腺IPMN中的应用

胰腺IPMN具有恶变潜能,临床需借助相关检查手段并结合肿瘤标志物、肿瘤代谢物[39]等对病变风险度分层进行预测,以尽量给予患者更合适的临床处置方案。基于影像组学的图像分析可以提取肉眼无法观察到的图像特征并进行定量分析,为胰腺IPMN的危险度分层及恶性程度预测提供新的方向和思路[40, 41, 42],为胰腺IPMN的治疗、随访、预后等提供更为准确的信息。

Tobaly等[43]从胰腺IPMN患者的术前CT扫描图像中提取影像组学特征,将胰腺IPMN分为两组:良性IPMN和恶性IPMN,并采用逻辑回归方法建立多元模型。在训练队列中,85/107的影像组学特征在良性和恶性IPMN患者之间存在显著差异。无监督聚类分析显示四个不同的患者具有相似的影像组学特征,其中恶性肿瘤具有最显著的相关性。多变量模型可区分良恶性肿瘤的AUC为0.84、敏感度为82%、特异度为74%。这项研究证实了术前基于CT的影像组学分析对鉴别良恶性IPMN的高诊断性能。

Hanania等[44]对提取出的影像组学特征分析发现基于灰度共生矩阵的14个成像生物标志物能够有效区分低级别与高级别病变。当联合评价时,两组间的区分准确率可达96%,该方法的准确性优于单纯福冈标准(假阳性率为36%),提供了一种精确评估胰腺IPMN恶性潜能的定量成像方法。

Permuth等[45]从胰腺IPMN患者的术前CT图像中提取出纹理特征,然后将来自该患者队列的影像组学特征与其microRNA数据相结合,发现综合模型预测效能更高,与共识指南中的影像学特征[46]相比,无创性影像基因组学方法可以更准确地预测胰腺IPMN的病理分级。同样地,Chakraborty等[47]使用从术前CT中提取的影像组学特征作为标记物来评估胰腺IPMN恶性肿瘤的风险,结果有12个纹理特征显示出与恶性肿瘤的风险密切相关。作者还总结出与胰腺IPMN恶性风险相关的5个临床变量(手术年龄、囊肿大小、实性成分、有无腹痛和性别)。最终结果显示,将影像组学特征与临床变量相结合的组合模型预测胰腺IPMN恶性风险准确率更高。Cui等[48]从患者MRI图像中提取出影像组学特征与临床变量(CA19-9水平和MPD大小)相结合,构建联合列线图模型,研究其对BD-IPMN患者术前预测肿瘤病理分级的能力。研究结果显示该联合列线图模型可显著提高预测性能(训练队列AUC为0.903)。作者还利用来自其他机构的独立数据集,验证了所开发的预测模型的可重复性和可靠性。以上研究都证实了将生物学特征、临床特征与影像组学特征相结合的综合模型在预测肿瘤类型、分级、治疗反应和预后方面具有更高的效能,值得进一步研究。

综上,相较于传统影像学检查,影像组学可明显提高对胰腺IPMN良恶性鉴别的准确率[49]。上述影像组学研究大多基于CT成像,可能是因为CT扫描标准化程度较高。MRI在识别壁结节、病灶分隔和肿瘤多灶性方面具有高敏感性,并且可以多模态、多序列、多方位成像,被认为是诊断胰腺IPMN首选的无创检查手段。未来应考虑基于MRI的影像组学研究。另外上述研究多为单中心回顾性研究并且样本量较小,选取的样本均来自于术后病理证实的病例,不能除外模型过拟合及选择偏倚的可能,实际临床应用价值还有待进一步验证。

3 总结及展望

影像组学在多种疾病的诊断、定量成像、特征提取和模型构建等方面应用前景广大,现已逐渐应用于对胰腺常见肿瘤的诊断、生物学行为评价、预后评估、治疗反应与疗效评估等领域。影像组学的数据可以从已经获得的扫描图像中计算出来,因此它不会增加患者额外的成本或负担。影像组学可以明显提高对胰腺IPMN危险度分层诊断效能,但目前尚缺乏对HGD与IPMC的鉴别,有待进一步研究探索。目前影像组学研究范围主要是肿瘤区域,未来还需要从瘤周区域提取特征进行分析。

目前影像组学发展尚未成熟,缺乏关于图像采集和特征提取的既定协议和指导方针,同时也缺乏验证结果的指南标准,并且肿瘤区域的分割大多由放射科医生手动进行,可重复性差且工作量大。此外,几乎所有的研究都是回顾性的单中心研究,患者队列有限。未来应制订可重复的指南标准,能够使不同扫描设备、后处理软件和时间点产生一致的结果。最终目标应该是开发一个能够自动识别感兴趣区域的模型,并能够精确地描述评估相关病变。

我们期待将影像组学与基因组学、蛋白组学等其他组学结合起来,为精准化、个性化医疗提供更多有用信息,推动胰腺IPMN的精准诊疗进程。

志      谢
ACKNOWLEDGMENTS

Shandong Province Medical and Health Science and Technology Development Plan Project (No. 202109010537); Medical Scientific Research Development Fund Project: Roentgen Imaging Research Project (No. SD-202008-013).

利益冲突
作者利益冲突声明:

全体作者均声明无利益冲突。

参考文献References
[1]
European Study Group on Cystic Tumours of the Pancreas. European evidence-based guidelines on pancreatic cystic neoplasms[J]. Gut, 2018, 67(5): 789-804. DOI: 10.1136/gutjnl-2018-316027.
[2]
WashingtonMK, GoldbergRM, ChangGJ, et al. Diagnosis of digestive system tumours[J]. Int J Cancer, 2021, 148(5): 1040-1050. DOI: 10.1002/ijc.33210.
[3]
PorukKE, ValeroV, HeJ, et al. Circulating epithelial cells in intraductal papillary mucinous neoplasms and cystic pancreatic lesions[J]. Pancreas, 2017, 46(7): 943-947. DOI: 10.1097/MPA.0000000000000869.
[4]
WuJY, WangYF, LiZT, et al. Accuracy of Fukuoka and American gastroenterological association guidelines for predicting advanced neoplasia in pancreatic cyst neoplasm: a Meta-analysis[J]. Ann Surg Oncol, 2019, 26(13): 4522-4536. DOI: 10.1245/s10434-019-07921-8.
[5]
HasanA, VisrodiaK, FarrellJJ, et al. Overview and comparison of guidelines for management of pancreatic cystic neoplasms[J]. World J Gastroenterol, 2019, 25(31): 4405-4413. DOI: 10.3748/wjg.v25.i31.4405.
[6]
NagtegaalID, OdzeRD, KlimstraD, et al. The 2019 WHO classification of tumours of the digestive system[J]. Histopathology, 2020, 76(2): 182-188. DOI: 10.1111/his.13975.
[7]
ScheimanJM, HwangJH, MoayyediP. American gastroenterological association technical review on the diagnosis and management of asymptomatic neoplastic pancreatic cysts[J]. Gastroenterology, 2015, 148(4): 824-848.e22. DOI: 10.1053/j.gastro.2015.01.014.
[8]
CaprettiG, NebbiaM, GavazziF, et al. Invasive IPMN relapse later and more often in lungs in comparison to pancreatic ductal adenocarcinoma[J]. Pancreatology, 2022, 22(6): 782-788. DOI: 10.1016/j.pan.2022.05.006.
[9]
KaiserJ, ScheifeleC, HinzU, et al. IPMN-associated pancreatic cancer: survival, prognostic staging and impact of adjuvant chemotherapy[J]. Eur J Surg Oncol, 2022, 48(6): 1309-1320. DOI: 10.1016/j.ejso.2021.12.009.
[10]
O'ReillyD, FouLY, HaslerE, et al. Diagnosis and management of pancreatic cancer in adults: a summary of guidelines from the UK National Institute for Health and Care Excellence[J]. Pancreatology, 2018, 18(8): 962-970. DOI: 10.1016/j.pan.2018.09.012.
[11]
ExarchakouA, PapacleovoulouG, RousB, et al. Pancreatic cancer incidence and survival and the role of specialist centres in resection rates in England, 2000 to 2014: a population-based study[J]. Pancreatology, 2020, 20(3): 454-461. DOI: 10.1016/j.pan.2020.01.012.
[12]
ChungWY, CorreaE, YoshimuraK, et al. Using probe electrospray ionization mass spectrometry and machine learning for detecting pancreatic cancer with high performance[J]. Am J Transl Res, 2020, 12(1): 171-179.
[13]
周英文, 征锦. 胰腺导管内乳头状黏液瘤的影像学诊断进展[J/OL]. 中华消化病与影像杂志(电子版), 2015, 5(6): 45-48 [2022-04-07]. http://www.zhxhbyyxzz.com/Magazine/Show.aspx?ID=50744. DOI: 10.3877/cma.j.issn.2095-2015.2015.06.013.
ZhouYW, ZhengJ. Imaging diagnosis progress of intraductal papillary mucinous neoplasm of pancreas[J/OL]. Chin J Dig Med Imageology Electron Ed, 2015, 5(6): 45-48 [2022-04-07]. http://www.zhxhbyyxzz.com/Magazine/Show.aspx?ID=50744. DOI: 10.3877/cma.j.issn.2095-2015.2015.06.013.
[14]
刘英娜, 肖新广, 李润华, . 多层螺旋CT对胰腺导管内乳头状粘液性肿瘤的诊断及鉴别[J]. 现代医用影像学, 2019, 28(2): 310.
LiuYN, XiaoXG, LiRH, et al. Multi-slice spiral CT diagnosis and differential diagnosis of intraductal papillary mucinous tumor of pancreas[J]. Mod Med Imageology, 2019, 28(2): 310.
[15]
LeeJE, ChoiSY, MinJH, et al. Determining malignant potential of intraductal papillary mucinous neoplasm of the pancreas: CT versus MRI by using revised 2017 international consensus guidelines[J]. Radiology, 2019, 293(1): 134-143. DOI: 10.1148/radiol.2019190144.
[16]
BoughtonCK, HovorkaR. Advances in artificial pancreas systems[J/OL]. Sci Transl Med, 2019, 11(484) [2022-04-07]. https://www.science.org/doi/10.1126/scitranslmed.aaw4949?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200pubmed. DOI: 10.1126/scitranslmed.aaw4949.
[17]
孙勤学, 陈振东, 赵亦军, . 胰腺导管内乳头状粘液性肿瘤的影像分析[J]. 医学影像学杂志, 2019, 29(6): 993-996.
SunQX, ChenZD, ZhaoYJ, et al. Imaging features of intraductal papillary mucinous neoplasm of the pancreas[J]. J Med Imaging, 2019, 29(6): 993-996.
[18]
HoffmanDH, ReamJM, HajduCH, et al. Utility of whole-lesion ADC histogram metrics for assessing the malignant potential of pancreatic intraductal papillary mucinous neoplasms (IPMNs)[J]. Abdom Radiol, 2017, 42(4): 1222-1228. DOI: 10.1007/s00261-016-1001-7.
[19]
D'OnofrioM, TedescoG, CardobiN, et al. Magnetic resonance (MR) for mural nodule detection studying Intraductal papillary mucinous neoplasms (IPMN) of pancreas: imaging-pathologic correlation[J]. Pancreatology, 2021, 21(1): 180-187. DOI: 10.1016/j.pan.2020.11.024.
[20]
MinJH, KimYK, KimSK, et al. Intraductal papillary mucinous neoplasm of the pancreas: diagnostic performance of the 2017 international consensus guidelines using CT and MRI[J]. Eur Radiol, 2021, 31(7): 4774-4784. DOI: 10.1007/s00330-020-07583-1.
[21]
MarchegianiG, AndrianelloS, BorinA, et al. Systematic review, meta-analysis, and a high-volume center experience supporting the new role of mural nodules proposed by the updated 2017 international guidelines on IPMN of the pancreas[J]. Surgery, 2018, 163(6): 1272-1279. DOI: 10.1016/j.surg.2018.01.009.
[22]
OhnoE, HirookaY, ItohA, et al. Intraductal papillary mucinous neoplasms of the pancreas: differentiation of malignant and benign tumors by endoscopic ultrasound findings of mural nodules[J]. Ann Surg, 2009, 249(4): 628-634. DOI: 10.1097/SLA.0b013e3181a189a8.
[23]
LimJ, AllenPJ. The diagnosis and management of intraductal papillary mucinous neoplasms of the pancreas: has progress been made?[J]. Updates Surg, 2019, 71(2): 209-216. DOI: 10.1007/s13304-019-00661-0.
[24]
WilsonCBJH. PET scanning in oncology[J]. Eur J Cancer, 1992, 28(2/3): 508-510. DOI: 10.1016/S0959-8049(05)80089-9.
[25]
HongHS, YunMJ, ChoA, et al. The utility of F-18 FDG PET/CT in the evaluation of pancreatic intraductal papillary mucinous neoplasm[J]. Clin Nucl Med, 2010, 35(10): 776-779. DOI: 10.1097/RLU.0b013e3181e4da32.
[26]
YamashitaYI, OkabeH, HayashiH, et al. Usefulness of 18-FDG PET/CT in detecting malignancy in intraductal papillary mucinous neoplasms of the pancreas[J]. Anticancer Res, 2019, 39(5): 2493-2499. DOI: 10.21873/anticanres.13369.
[27]
OhtaK, TanadaM, SugawaraY, et al. Usefulness of positron emission tomography (PET)/contrast-enhanced computed tomography (CE-CT) in discriminating between malignant and benign intraductal papillary mucinous neoplasms (IPMNs)[J]. Pancreatology, 2017, 17(6): 911-919. DOI: 10.1016/j.pan.2017.09.010.
[28]
GilliesRJ, AndersonAR, GatenbyRA, et al. The biology underlying molecular imaging in oncology: from genome to anatome and back again[J]. Clin Radiol, 2010, 65(7): 517-521. DOI: 10.1016/j.crad.2010.04.005.
[29]
LambinP, Rios-VelazquezE, LeijenaarR, et al. Radiomics: extracting more information from medical images using advanced feature analysis[J]. Eur J Cancer, 2012, 48(4): 441-446. DOI: 10.1016/j.ejca.2011.11.036.
[30]
Amisha, MalikP, PathaniaM, et al. Overview of artificial intelligence in medicine[J]. J Family Med Prim Care, 2019, 8(7): 2328-2331. DOI: 10.4103/jfmpc.jfmpc_440_19.
[31]
SabaLC, BiswasM, KuppiliV, et al. The present and future of deep learning in radiology[J]. Eur J Radiol, 2019, 114: 14-24. DOI: 10.1016/j.ejrad.2019.02.038.
[32]
GyawaliB. Does global oncology need artificial intelligence?[J]. Lancet Oncol, 2018, 19(5): 599-600. DOI: 10.1016/S1470-2045(18)30269-9.
[33]
ShimizuY, HijiokaS, HironoS, et al. New model for predicting malignancy in patients with intraductal papillary mucinous neoplasm[J]. Ann Surg, 2020, 272(1): 155-162. DOI: 10.1097/SLA.0000000000003108.
[34]
Al EfishatMA, AttiyehMA, EatonAA, et al. Multi-institutional validation study of pancreatic cyst fluid protein analysis for prediction of high-risk intraductal papillary mucinous neoplasms of the pancreas[J]. Ann Surg, 2018, 268(2): 340-347. DOI: 10.1097/SLA.0000000000002421.
[35]
ByrneMF, ChapadosN, SoudanF, et al. Real-time differentiation of adenomatous and hyperplastic diminutive colorectal polyps during analysis of unaltered videos of standard colonoscopy using a deep learning model[J]. Gut, 2019, 68(1): 94-100. DOI: 10.1136/gutjnl-2017-314547.
[36]
ChassagnonG, VakalopoulouM, ParagiosN, et al. Artificial intelligence applications for thoracic imaging[J/OL]. Eur J Radiol, 2020, 123 [2022-04-07]. https://www.ejradiology.com/article/S0720-048X(19)30424-3/fulltext. DOI: 10.1016/j.ejrad.2019.108774.
[37]
WeisbergEM, ChuLC, ParkS, et al. Deep lessons learned: Radiology, oncology, pathology, and computer science experts unite around artificial intelligence to strive for earlier pancreatic cancer diagnosis[J]. Diagn Interv Imaging, 2020, 101(2): 111-115. DOI: 10.1016/j.diii.2019.09.002.
[38]
MazurowskiMA, BudaM, SahaA, et al. Deep learning in radiology: an overview of the concepts and a survey of the state of the art with focus on MRI[J]. J Magn Reson Imaging, 2019, 49(4): 939-954. DOI: 10.1002/jmri.26534.
[39]
HarringtonK, WilliamsT, LawrenceSA, et al. Multimodal radiomics and cyst fluid inflammatory markers model to predict preoperative risk in intraductal papillary mucinous neoplasms[J/OL]. J Med Imaging, 2020, 7 [2022-04-07]. https://www.spiedigitallibrary.org/journals/journal-of-medical-imaging/volume-7/issue-3/031507/Multimodal-radiomics-and-cyst-fluid-inflammatory-markers-model-to-predict/10.1117/1.JMI.7.3.031507.short?SSO=1. DOI: 10.1117/1.JMI.7.3.031507.
[40]
AttiyehMA, ChakrabortyJ, GazitL, et al. Preoperative risk prediction for intraductal papillary mucinous neoplasms by quantitative CT image analysis[J]. HPB, 2019, 21(2): 212-218. DOI: 10.1016/j.hpb.2018.07.016.
[41]
GaoX, WangXL. Deep learning for World Health Organization grades of pancreatic neuroendocrine tumors on contrast-enhanced magnetic resonance images: a preliminary study[J]. Int J CARS, 2019, 14(11): 1981-1991. DOI: 10.1007/s11548-019-02070-5.
[42]
BiWL, HosnyA, SchabathMB, et al. Artificial intelligence in cancer imaging: clinical challenges and applications[J]. CA Cancer J Clin, 2019, 69(2): 127-157. DOI: 10.3322/caac.21552.
[43]
TobalyD, SantinhaJ, SartorisR, et al. CT-based radiomics analysis to predict malignancy in patients with intraductal papillary mucinous neoplasm (IPMN) of the pancreas[J/OL]. Cancers (Basel), 2020, 12(11) [2022-04-07. https://www.mdpi.com/resolver?pii=cancers12113089. DOI: 10.3390/cancers12113089.
[44]
HananiaAN, BantisLE, FengZD, et al. Quantitative imaging to evaluate malignant potential of IPMNs[J]. Oncotarget, 2016, 7(52): 85776-85784. DOI: 10.18632/oncotarget.11769.
[45]
PermuthJB, ChoiJ, BalarunathanY, et al. Combining radiomic features with a miRNA classifier may improve prediction of malignant pathology for pancreatic intraductal papillary mucinous neoplasms[J]. Oncotarget, 2016, 7(52): 85785-85797. DOI: 10.18632/oncotarget.11768.
[46]
HechtEM, KhatriG, MorganD, et al. Intraductal papillary mucinous neoplasm (IPMN) of the pancreas: recommendations for Standardized Imaging and Reporting from the Society of Abdominal Radiology IPMN disease focused panel[J].Abdom Radiol, 2021, 46(4): 1586-1606. DOI: 10.1007/s00261-020-02853-4.
[47]
ChakrabortyJ, MidyaA, GazitL, et al. CTradiomics to predict high-risk intraductal papillary mucinous neoplasms of the pancreas[J]. Med Phys, 2018, 45(11): 5019-5029. DOI: 10.1002/mp.13159.
[48]
CuiSJ, TangTY, SuQM, et al. Radiomic nomogram based on MRI to predict grade of branching type intraductal papillary mucinous neoplasms of the pancreas: a multicenter study[J].Cancer Imaging, 2021, 21(1): 1-13. DOI: 10.1186/s40644-021-00395-6.
[49]
KuwaharaT, HaraK, MizunoN, et al. Usefulness of deep learning analysis for the diagnosis of malignancy in intraductal papillary mucinous neoplasms of the pancreas[J]. Clin Transl Gastroenterol, 2019, 10(5): 1-8. DOI: 10.14309/ctg.0000000000000045.
 
 
展开/关闭提纲
查看图表详情
回到顶部
放大字体
缩小字体