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:

唐立钧,E-mail:tanglijun@njmu.edu.cn

中图分类号:R737.25

文献标识码:A

文章编号:1007-4368(2024)07-1010-08

DOI:10.7655/NYDXBNSN240239

参考文献 1
JEMAL A,SIEGEL R,WARD E,et al.Cancer statistics,2009[J].CA A Cancer J Clin,2009,59(4):225-249
参考文献 2
WANG G C,ZHAO D,SPRING D J,et al.Genetics and biology of prostate cancer[J].Genes Dev,2018,32(17/18):1105-1140
参考文献 3
MILOT L.Variation of PI ⁃RADS interpretations between experts:a significant limitation[J].Radiology,2020,296(1):85-86
参考文献 4
PICKERSGILL N A,VETTER J M,ANDRIOLE G L,et al.Accuracy and variability of prostate multiparametric magnetic resonance imaging interpretation using the prostate imaging reporting and data system:a blinded comparison of radiologists[J].Eur Urol Focus,2020,6(2):267-272
参考文献 5
BI W L,HOSNY A,SCHABATH M B,et al.Artificial intelligence in cancer imaging:clinical challenges and applications[J].CA Cancer J Clin,2019,69(2):127-157
参考文献 6
ACHARYA U R,HAGIWARA Y,SUDARSHAN V K,et al.Towards precision medicine:from quantitative imaging to radiomics[J].J Zhejiang Univ Sci B,2018,19(1):6-24
参考文献 7
LAMBIN P,RIOS⁃VELAZQUEZ E,LEIJENAAR R,et al.Radiomics:extracting more information from medical images using advanced feature analysis[J].Eur J Cancer,2012,48(4):441-446
参考文献 8
SPOHN S K B,BETTERMANN A S,BAMBERG F,et al.Radiomics in prostate cancer imaging for a personalized treatment approach-current aspects of methodology and a systematic review on validated studies[J].Theranostics,2021,11(16):8027-8042
参考文献 9
LAMBIN P,LEIJENAAR R T H,DEIST T M,et al.Ra-diomics:the bridge between medical imaging and personalized medicine[J].Nat Rev Clin Oncol,2017,14(12):749-762
参考文献 10
FUJIHARA A,IWATA T,SHAKIR A,et al.Multipara-metric magnetic resonance imaging facilitates reclassification during active surveillance for prostate cancer[J].BJU Int,2021,127(6):712-721
参考文献 11
WU R C,LEBASTCHI A H,HADASCHIK B A,et al.Role of MRI for the detection of prostate cancer[J].World J Urol,2021,39(3):637-649
参考文献 12
PLOUSSARD G,ROUVIÈRE O,ROUPRÊT M,et al.The current role of MRI for guiding active surveillance in prostate cancer[J].Nat Rev Urol,2022,19(6):357-365
参考文献 13
QI Y F,ZHANG S T,WEI J W,et al.Multiparametric MRI-based radiomics for prostate cancer screening with PSA in 4 ⁃ 10 ng/mL to reduce unnecessary biopsies[J].J Magn Reson Imaging,2020,51(6):1890-1899
参考文献 14
李梦娟,张彩元,赵文露,等.基于双参数MRI的影像组学模型在临床显著性前列腺癌中的诊断价值探讨[J].磁共振成像,2022,13(11):76-81
参考文献 15
CHEN T,ZHANG Z Y,TAN S X,et al.MRI based ra-diomics compared with the PI⁃RADS V2.1 in the prediction of clinically significant prostate cancer:biparametric vs multiparametric MRI[J].Front Oncol,2021,11:792456
参考文献 16
BASS E J,PANTOVIC A,CONNOR M,et al.A systematic review and meta-analysis of the diagnostic accuracy of biparametric prostate MRI for prostate cancer in men at risk[J].Prostate Cancer Prostatic Dis,2021,24(3):596-611
参考文献 17
WOO S,SUH C H,KIM S Y,et al.Head⁃to⁃head comparison between biparametric and multiparametric MRI for the diagnosis of prostate cancer:a systematic review and meta⁃analysis[J].AJR Am J Roentgenol,2018,211(5):W226-W241
参考文献 18
ALABOUSI M,SALAMEH J P,GUSENBAUER K,et al.Biparametric vs multiparametric prostate magnetic resonance imaging for the detection of prostate cancer in treatment⁃naïve patients:a diagnostic test accuracy systematic review and meta-analysis[J].BJU Int,2019,124(2):209-220
参考文献 19
BLEKER J,KWEE T C,DIERCKX R A J O,et al.Multi-parametric MRI and auto ⁃fixed volume of interest-based radiomics signature for clinically significant peripheral zone prostate cancer[J].Eur Radiol,2020,30(3):1313-1324
参考文献 20
CASTILLO T J M,ARIF M,STARMANS M P A,et al.Classification of clinically significant prostate cancer on multi⁃parametric MRI:a validation study comparing deep learning and radiomics[J].Cancers,2021,14(1):12
参考文献 21
MOSES K A,SPRENKLE P C,BAHLER C,et al.NCCN guidelines® insights:prostate cancer early detection,version 1.2023[J].J Natl Compr Canc Netw,2023,21(3):236-246
参考文献 22
EPSTEIN J I,ZELEFSKY M J,SJOBERG D D,et al.A contemporary prostate cancer grading system:a validated alternative to the gleason score[J].Eur Urol,2016,69(3):428-435
参考文献 23
MOTTET N,VAN DEN BERGH R C N,BRIERS E,et al.EAU⁃EANM⁃ESTRO⁃ESUR⁃SIOG guidelines on prostate cancer ⁃ 2020 update.part 1:screening,diagnosis,and local treatment with curative intent[J].Eur Urol,2021,79(2):243-262
参考文献 24
KENDRICK J,FRANCIS R,HASSAN G M,et al.Ra-diomics for identification and prediction in metastatic prostate cancer:a review of studies[J].Front Oncol,2021,11:771787
参考文献 25
SHANKAR P R,KAZA R K,AL ⁃HAWARY M M,et al.Impact of clinical history on maximum PI ⁃RADS version 2 score:a six⁃reader 120⁃case sham history retrospective evaluation[J].Radiology,2018,288(1):158-163
参考文献 26
GONG L X,XU M,FANG M J,et al.Noninvasive prediction of high-grade prostate cancer via biparametric MRI radiomics[J].J Magn Reson Imaging,2020,52(4):1102-1109
参考文献 27
WANG X C,ZHANG Y,JI Z G,et al.Old men with prostate cancer have higher risk of gleason score upgrading and pathological upstaging after initial diagnosis:a systematic review and meta-analysis[J].World J Surg On-col,2021,19(1):18
参考文献 28
ZHANG G M,HAN Y Q,WEI J W,et al.Radiomics based on MRI as a biomarker to guide therapy by predicting upgrading of prostate cancer from biopsy to radical prostatectomy[J].J Magn Reson Imaging,2020,52(4):1239-1248
参考文献 29
OGBONNAYA C N,ZHANG X Y,ALSAEDI B S O,et al.Prediction of clinically significant cancer using radiomics features of pre⁃biopsy of multiparametric MRI in men suspected of prostate cancer[J].Cancers,2021,13(24):6199
参考文献 30
MAKOWSKI M R,BRESSEM K K,FRANZ L,et al.De novo radiomics approach using image augmentation and features from T1 mapping to predict gleason scores in prostate cancer[J].Invest Radiol,2021,56(10):661-668
参考文献 31
CASTILLO T J M,STARMANS M P A,ARIF M,et al.A multi⁃center,multi⁃vendor study to evaluate the generalizability of a radiomics model for classifying prostate cancer:high grade vs.low grade[J].Diagnostics,2021,11(2):369
参考文献 32
ALBERTSEN P C.Radical prostatectomy or watchful waiting in prostate cancer ⁃ 29⁃ year follow⁃up[J].Eur Urol,2020,78(3):471
参考文献 33
JEONG B C,CHALFIN H J,LEE S B,et al.The relation⁃ship between the extent of extraprostatic extension and survival following radical prostatectomy[J].Eur Urol,2015,67(2):342-346
参考文献 34
LITWIN M S,TAN H J.The diagnosis and treatment of prostate cancer:a review[J].JAMA,2017,317(24):2532-2542
参考文献 35
ROOIJ M D,HAMOEN E H,WITJES J A,et al.Accuracy of magnetic resonance imaging for local staging of prostate cancer:a diagnostic meta-analysis[J].Eur Urol,2016,70(2):233-245
参考文献 36
STANZIONE A,CUOCOLO R,COCOZZA S,et al.Detection of extraprostatic extension of cancer on biparametric MRI combining texture analysis and machine learning:preliminary results[J].Acad Radiol,2019,26(10):1338-1344
参考文献 37
MA S,XIE H H,WANG H H,et al.Preoperative prediction of extracapsular extension:radiomics signature based on magnetic resonance imaging to stage prostate cancer[J].Mol Imaging Biol,2020,22(3):711-721
参考文献 38
LOSNEGÅRD A,REISAETER L A R,HALVORSEN O J,et al.Magnetic resonance radiomics for prediction of ex-traprostatic extension in non ⁃favorable intermediate-and high⁃risk prostate cancer patients[J].Acta Radiol,2020,61(11):1570-1579
参考文献 39
MA S,XIE H H,WANG H H,et al.MRI⁃based radiomics signature for the preoperative prediction of extracapsular extension of prostate cancer[J].J Magn Reson Imaging,2019,50(6):1914-1925
参考文献 40
XU L L,ZHANG G,ZHAO L,et al.Radiomics based on multiparametric magnetic resonance imaging to predict extraprostatic extension of prostate cancer[J].Front On-col,2020,10:940
参考文献 41
HE D,WANG X M,FU C C,et al.MRI⁃based radiomics models to assess prostate cancer,extracapsular extension and positive surgical margins[J].Cancer Imaging,2021,21(1):46
参考文献 42
BAI H L,XIA W,JI X F,et al.Multiparametric magnetic resonance imaging⁃based peritumoral radiomics for preoperative prediction of the presence of extracapsular extension with prostate cancer[J].J Magn Reson Imaging,2021,54(4):1222-1230
参考文献 43
ALGOHARY A,SHIRADKAR R,PAHWA S,et al.Combination of peri ⁃tumoral and intra ⁃tumoral radiomic features on Bi-parametric MRI accurately stratifies prostate cancer risk:a multi⁃site study[J].Cancers,2020,12(8):2200
参考文献 44
ZHAI T,HU L,MA W,et al.Peri⁃prostatic adipose tissue measurements using MRI predict prostate cancer aggressiveness in men undergoing radical prostatectomy[J].J Endocrinol Invest,2021,44(2):287-296
参考文献 45
SALJI M,HENDRY J,PATEL A,et al.Peri⁃prostatic fat volume measurement as a predictive tool for castration resistance in advanced prostate cancer[J].Eur Urol Focus,2018,4(6):858-866
参考文献 46
CUOCOLO R,STANZIONE A,FALETTI R,et al.MRI index lesion radiomics and machine learning for detection of extraprostatic extension of disease:a multicenter study[J].Eur Radiol,2021,31(10):7575-7583
参考文献 47
FAN X H,XIE N,CHEN J W,et al.Multiparametric MRI and machine learning based radiomic models for preoperative prediction of multiple biological characteristics in prostate cancer[J].Front Oncol,2022,12:839621
参考文献 48
ZHONG Q Z,LONG L H,LIU A,et al.Radiomics of mul-tiparametric MRI to predict biochemical recurrence of localized prostate cancer after radiation therapy[J].Front Oncol,2020,10:731
参考文献 49
LI L,SHIRADKAR R,LEO P,et al.A novel imaging based nomogram for predicting post ⁃surgical biochemical recurrence and adverse pathology of prostate cancer from preoperative bi-parametric MRI[J].EBio Med,2021,63:103163
参考文献 50
SHIRADKAR R,PODDER T K,ALGOHARY A,et al.Radiomics based targeted radiotherapy planning(Rad-TRaP):a computational framework for prostate cancer treatment planning with MRI[J].Radiat Oncol,2016,11(1):148
参考文献 51
DANIEL M,KUESS P,ANDRZEJEWSKI P,et al.Impact of androgen deprivation therapy on apparent diffusion coefficient and T2W MRI for histogram and texture analysis with respect to focal radiotherapy of prostate cancer[J].Al,2019,195(5):402-411
参考文献 52
ABDOLLAHI H,TANHA K,MOFID B,et al.MRI ra-diomic analysis of IMRT⁃induced bladder wall changes in prostate cancer patients:a relationship with radiation dose and toxicity[J].J Med Imaging Radiat Sci,2019,50(2):252-260
参考文献 53
ABDOLLAHI H,MOFID B,SHIRI I,et al.Machine learning⁃based radiomic models to predict intensity⁃modulated radiation therapy response,Gleason score and stage in prostate cancer[J].Radiol Med,2019,124(6):555-567
参考文献 54
SULLIVAN L,WILLIAMS S G,TAI K H,et al.Urethral stricture following high dose rate brachytherapy for prostate cancer[J].Radiother Oncol,2009,91(2):232-236
参考文献 55
TSANG Y M,VIGNARAJAH D,MCWILLIAM A,et al.A pilot study on dosimetric and radiomics analysis of urethral strictures following HDR brachytherapy as monotherapy for localized prostate cancer[J].Br J Radiol,2020,93(1106):20190760
参考文献 56
ABDOLLAHI H,MAHDAVI S R,SHIRI I,et al.Magnetic resonance imaging radiomic feature analysis of radiation ⁃induced femoral head changes in prostate cancer radio⁃therapy[J].J Cancer Res Ther,2019,15(Supplement):S11-S19
参考文献 57
THARMALINGAM H,TSANG Y M,ALONZI R,et al.Changes in magnetic resonance imaging radiomic features in response to androgen deprivation therapy in patients with intermediate-and high⁃risk prostate cancer[J].Clin Oncol,2022,34(6):246-253
参考文献 58
ZAMBOGLOU C,BETTERMANN A S,GRATZKE C,et al.Uncovering the invisible-prevalence,characteristics,and radiomics feature-based detection of visually undetectable intraprostatic tumor lesions in 68GaPSMA⁃11 PET images of patients with primary prostate cancer[J].Eur J Nucl Med Mol Imaging,2021,48(6):1987-1997
参考文献 59
FELICIANI G,CELLI M,FERRONI F,et al.Radiomics analysis on[68Ga]Ga-PSMA ⁃11 PET and MRI ⁃ADC for the prediction of prostate cancer ISUP grades:preliminary results of the BIOPSTAGE trial[J].Cancers,2022,14(8):1888
参考文献 60
ERLE A,MOAZEMI S,LÜTJE S,et al.Evaluating a machine learning tool for the classification of pathological uptake in whole⁃body PSMA⁃PET⁃CT scans[J].Tomography,2021,7(3):301-312
参考文献 61
CYSOUW M C F,JANSEN B H E,VAN DE BRUG T,et al.Machine learning⁃based analysis of[18F]DCFPyL PET radiomics for risk stratification in primary prostate cancer[J].Eur J Nucl Med Mol Imaging,2021,48(2):340-349
参考文献 62
OSMAN S S,LEIJENAAR R T,COLE A J,et al.Computed tomography-based radiomics for risk stratification in prostate cancer[J].Int J Radiat Oncol Biol Phys,2019,105(2):448-456
参考文献 63
ACAR E,LEBLEBICI A,ELLIDOKUZ B E,et al.Machine learning for differentiating metastatic and completely responded sclerotic bone lesion in prostate cancer:a retrospective radiomics study[J].Br J Radiol,2019,92(1101):20190286
参考文献 64
PEEKEN J C,SHOUMAN M A,KROENKE M,et al.A CT-based radiomics model to detect prostate cancer lymph node metastases in PSMA radioguided surgery patients[J].Eur J Nucl Med Mol Imaging,2020,47(13):2968-2977
参考文献 65
MOSTAFAEI S,ABDOLLAHI H,KAZEMPOUR DEH-KORDI S,et al.CT imaging markers to improve radiation toxicity prediction in prostate cancer radiotherapy by stacking regression algorithm[J].Radiol Med,2020,125(1):87-97
参考文献 66
OU W,LEI J H,LI M H,et al.Ultrasound-based ra-diomics score for pre⁃biopsy prediction of prostate cancer to reduce unnecessary biopsies[J].Prostate,2023,83(1):109-118
参考文献 67
LIANG L,ZHI X,SUN Y,et al.A nomogram based on a multiparametric ultrasound radiomics model for discrimination between malignant and benign prostate lesions[J].Front Oncol,2021,11:610785
参考文献 68
WILDEBOER R R,MANNAERTS C K,VAN SLOUN R J G,et al.Automated multiparametric localization of prostate cancer based on B ⁃mode,shear ⁃wave elastography,and contrast ⁃enhanced ultrasound radiomics[J].Eur Ra-diol,2020,30(2):806⁃815
参考文献 69
ZHANG Q,XIONG J Y,CAI Y H,et al.Multimodal feature learning and fusion on B ⁃mode ultrasonography and sonoelastography using point-wise gated deep networks for prostate cancer diagnosis[J].Biomed Tech,2020,65(1):87-98
目录contents

    摘要

    前列腺癌(prostate cancer,PCa)是全世界第二大常见的男性恶性肿瘤,对其早期筛查、辅助诊断、风险分层、治疗指导和疗效评估需要更加全面和一体化的方法。影像组学是一种以定量方式高通量提取和分析图像特征的技术,目前已有大量研究表明它对PCa的无创诊断、侵袭性评估、疗效及预后评估均有重要的临床价值。随着当今成像方式的快速进步和人工智能技术的迅速发展,影像组学技术拥有巨大潜力。文章回顾了基于多参数磁共振成像、正电子发射计算机断层扫描、计算机断层扫描、超声等不同成像方式的影像组学在PCa中的现有研究,探讨其在未来应用中的潜力。

    Abstract

    Prostate cancer(PCa)is the second most common malignancy among men worldwide. A more comprehensive and integrated approach is needed for its early screening,auxiliary diagnosis,risk stratification,treatment guidance,and efficacy assessment. Radiomics,a technique that extracts and analyzes image features in a quantitative and high-throughput manner,has shown significant clinical value in the non-invasive diagnosis,aggressiveness assessment,treatment efficacy,and prognosis evaluation of PCa. With the rapid development of imaging modalities and artificial intelligence technologies,radiomics has tremendous potential. This article reviews the existing researches on radiomics based on multiparametric magnetic resonance imaging,positron emission tomography/computed tomography,computed tomography,ultrasound and other imaging modalities in PCa,emphasizing its potential in future applications.

  • 前列腺癌(prostate cance,PCa)是男性泌尿生殖系统最常见的恶性肿瘤,是全球男性第二大常见癌症,也是男性癌症相关死亡的第五大原因[1]。由于早期诊断的PCa患者生存率远高于晚期诊断患者,及时发现PCa对降低病死率非常重要[2]。随着成像方式的快速进步,多参数磁共振成像(multi⁃parame⁃ tric magnetic resonance imaging,mpMRI)已经成为 PCa检出和预测侵袭性的主要工具,但mpMRI存在假阳性和假阴性情况[3],且观察者间一致性较差[4],很大程度依赖影像医师的经验和主观判断。前列腺特异性膜抗原(prostate specific membrane antigen, PSMA)正电子发射计算机断层扫描(positron emis⁃ sion tomography/computed tomography,PET/CT)可用于PCa的分期和复发性PCa治疗后再分期,比mpMRI 灵敏度更高,但特异度较低。计算机断层扫描(com⁃ puted tomography,CT)和超声也可协助PCa的筛查,但灵敏度和特异度较低。而影像组学技术通过分析医学图像中获得的数据,转定性评估为定量评估,为PCa的无创精准检出提供了新思路。人工智能(artificial intelligence,AI)技术是影像组学的重要方法学基础[5],影像组学和AI技术的有力结合使一些影像组学特征(radiomics features,RF)成为稳健可靠的标志物,反映肿瘤内部的异质性和生物学表型[6]。文章基于 mpMRI、PET/CT、CT、超声等不同成像方式影像组学的最新研究进行综述,并介绍AI技术如何改善影像组学并转化为临床实践。

  • 1 影像组学概述

  • 影像组学概念最早在2012年由Lambin等[7] 提出,它是一种从标准医学图像中高通量挖掘定量图像特征的技术,通过原始图像信息、算法、统计分析等工具的融合,提供比传统影像学更加丰富的深层信息,从而实现对疾病的精准诊疗。影像组学的基本步骤包括:图像的采集和处理、感兴趣区域(region of interes, ROI)的分割、特征的提取和选择、模型的构建和验证。影像组学研究能够成立的最基本假设是,肿瘤的宏观影像特征与基因、蛋白质、细胞、代谢等微观信息在时间和空间水平上的改变息息相关[8-9]

  • 2 MRI影像组学在PCa中的应用

  • 2.1 PCa的诊断

  • mpMRI目前是诊断PCa的首选影像学方法,主要包含T2加权图像(T2 weighted imaging,T2WI)、弥散加权图像(diffusion weighted imaging,DWI)、表观弥散系数(apparent diffusion coefficient,ADC)图像和动态增强(dynamic contrast enhancement,DCE)图像,在PCa早期筛查、辅助诊断、风险分层、指导治疗和疗效评估中发挥了重要作用[10-12]

  • Qi 等[13] 评估整合了 mpMRI 的 RF 和临床⁃影像学危险因素的联合模型在前列腺特异性抗原(pros⁃ tate specific antigen,PSA)为 4~10 ng/mL 的可疑 PCa 患者中的诊断效能,结果显示联合模型优于单纯的临床⁃影像学模型(P <0.05),并识别出更多的阴性患者,使18.4%的患者避免不必要的穿刺。李梦娟等[14] 基于双参数MRI(biparameter MRI,bpMRI)图像构建影像组学模型来预测临床显著性前列腺癌(clin⁃ ically significant prostate cancer,csPCa),发现灰度共生矩阵(gray⁃level co⁃occurrence matrix,GLCM)和形态特征在所有特征中占比较大,但该研究样本量小,未对病灶位置进行分类,且缺乏组织病理分析。Chen 等[15] 比较了影像组学与前列腺影像报告和数据系统(prostate imaging reporting and data sys⁃ tem,PI⁃RADS)评分在 bpMRI 与 mpMRI 检测 csPCa 中的表现,最终显示bpMRI及mpMRI影像组学模型性能均优于 PI ⁃ RADS 评分系统(AUC:0.953 vs. 0.853、0.968 vs. 0.863),且两个影像组学模型间性能差异无统计学意义(P >0.05)。这符合目前的研究趋势,bpMRI省略DCE成像,可以减少时间成本和检查费用,同时获得近似mpMRI的诊断准确性[16-18]

  • 随着AI技术与影像组学方法的结合,研究者可以通过机器学习(machine learning,ML)甚至深度学习(deep learning,DL)方法评价医学图像,有效减少了各种潜在偏差。Bleker等[19] 创建了一种自动固定 ROI 的方法,使用 ML 算法基于 mpMRI 图像构建影像组学模型以识别外周带 csPCa,结果显示多变量特征选择和极致梯度提升算法的性能优于单变量特征选择和随机森林算法(AUC:0.870 vs. 0.780),且 DCE特征并不能显著提高诊断性能,这与先前得到的结论相同[15]。但该模型仅适用于外周带病变,移行带病变仍需要构建专用模型,且需要在外部验证集中进一步评估该模型的普适性。Castillo 等[20] 为了比较 DL 模型和影像组学模型对 csPCa 的诊断性能,分别对3个不同机构的患者进行测试,结果显示 DL模型在训练集中表现更好,但影像组学模型在所有验证集中均优于DL模型。这是第1个进行这种比较的研究,但需要注意的是,DL模型识别整个前列腺图像中的 csPCa,而影像组学模型识别勾画的 ROI中的csPCa。

  • 2.2 PCa的Gleason 评分(Gleason score,GS)及侵袭性预测

  • 前列腺穿刺活检得到GS是目前最广泛采用的衡量PCa侵袭性水平的指标[21]。研究表明,低风险 (GS<7)、中风险(GS=7)和高风险(GS>7)PCa的5年无生化复发(biochemical recurrence,BCR)率差异较大[22],治疗策略也差异较大[23],因此准确的风险分层对PCa治疗决策的制定和预后判断有至关重要的指导意义。但活检并不能反映整个病灶的异质性,并且有出血、感染等风险[24]。而mpMRI由于其阳性预测值较低[25],往往又会导致很多不必要的活检。因此,影像组学的出现为 PCa 侵袭性的无创预测提供可能。

  • Gong等[26] 基于术前bpMRI 图像分别构建T2WI 模型、DWI模型和联合模型来无创预测高风险PCa,结果表明联合模型预测能力最佳(AUC=0.788),但较DWI模型并未显著提高(P >0.05)。众所周知病变活检后的GS与相应根治性前列腺切除术(radical prostatectomy,RP)后的 GS 不一致仍然是一个未解决的问题[27],而Zhang等[28] 研究表明,基于mpMRI的影像组学模型对 PCa 从活检到 RP 的 GS 升级有良好的预测性能,并开发了一种简单易用的诺莫图,帮助临床医生根据患者 GS 升级的风险决定继续实行主动监测或手术根治等。Ogbonnaya等[29] 全部使用 RP 标本作为参考标准,从 bpMRI 图像中提取 GLCM特征,筛选能独立预测csPCa的RF并构建模型,该研究是第1个使用3D特异性模具将影像学和组织病理学信息整合在一起的前瞻性研究。

  • 大量研究表明 mpMRI 的计算机辅助分析可以帮助进行PCa的分级及风险分层,Makowski等[30] 使用支持向量机(support vector machine,SVM)、随机梯度提升、随机森林和 K ⁃最近邻(K ⁃ nearest neighbor, KNN)这4种不同的ML算法训练模型对PCa进行风险分层,并评估T1 mapping特征的附加价值,结果显示SVM的准确率最高(AUC=0.92),且T1 mapping定量技术可提高影像组学模型的准确性,但该研究样本量小,无法充分确定T1 mapping定量技术的潜在价值。最近,Castillo等[31] 在一项多中心研究中使用 ML算法构建区分高风险和低风险PCa的影像组学模型,并与影像医师的诊断进行比较,结果显示影像组学模型优于影像医师的诊断。这是第1个将影像组学模型的普适性作为主要关注点的研究,也提示要在多中心环境中训练验证影像组学模型后,才能将其应用于临床决策系统中。

  • 2.3 PCa 包膜外侵犯(extracapsular extension,ECE) 的预测

  • ECE 会导致 PCa 患者发生手术切缘阳性(posi⁃ tive surgical margins,PSM)和微转移的概率增加,也与不良预后和BCR相关[32-33],对于这些患者需要考虑切除血管神经束或辅助治疗等[34]。因此,术前准确识别 ECE 对于治疗方案的制定至关重要。一项 Meta分析显示mpMRI术前预测ECE的灵敏度仅为 57%[35],而影像组学的出现提高了ECE检测的准确性[36-38]

  • Ma等[39] 基于T2WI图像构建影像组学模型,训练集和验证集的AUC分别为0.902和0.883,与影像医师的诊断灵敏度(46.88%~50.00%)相比,灵敏度 (75.00%)显著提高,特异度差异无统计学意义,与先前的研究结果一致[35],但该研究仅评估了 T2WI 图像的RF。Xu等[40] 基于mpMRI图像构建影像组学模型以术前预测ECE,结果显示其诊断效能明显优于临床模型(0.865 vs. 0.658,P=0.02),决策曲线分析也证实该模型有良好的临床实用性。He 等[41] 基于 T2WI和ADC图像构建影像组学模型分别预测ECE 和 PSM,结果显示 ADC 模型性能最佳(AUC:0.625、 0.733),结合临床因素后的联合模型性能提高到 0.728 和 0.766,也说明联合模型可提高对 ECE 和 PSM 的风险预测能力。另外该研究从原发灶中提取 RF,说明 ECE 的发生与瘤内的异质性相关。而 Bai 等[42] 基于瘤周区域构建影像组学模型来预测 ECE,结果显示瘤周区域的RF较瘤内区域能更好地预测ECE(P <0.05)。研究认为瘤周组织及微环境能够提供肿瘤诊断和预后相关的信息[43],如瘤周血管和软组织、脂肪组织的变化可能与包膜受侵袭程度有关[44-45]

  • 目前预测ECE的 ML研究相对较少,Cuocolo等[46] 基于 T2WI 和 ADC 图像采用 SVM 算法构建 ECE 预测模型,并在多中心验证模型,结果显示模型在训练集中的准确率达83%,在两个外部验证集中的准确率达79%和74%,且与影像医师的诊断准确率差异无统计学意义,多中心验证的设计直接证明该模型具有良好的普适性,将其纳入ECE评分系统有望进一步提高诊断的准确性。Fan等[47] 基于mpMRI图像使用6种算法分别建立与PCa侵袭性相关的5种生物学特征(Ki⁃67、S100、ECE、PSM和神经周围侵犯) 的影像组学模型并验证诊断效能,结果显示随机森林算法对Ki⁃67、S100、ECE、神经周围侵犯的预测性能最好(AUC:0.87、0.80、0.85、0.82),SVM对PSM的预测性能最好(AUC:0.77)。该研究证明了影像组学在预测PCa多种生物学特征方面的适用性,并提供ML参数为未来的标准化工作提供参考。

  • 2.4 PCa BCR的预测

  • BCR是PCa患者治疗失败及死亡的主要原因,早期准确预测 BCR 有助于指导临床合理制定治疗方案,改善预后。

  • Zhong 等[48] 使用基于卷积神经网络(convolu⁃ tional neural network,CNN)的 ML,构建 mpMRI 影像组学模型来预测局限性 PCa 患者放射治疗(radia⁃ tion therapy,RT)后的BCR,结果表明该模型的总体准确率为74.1%,首次证明了mpMRI影像组学在评估局限性PCa患者RT 后预后方面的潜力。在此基础上,Li 等[49] 联合术前 bpMRI 的 RF、GS 分级分组和 PSA 水平建立 RadClip 诺莫图来预测 RP 术后的无 BCR 生存和不良病理,结果显示 RadClip 的模型 C 指数为 0.77,AUC 为 0.71,其表现优于现有的预测工具如前列腺癌风险评估评分和 Decipher 基因检测,有助于早期识别可以从新辅助治疗中获益的潜在患者。

  • 2.5 PCa的疗效评估

  • 对于局部进展性 PCa,RP 联合内分泌治疗、放化疗、免疫治疗、靶向治疗等是目前常用的治疗方法,可以术前降低肿瘤分期,术后消灭残留的病灶及微小转移灶,提高患者生存率[23]。随着诸多新型治疗方法的问世,这些方法对PCa的疗效和不良反应评估成为了关键[50-52]

  • 在RT方面,Abdollahi等[53] 根据RT前后ADC值的变化来评估 33 例接受调强 RT 的 PCa 患者的疗效,并基于 RT 前后的 bpMRI 图像开发影像组学模型,发现 RT 前后 T2WI 和 ADC 图像中提取的 RF 均可以预测早期 RT 反应。与以往研究不同的是,该研究结合 RT 后的图像特征,量化了肿瘤治疗后内部异质性的变化。尽管该研究样本较少,未设置验证组,但其先进的ML和影像组学方法,开拓了PCa 个体化治疗的第1步。关于RT后的不良反应,尿道狭窄是PCa高剂量近距离RT的特异性晚期反应[54], Tsang等[55] 探索非恶性尿道狭窄形成与沿尿道的剂量分布及 RF 的关系,发现剂量分布与尿道狭窄并无关联,而同质性和对比度特征能够识别 RT 后尿道狭窄。另外Abdollahi等[56] 还发现mpMRI的RF有早期预测RT后股骨骨折的潜力。

  • 在内分泌治疗方面,鉴于接受雄激素剥夺疗法 (androgen deprivation therapy,ADT)后的 PCa 患者 PSA通常降至较低水平,限制了PSA在识别治疗反应梯度的作用。因此 Tharmalingam 等[57] 研究 ADT 前后MRI纹理特征的差异,发现能量及同质性特征在治疗后良性组织中显著增加,在恶性组织中显著降低,证实了 RF 可以作为预测 PCa 对 ADT 敏感性的标志物。

  • 3 PSMA⁃PET/CT影像组学在PCa中的应用

  • PSMA⁃PET/CT是PCa最有前景的成像方式,多项研究证明PSMA⁃PET/CT 在NM 分期上比MRI、增强CT或Choline⁃PET/CT更准确,PSMA⁃PET/CT已成为复发性PCa治疗后再分期的金标准[23]。但目前基于PSMA⁃PET/CT的影像组学研究仍然不多。

  • Zamboglou等[58] 探索了基于68Ga⁃PSMA⁃11 PET/CT 的 RF 检测 CT 无法检出的 csPCa 的可行性,最终研究发现两个RF(大小区域不均匀归一化和小区域强调)对漏诊的病灶有优秀的检测能力(AUC≥0.93)。Feliciani 等[59] 基于 68Ga⁃PSMA⁃11 PET/CT 和 ADC 图像,筛选能够区分低级别(GS=3+3)和高级别(GS>3+3)PCa 的 RF,以减少对早期惰性 PCa 的过度诊疗,但该研究的局限性为 PET/CT 和 ADC 图像一致的病变数量较少。Erle 等[60] 基于决策树分类器构建 68Ga⁃PSMA⁃11 PET/CT 影像组学模型来鉴别晚期 PCa患者的生理性和病理性摄取,证实了影像组学模型对晚期 PCa 患者 M 和 N 分期的准确性。Cysouw 等[61]18F⁃DCFPyL PET/CT 图像上评估影像组学模型能否预测PCa的淋巴结转移、远处转移、GS≥8和 ECE,结果显示影像组学模型对PCa的GS评估和转移倾向有良好的预测能力(AUC:0.86),优于传统 PET参数模型(AUC:0.81)。

  • 4 CT影像组学在PCa中的应用

  • 虽然 CT成像在检测PCa病变方面缺乏特异度和灵敏度,但 CT 图像可以提供 NM 分期信息,特别是对有MRI禁忌证的患者,CT影像组学可以提供补充信息来完善PCa的诊疗方式。

  • Osman 等[62] 对 PCa 风险分层进行了高精度分类,探索基于CT的RF对PCa侵袭性预测的价值,发现区分GS≤6与GS≥7、GS=3+4与GS=4+3、低风险组与高风险组、低风险组与中风险组的 AUC 分别为 0.90、0.98、0.96、1.00,具有良好的灵敏度和特异度。Acar 等[63] 使用基于 ML 的 CT 纹理特征来鉴别 68Ga⁃PSMA PET/CT图像上有PSMA表达的骨转移灶和无PSMA表达的硬化灶,研究结果显示短区域高灰度强调和峰值是最有意义的两个 RF,且 KNN 模型预测性能最佳(AUC:0.76),证实 CT 影像组学在鉴别骨转移灶中的潜力。Peeken 等[64] 开发了 1 个 CT影像组学模型,以预测接受PSMA引导手术治疗 (即在术中检测和切除PSMA⁃PET/CT阳性淋巴结的新型手术方法)后的复发性PCa患者的淋巴结转移状态,结果显示联合影像组学模型的性能最佳 (AUC:0.95),优于传统 CT 参数如淋巴结短径为基础的模型(AUC:0.84)等。Mostafaei 等[65] 在一项前瞻性研究中探索了CT的RF和临床⁃剂量学参数对 RT引起的泌尿道和胃肠道不良反应(≥1级膀胱炎和直肠炎)的预测能力,结果显示CT影像组学模型对 RT 不良反应有较好的预测潜力(AUC:0.71、 0.71),优于临床模型(AUC:0.67、0.66)。

  • 5 超声影像组学在PCa中的应用

  • 超声作为一种非侵入性、无辐射、经济实用的成像方法,在前列腺活检中起着不可或缺的作用。近年来随着超声成像方式的增多,其影像组学研究也在逐渐增多。

  • Ou 等[66] 开发了一个基于经直肠超声(transrec⁃ tal ultrasound,TRUS)的影像组学模型,并联合临床危险因素构建诺莫图以在活检前预测PCa,研究结果显示部分RF是预测PCa的独立因素。Liang等[67] 通过TRUS和剪切波弹性成像(shear⁃wave elastogra⁃ phy,SWE)构建多参数超声影像组学模型来鉴别前列腺良恶性病变,结果显示该模型的 AUC 为 0.85,结合了临床参数的联合模型AUC为0.90,证实了多参数超声成像模型对PCa诊断的准确性。与之相似的,Wildeboer等[68] 通过ML将B超、SWE、动态对比增强超声造影(dynamic contrast ⁃ enhanced ultrasound, DCE⁃US)自身的定量参数与基于它们的RF结合构建多参数超声影像组学模型,评估其在PCa诊断中的潜力,研究结果显示该模型对 PCa 和 csPCa 的诊断性能良好(AUC:0.75、0.90),优于表现最佳的定量参数即灌注速度构建的模型(AUC:0.69、0.76),该模型性能的改善应归功于 RF 以及多参数联合。 Zhang 等[69] 构建了 1 个基于神经网络的DL技术,用于学习和融合从B超和实时弹性成像中提取的RF,以在计算机辅助分析下诊断PCa,为疾病诊断提供较少的主观和人为干预,最终显示出良好的诊断能力(AUC:0.851)。

  • 6 问题与展望

  • PCa 的影像组学研究越来越受到重视,但在未来的研究和应用中仍存在许多挑战。第一,目前大多数研究是单中心、回顾性研究,样本量小,缺乏外部验证,限制了研究结果的准确性;第二,由于PCa 的多灶性及异质性,穿刺活检作为金标准有可能导致漏诊或误诊,直接影响影像组学诊断的准确性; 第三,虽然人工分割是目前描绘ROI最准确的方法,但观察者间和观察者内的变异性难以避免,未来需要进一步改进自动分割算法的可重复性以提高定位的准确性;第四,由于缺乏统一标准,不同的机器及采集方案等限制了影像组学模型的临床应用,但 AI和ML技术的兴起可以限制这种偏差,未来仍需将这些技术与传统成像方式进行比较;第五,由于所选特征与临床结果之间的关系缺乏透明的解释,未来需要进一步了解影像组学背后的决策过程以提高可信度。随着影像组学应用潜力的证实,未来研究应侧重于影像组学与其他非成像生物标志物的结合,比如基因组学,甚至可以综合基因组学、转录组学、蛋白质组学、代谢组学等,开发新的生物标志物,以加强对患者的临床管理。

  • 参考文献

    • [1] JEMAL A,SIEGEL R,WARD E,et al.Cancer statistics,2009[J].CA A Cancer J Clin,2009,59(4):225-249

    • [2] WANG G C,ZHAO D,SPRING D J,et al.Genetics and biology of prostate cancer[J].Genes Dev,2018,32(17/18):1105-1140

    • [3] MILOT L.Variation of PI ⁃RADS interpretations between experts:a significant limitation[J].Radiology,2020,296(1):85-86

    • [4] PICKERSGILL N A,VETTER J M,ANDRIOLE G L,et al.Accuracy and variability of prostate multiparametric magnetic resonance imaging interpretation using the prostate imaging reporting and data system:a blinded comparison of radiologists[J].Eur Urol Focus,2020,6(2):267-272

    • [5] BI W L,HOSNY A,SCHABATH M B,et al.Artificial intelligence in cancer imaging:clinical challenges and applications[J].CA Cancer J Clin,2019,69(2):127-157

    • [6] ACHARYA U R,HAGIWARA Y,SUDARSHAN V K,et al.Towards precision medicine:from quantitative imaging to radiomics[J].J Zhejiang Univ Sci B,2018,19(1):6-24

    • [7] LAMBIN P,RIOS⁃VELAZQUEZ E,LEIJENAAR R,et al.Radiomics:extracting more information from medical images using advanced feature analysis[J].Eur J Cancer,2012,48(4):441-446

    • [8] SPOHN S K B,BETTERMANN A S,BAMBERG F,et al.Radiomics in prostate cancer imaging for a personalized treatment approach-current aspects of methodology and a systematic review on validated studies[J].Theranostics,2021,11(16):8027-8042

    • [9] LAMBIN P,LEIJENAAR R T H,DEIST T M,et al.Ra-diomics:the bridge between medical imaging and personalized medicine[J].Nat Rev Clin Oncol,2017,14(12):749-762

    • [10] FUJIHARA A,IWATA T,SHAKIR A,et al.Multipara-metric magnetic resonance imaging facilitates reclassification during active surveillance for prostate cancer[J].BJU Int,2021,127(6):712-721

    • [11] WU R C,LEBASTCHI A H,HADASCHIK B A,et al.Role of MRI for the detection of prostate cancer[J].World J Urol,2021,39(3):637-649

    • [12] PLOUSSARD G,ROUVIÈRE O,ROUPRÊT M,et al.The current role of MRI for guiding active surveillance in prostate cancer[J].Nat Rev Urol,2022,19(6):357-365

    • [13] QI Y F,ZHANG S T,WEI J W,et al.Multiparametric MRI-based radiomics for prostate cancer screening with PSA in 4 ⁃ 10 ng/mL to reduce unnecessary biopsies[J].J Magn Reson Imaging,2020,51(6):1890-1899

    • [14] 李梦娟,张彩元,赵文露,等.基于双参数MRI的影像组学模型在临床显著性前列腺癌中的诊断价值探讨[J].磁共振成像,2022,13(11):76-81

    • [15] CHEN T,ZHANG Z Y,TAN S X,et al.MRI based ra-diomics compared with the PI⁃RADS V2.1 in the prediction of clinically significant prostate cancer:biparametric vs multiparametric MRI[J].Front Oncol,2021,11:792456

    • [16] BASS E J,PANTOVIC A,CONNOR M,et al.A systematic review and meta-analysis of the diagnostic accuracy of biparametric prostate MRI for prostate cancer in men at risk[J].Prostate Cancer Prostatic Dis,2021,24(3):596-611

    • [17] WOO S,SUH C H,KIM S Y,et al.Head⁃to⁃head comparison between biparametric and multiparametric MRI for the diagnosis of prostate cancer:a systematic review and meta⁃analysis[J].AJR Am J Roentgenol,2018,211(5):W226-W241

    • [18] ALABOUSI M,SALAMEH J P,GUSENBAUER K,et al.Biparametric vs multiparametric prostate magnetic resonance imaging for the detection of prostate cancer in treatment⁃naïve patients:a diagnostic test accuracy systematic review and meta-analysis[J].BJU Int,2019,124(2):209-220

    • [19] BLEKER J,KWEE T C,DIERCKX R A J O,et al.Multi-parametric MRI and auto ⁃fixed volume of interest-based radiomics signature for clinically significant peripheral zone prostate cancer[J].Eur Radiol,2020,30(3):1313-1324

    • [20] CASTILLO T J M,ARIF M,STARMANS M P A,et al.Classification of clinically significant prostate cancer on multi⁃parametric MRI:a validation study comparing deep learning and radiomics[J].Cancers,2021,14(1):12

    • [21] MOSES K A,SPRENKLE P C,BAHLER C,et al.NCCN guidelines® insights:prostate cancer early detection,version 1.2023[J].J Natl Compr Canc Netw,2023,21(3):236-246

    • [22] EPSTEIN J I,ZELEFSKY M J,SJOBERG D D,et al.A contemporary prostate cancer grading system:a validated alternative to the gleason score[J].Eur Urol,2016,69(3):428-435

    • [23] MOTTET N,VAN DEN BERGH R C N,BRIERS E,et al.EAU⁃EANM⁃ESTRO⁃ESUR⁃SIOG guidelines on prostate cancer ⁃ 2020 update.part 1:screening,diagnosis,and local treatment with curative intent[J].Eur Urol,2021,79(2):243-262

    • [24] KENDRICK J,FRANCIS R,HASSAN G M,et al.Ra-diomics for identification and prediction in metastatic prostate cancer:a review of studies[J].Front Oncol,2021,11:771787

    • [25] SHANKAR P R,KAZA R K,AL ⁃HAWARY M M,et al.Impact of clinical history on maximum PI ⁃RADS version 2 score:a six⁃reader 120⁃case sham history retrospective evaluation[J].Radiology,2018,288(1):158-163

    • [26] GONG L X,XU M,FANG M J,et al.Noninvasive prediction of high-grade prostate cancer via biparametric MRI radiomics[J].J Magn Reson Imaging,2020,52(4):1102-1109

    • [27] WANG X C,ZHANG Y,JI Z G,et al.Old men with prostate cancer have higher risk of gleason score upgrading and pathological upstaging after initial diagnosis:a systematic review and meta-analysis[J].World J Surg On-col,2021,19(1):18

    • [28] ZHANG G M,HAN Y Q,WEI J W,et al.Radiomics based on MRI as a biomarker to guide therapy by predicting upgrading of prostate cancer from biopsy to radical prostatectomy[J].J Magn Reson Imaging,2020,52(4):1239-1248

    • [29] OGBONNAYA C N,ZHANG X Y,ALSAEDI B S O,et al.Prediction of clinically significant cancer using radiomics features of pre⁃biopsy of multiparametric MRI in men suspected of prostate cancer[J].Cancers,2021,13(24):6199

    • [30] MAKOWSKI M R,BRESSEM K K,FRANZ L,et al.De novo radiomics approach using image augmentation and features from T1 mapping to predict gleason scores in prostate cancer[J].Invest Radiol,2021,56(10):661-668

    • [31] CASTILLO T J M,STARMANS M P A,ARIF M,et al.A multi⁃center,multi⁃vendor study to evaluate the generalizability of a radiomics model for classifying prostate cancer:high grade vs.low grade[J].Diagnostics,2021,11(2):369

    • [32] ALBERTSEN P C.Radical prostatectomy or watchful waiting in prostate cancer ⁃ 29⁃ year follow⁃up[J].Eur Urol,2020,78(3):471

    • [33] JEONG B C,CHALFIN H J,LEE S B,et al.The relation⁃ship between the extent of extraprostatic extension and survival following radical prostatectomy[J].Eur Urol,2015,67(2):342-346

    • [34] LITWIN M S,TAN H J.The diagnosis and treatment of prostate cancer:a review[J].JAMA,2017,317(24):2532-2542

    • [35] ROOIJ M D,HAMOEN E H,WITJES J A,et al.Accuracy of magnetic resonance imaging for local staging of prostate cancer:a diagnostic meta-analysis[J].Eur Urol,2016,70(2):233-245

    • [36] STANZIONE A,CUOCOLO R,COCOZZA S,et al.Detection of extraprostatic extension of cancer on biparametric MRI combining texture analysis and machine learning:preliminary results[J].Acad Radiol,2019,26(10):1338-1344

    • [37] MA S,XIE H H,WANG H H,et al.Preoperative prediction of extracapsular extension:radiomics signature based on magnetic resonance imaging to stage prostate cancer[J].Mol Imaging Biol,2020,22(3):711-721

    • [38] LOSNEGÅRD A,REISAETER L A R,HALVORSEN O J,et al.Magnetic resonance radiomics for prediction of ex-traprostatic extension in non ⁃favorable intermediate-and high⁃risk prostate cancer patients[J].Acta Radiol,2020,61(11):1570-1579

    • [39] MA S,XIE H H,WANG H H,et al.MRI⁃based radiomics signature for the preoperative prediction of extracapsular extension of prostate cancer[J].J Magn Reson Imaging,2019,50(6):1914-1925

    • [40] XU L L,ZHANG G,ZHAO L,et al.Radiomics based on multiparametric magnetic resonance imaging to predict extraprostatic extension of prostate cancer[J].Front On-col,2020,10:940

    • [41] HE D,WANG X M,FU C C,et al.MRI⁃based radiomics models to assess prostate cancer,extracapsular extension and positive surgical margins[J].Cancer Imaging,2021,21(1):46

    • [42] BAI H L,XIA W,JI X F,et al.Multiparametric magnetic resonance imaging⁃based peritumoral radiomics for preoperative prediction of the presence of extracapsular extension with prostate cancer[J].J Magn Reson Imaging,2021,54(4):1222-1230

    • [43] ALGOHARY A,SHIRADKAR R,PAHWA S,et al.Combination of peri ⁃tumoral and intra ⁃tumoral radiomic features on Bi-parametric MRI accurately stratifies prostate cancer risk:a multi⁃site study[J].Cancers,2020,12(8):2200

    • [44] ZHAI T,HU L,MA W,et al.Peri⁃prostatic adipose tissue measurements using MRI predict prostate cancer aggressiveness in men undergoing radical prostatectomy[J].J Endocrinol Invest,2021,44(2):287-296

    • [45] SALJI M,HENDRY J,PATEL A,et al.Peri⁃prostatic fat volume measurement as a predictive tool for castration resistance in advanced prostate cancer[J].Eur Urol Focus,2018,4(6):858-866

    • [46] CUOCOLO R,STANZIONE A,FALETTI R,et al.MRI index lesion radiomics and machine learning for detection of extraprostatic extension of disease:a multicenter study[J].Eur Radiol,2021,31(10):7575-7583

    • [47] FAN X H,XIE N,CHEN J W,et al.Multiparametric MRI and machine learning based radiomic models for preoperative prediction of multiple biological characteristics in prostate cancer[J].Front Oncol,2022,12:839621

    • [48] ZHONG Q Z,LONG L H,LIU A,et al.Radiomics of mul-tiparametric MRI to predict biochemical recurrence of localized prostate cancer after radiation therapy[J].Front Oncol,2020,10:731

    • [49] LI L,SHIRADKAR R,LEO P,et al.A novel imaging based nomogram for predicting post ⁃surgical biochemical recurrence and adverse pathology of prostate cancer from preoperative bi-parametric MRI[J].EBio Med,2021,63:103163

    • [50] SHIRADKAR R,PODDER T K,ALGOHARY A,et al.Radiomics based targeted radiotherapy planning(Rad-TRaP):a computational framework for prostate cancer treatment planning with MRI[J].Radiat Oncol,2016,11(1):148

    • [51] DANIEL M,KUESS P,ANDRZEJEWSKI P,et al.Impact of androgen deprivation therapy on apparent diffusion coefficient and T2W MRI for histogram and texture analysis with respect to focal radiotherapy of prostate cancer[J].Al,2019,195(5):402-411

    • [52] ABDOLLAHI H,TANHA K,MOFID B,et al.MRI ra-diomic analysis of IMRT⁃induced bladder wall changes in prostate cancer patients:a relationship with radiation dose and toxicity[J].J Med Imaging Radiat Sci,2019,50(2):252-260

    • [53] ABDOLLAHI H,MOFID B,SHIRI I,et al.Machine learning⁃based radiomic models to predict intensity⁃modulated radiation therapy response,Gleason score and stage in prostate cancer[J].Radiol Med,2019,124(6):555-567

    • [54] SULLIVAN L,WILLIAMS S G,TAI K H,et al.Urethral stricture following high dose rate brachytherapy for prostate cancer[J].Radiother Oncol,2009,91(2):232-236

    • [55] TSANG Y M,VIGNARAJAH D,MCWILLIAM A,et al.A pilot study on dosimetric and radiomics analysis of urethral strictures following HDR brachytherapy as monotherapy for localized prostate cancer[J].Br J Radiol,2020,93(1106):20190760

    • [56] ABDOLLAHI H,MAHDAVI S R,SHIRI I,et al.Magnetic resonance imaging radiomic feature analysis of radiation ⁃induced femoral head changes in prostate cancer radio⁃therapy[J].J Cancer Res Ther,2019,15(Supplement):S11-S19

    • [57] THARMALINGAM H,TSANG Y M,ALONZI R,et al.Changes in magnetic resonance imaging radiomic features in response to androgen deprivation therapy in patients with intermediate-and high⁃risk prostate cancer[J].Clin Oncol,2022,34(6):246-253

    • [58] ZAMBOGLOU C,BETTERMANN A S,GRATZKE C,et al.Uncovering the invisible-prevalence,characteristics,and radiomics feature-based detection of visually undetectable intraprostatic tumor lesions in 68GaPSMA⁃11 PET images of patients with primary prostate cancer[J].Eur J Nucl Med Mol Imaging,2021,48(6):1987-1997

    • [59] FELICIANI G,CELLI M,FERRONI F,et al.Radiomics analysis on[68Ga]Ga-PSMA ⁃11 PET and MRI ⁃ADC for the prediction of prostate cancer ISUP grades:preliminary results of the BIOPSTAGE trial[J].Cancers,2022,14(8):1888

    • [60] ERLE A,MOAZEMI S,LÜTJE S,et al.Evaluating a machine learning tool for the classification of pathological uptake in whole⁃body PSMA⁃PET⁃CT scans[J].Tomography,2021,7(3):301-312

    • [61] CYSOUW M C F,JANSEN B H E,VAN DE BRUG T,et al.Machine learning⁃based analysis of[18F]DCFPyL PET radiomics for risk stratification in primary prostate cancer[J].Eur J Nucl Med Mol Imaging,2021,48(2):340-349

    • [62] OSMAN S S,LEIJENAAR R T,COLE A J,et al.Computed tomography-based radiomics for risk stratification in prostate cancer[J].Int J Radiat Oncol Biol Phys,2019,105(2):448-456

    • [63] ACAR E,LEBLEBICI A,ELLIDOKUZ B E,et al.Machine learning for differentiating metastatic and completely responded sclerotic bone lesion in prostate cancer:a retrospective radiomics study[J].Br J Radiol,2019,92(1101):20190286

    • [64] PEEKEN J C,SHOUMAN M A,KROENKE M,et al.A CT-based radiomics model to detect prostate cancer lymph node metastases in PSMA radioguided surgery patients[J].Eur J Nucl Med Mol Imaging,2020,47(13):2968-2977

    • [65] MOSTAFAEI S,ABDOLLAHI H,KAZEMPOUR DEH-KORDI S,et al.CT imaging markers to improve radiation toxicity prediction in prostate cancer radiotherapy by stacking regression algorithm[J].Radiol Med,2020,125(1):87-97

    • [66] OU W,LEI J H,LI M H,et al.Ultrasound-based ra-diomics score for pre⁃biopsy prediction of prostate cancer to reduce unnecessary biopsies[J].Prostate,2023,83(1):109-118

    • [67] LIANG L,ZHI X,SUN Y,et al.A nomogram based on a multiparametric ultrasound radiomics model for discrimination between malignant and benign prostate lesions[J].Front Oncol,2021,11:610785

    • [68] WILDEBOER R R,MANNAERTS C K,VAN SLOUN R J G,et al.Automated multiparametric localization of prostate cancer based on B ⁃mode,shear ⁃wave elastography,and contrast ⁃enhanced ultrasound radiomics[J].Eur Ra-diol,2020,30(2):806⁃815

    • [69] ZHANG Q,XIONG J Y,CAI Y H,et al.Multimodal feature learning and fusion on B ⁃mode ultrasonography and sonoelastography using point-wise gated deep networks for prostate cancer diagnosis[J].Biomed Tech,2020,65(1):87-98

  • 参考文献

    • [1] JEMAL A,SIEGEL R,WARD E,et al.Cancer statistics,2009[J].CA A Cancer J Clin,2009,59(4):225-249

    • [2] WANG G C,ZHAO D,SPRING D J,et al.Genetics and biology of prostate cancer[J].Genes Dev,2018,32(17/18):1105-1140

    • [3] MILOT L.Variation of PI ⁃RADS interpretations between experts:a significant limitation[J].Radiology,2020,296(1):85-86

    • [4] PICKERSGILL N A,VETTER J M,ANDRIOLE G L,et al.Accuracy and variability of prostate multiparametric magnetic resonance imaging interpretation using the prostate imaging reporting and data system:a blinded comparison of radiologists[J].Eur Urol Focus,2020,6(2):267-272

    • [5] BI W L,HOSNY A,SCHABATH M B,et al.Artificial intelligence in cancer imaging:clinical challenges and applications[J].CA Cancer J Clin,2019,69(2):127-157

    • [6] ACHARYA U R,HAGIWARA Y,SUDARSHAN V K,et al.Towards precision medicine:from quantitative imaging to radiomics[J].J Zhejiang Univ Sci B,2018,19(1):6-24

    • [7] LAMBIN P,RIOS⁃VELAZQUEZ E,LEIJENAAR R,et al.Radiomics:extracting more information from medical images using advanced feature analysis[J].Eur J Cancer,2012,48(4):441-446

    • [8] SPOHN S K B,BETTERMANN A S,BAMBERG F,et al.Radiomics in prostate cancer imaging for a personalized treatment approach-current aspects of methodology and a systematic review on validated studies[J].Theranostics,2021,11(16):8027-8042

    • [9] LAMBIN P,LEIJENAAR R T H,DEIST T M,et al.Ra-diomics:the bridge between medical imaging and personalized medicine[J].Nat Rev Clin Oncol,2017,14(12):749-762

    • [10] FUJIHARA A,IWATA T,SHAKIR A,et al.Multipara-metric magnetic resonance imaging facilitates reclassification during active surveillance for prostate cancer[J].BJU Int,2021,127(6):712-721

    • [11] WU R C,LEBASTCHI A H,HADASCHIK B A,et al.Role of MRI for the detection of prostate cancer[J].World J Urol,2021,39(3):637-649

    • [12] PLOUSSARD G,ROUVIÈRE O,ROUPRÊT M,et al.The current role of MRI for guiding active surveillance in prostate cancer[J].Nat Rev Urol,2022,19(6):357-365

    • [13] QI Y F,ZHANG S T,WEI J W,et al.Multiparametric MRI-based radiomics for prostate cancer screening with PSA in 4 ⁃ 10 ng/mL to reduce unnecessary biopsies[J].J Magn Reson Imaging,2020,51(6):1890-1899

    • [14] 李梦娟,张彩元,赵文露,等.基于双参数MRI的影像组学模型在临床显著性前列腺癌中的诊断价值探讨[J].磁共振成像,2022,13(11):76-81

    • [15] CHEN T,ZHANG Z Y,TAN S X,et al.MRI based ra-diomics compared with the PI⁃RADS V2.1 in the prediction of clinically significant prostate cancer:biparametric vs multiparametric MRI[J].Front Oncol,2021,11:792456

    • [16] BASS E J,PANTOVIC A,CONNOR M,et al.A systematic review and meta-analysis of the diagnostic accuracy of biparametric prostate MRI for prostate cancer in men at risk[J].Prostate Cancer Prostatic Dis,2021,24(3):596-611

    • [17] WOO S,SUH C H,KIM S Y,et al.Head⁃to⁃head comparison between biparametric and multiparametric MRI for the diagnosis of prostate cancer:a systematic review and meta⁃analysis[J].AJR Am J Roentgenol,2018,211(5):W226-W241

    • [18] ALABOUSI M,SALAMEH J P,GUSENBAUER K,et al.Biparametric vs multiparametric prostate magnetic resonance imaging for the detection of prostate cancer in treatment⁃naïve patients:a diagnostic test accuracy systematic review and meta-analysis[J].BJU Int,2019,124(2):209-220

    • [19] BLEKER J,KWEE T C,DIERCKX R A J O,et al.Multi-parametric MRI and auto ⁃fixed volume of interest-based radiomics signature for clinically significant peripheral zone prostate cancer[J].Eur Radiol,2020,30(3):1313-1324

    • [20] CASTILLO T J M,ARIF M,STARMANS M P A,et al.Classification of clinically significant prostate cancer on multi⁃parametric MRI:a validation study comparing deep learning and radiomics[J].Cancers,2021,14(1):12

    • [21] MOSES K A,SPRENKLE P C,BAHLER C,et al.NCCN guidelines® insights:prostate cancer early detection,version 1.2023[J].J Natl Compr Canc Netw,2023,21(3):236-246

    • [22] EPSTEIN J I,ZELEFSKY M J,SJOBERG D D,et al.A contemporary prostate cancer grading system:a validated alternative to the gleason score[J].Eur Urol,2016,69(3):428-435

    • [23] MOTTET N,VAN DEN BERGH R C N,BRIERS E,et al.EAU⁃EANM⁃ESTRO⁃ESUR⁃SIOG guidelines on prostate cancer ⁃ 2020 update.part 1:screening,diagnosis,and local treatment with curative intent[J].Eur Urol,2021,79(2):243-262

    • [24] KENDRICK J,FRANCIS R,HASSAN G M,et al.Ra-diomics for identification and prediction in metastatic prostate cancer:a review of studies[J].Front Oncol,2021,11:771787

    • [25] SHANKAR P R,KAZA R K,AL ⁃HAWARY M M,et al.Impact of clinical history on maximum PI ⁃RADS version 2 score:a six⁃reader 120⁃case sham history retrospective evaluation[J].Radiology,2018,288(1):158-163

    • [26] GONG L X,XU M,FANG M J,et al.Noninvasive prediction of high-grade prostate cancer via biparametric MRI radiomics[J].J Magn Reson Imaging,2020,52(4):1102-1109

    • [27] WANG X C,ZHANG Y,JI Z G,et al.Old men with prostate cancer have higher risk of gleason score upgrading and pathological upstaging after initial diagnosis:a systematic review and meta-analysis[J].World J Surg On-col,2021,19(1):18

    • [28] ZHANG G M,HAN Y Q,WEI J W,et al.Radiomics based on MRI as a biomarker to guide therapy by predicting upgrading of prostate cancer from biopsy to radical prostatectomy[J].J Magn Reson Imaging,2020,52(4):1239-1248

    • [29] OGBONNAYA C N,ZHANG X Y,ALSAEDI B S O,et al.Prediction of clinically significant cancer using radiomics features of pre⁃biopsy of multiparametric MRI in men suspected of prostate cancer[J].Cancers,2021,13(24):6199

    • [30] MAKOWSKI M R,BRESSEM K K,FRANZ L,et al.De novo radiomics approach using image augmentation and features from T1 mapping to predict gleason scores in prostate cancer[J].Invest Radiol,2021,56(10):661-668

    • [31] CASTILLO T J M,STARMANS M P A,ARIF M,et al.A multi⁃center,multi⁃vendor study to evaluate the generalizability of a radiomics model for classifying prostate cancer:high grade vs.low grade[J].Diagnostics,2021,11(2):369

    • [32] ALBERTSEN P C.Radical prostatectomy or watchful waiting in prostate cancer ⁃ 29⁃ year follow⁃up[J].Eur Urol,2020,78(3):471

    • [33] JEONG B C,CHALFIN H J,LEE S B,et al.The relation⁃ship between the extent of extraprostatic extension and survival following radical prostatectomy[J].Eur Urol,2015,67(2):342-346

    • [34] LITWIN M S,TAN H J.The diagnosis and treatment of prostate cancer:a review[J].JAMA,2017,317(24):2532-2542

    • [35] ROOIJ M D,HAMOEN E H,WITJES J A,et al.Accuracy of magnetic resonance imaging for local staging of prostate cancer:a diagnostic meta-analysis[J].Eur Urol,2016,70(2):233-245

    • [36] STANZIONE A,CUOCOLO R,COCOZZA S,et al.Detection of extraprostatic extension of cancer on biparametric MRI combining texture analysis and machine learning:preliminary results[J].Acad Radiol,2019,26(10):1338-1344

    • [37] MA S,XIE H H,WANG H H,et al.Preoperative prediction of extracapsular extension:radiomics signature based on magnetic resonance imaging to stage prostate cancer[J].Mol Imaging Biol,2020,22(3):711-721

    • [38] LOSNEGÅRD A,REISAETER L A R,HALVORSEN O J,et al.Magnetic resonance radiomics for prediction of ex-traprostatic extension in non ⁃favorable intermediate-and high⁃risk prostate cancer patients[J].Acta Radiol,2020,61(11):1570-1579

    • [39] MA S,XIE H H,WANG H H,et al.MRI⁃based radiomics signature for the preoperative prediction of extracapsular extension of prostate cancer[J].J Magn Reson Imaging,2019,50(6):1914-1925

    • [40] XU L L,ZHANG G,ZHAO L,et al.Radiomics based on multiparametric magnetic resonance imaging to predict extraprostatic extension of prostate cancer[J].Front On-col,2020,10:940

    • [41] HE D,WANG X M,FU C C,et al.MRI⁃based radiomics models to assess prostate cancer,extracapsular extension and positive surgical margins[J].Cancer Imaging,2021,21(1):46

    • [42] BAI H L,XIA W,JI X F,et al.Multiparametric magnetic resonance imaging⁃based peritumoral radiomics for preoperative prediction of the presence of extracapsular extension with prostate cancer[J].J Magn Reson Imaging,2021,54(4):1222-1230

    • [43] ALGOHARY A,SHIRADKAR R,PAHWA S,et al.Combination of peri ⁃tumoral and intra ⁃tumoral radiomic features on Bi-parametric MRI accurately stratifies prostate cancer risk:a multi⁃site study[J].Cancers,2020,12(8):2200

    • [44] ZHAI T,HU L,MA W,et al.Peri⁃prostatic adipose tissue measurements using MRI predict prostate cancer aggressiveness in men undergoing radical prostatectomy[J].J Endocrinol Invest,2021,44(2):287-296

    • [45] SALJI M,HENDRY J,PATEL A,et al.Peri⁃prostatic fat volume measurement as a predictive tool for castration resistance in advanced prostate cancer[J].Eur Urol Focus,2018,4(6):858-866

    • [46] CUOCOLO R,STANZIONE A,FALETTI R,et al.MRI index lesion radiomics and machine learning for detection of extraprostatic extension of disease:a multicenter study[J].Eur Radiol,2021,31(10):7575-7583

    • [47] FAN X H,XIE N,CHEN J W,et al.Multiparametric MRI and machine learning based radiomic models for preoperative prediction of multiple biological characteristics in prostate cancer[J].Front Oncol,2022,12:839621

    • [48] ZHONG Q Z,LONG L H,LIU A,et al.Radiomics of mul-tiparametric MRI to predict biochemical recurrence of localized prostate cancer after radiation therapy[J].Front Oncol,2020,10:731

    • [49] LI L,SHIRADKAR R,LEO P,et al.A novel imaging based nomogram for predicting post ⁃surgical biochemical recurrence and adverse pathology of prostate cancer from preoperative bi-parametric MRI[J].EBio Med,2021,63:103163

    • [50] SHIRADKAR R,PODDER T K,ALGOHARY A,et al.Radiomics based targeted radiotherapy planning(Rad-TRaP):a computational framework for prostate cancer treatment planning with MRI[J].Radiat Oncol,2016,11(1):148

    • [51] DANIEL M,KUESS P,ANDRZEJEWSKI P,et al.Impact of androgen deprivation therapy on apparent diffusion coefficient and T2W MRI for histogram and texture analysis with respect to focal radiotherapy of prostate cancer[J].Al,2019,195(5):402-411

    • [52] ABDOLLAHI H,TANHA K,MOFID B,et al.MRI ra-diomic analysis of IMRT⁃induced bladder wall changes in prostate cancer patients:a relationship with radiation dose and toxicity[J].J Med Imaging Radiat Sci,2019,50(2):252-260

    • [53] ABDOLLAHI H,MOFID B,SHIRI I,et al.Machine learning⁃based radiomic models to predict intensity⁃modulated radiation therapy response,Gleason score and stage in prostate cancer[J].Radiol Med,2019,124(6):555-567

    • [54] SULLIVAN L,WILLIAMS S G,TAI K H,et al.Urethral stricture following high dose rate brachytherapy for prostate cancer[J].Radiother Oncol,2009,91(2):232-236

    • [55] TSANG Y M,VIGNARAJAH D,MCWILLIAM A,et al.A pilot study on dosimetric and radiomics analysis of urethral strictures following HDR brachytherapy as monotherapy for localized prostate cancer[J].Br J Radiol,2020,93(1106):20190760

    • [56] ABDOLLAHI H,MAHDAVI S R,SHIRI I,et al.Magnetic resonance imaging radiomic feature analysis of radiation ⁃induced femoral head changes in prostate cancer radio⁃therapy[J].J Cancer Res Ther,2019,15(Supplement):S11-S19

    • [57] THARMALINGAM H,TSANG Y M,ALONZI R,et al.Changes in magnetic resonance imaging radiomic features in response to androgen deprivation therapy in patients with intermediate-and high⁃risk prostate cancer[J].Clin Oncol,2022,34(6):246-253

    • [58] ZAMBOGLOU C,BETTERMANN A S,GRATZKE C,et al.Uncovering the invisible-prevalence,characteristics,and radiomics feature-based detection of visually undetectable intraprostatic tumor lesions in 68GaPSMA⁃11 PET images of patients with primary prostate cancer[J].Eur J Nucl Med Mol Imaging,2021,48(6):1987-1997

    • [59] FELICIANI G,CELLI M,FERRONI F,et al.Radiomics analysis on[68Ga]Ga-PSMA ⁃11 PET and MRI ⁃ADC for the prediction of prostate cancer ISUP grades:preliminary results of the BIOPSTAGE trial[J].Cancers,2022,14(8):1888

    • [60] ERLE A,MOAZEMI S,LÜTJE S,et al.Evaluating a machine learning tool for the classification of pathological uptake in whole⁃body PSMA⁃PET⁃CT scans[J].Tomography,2021,7(3):301-312

    • [61] CYSOUW M C F,JANSEN B H E,VAN DE BRUG T,et al.Machine learning⁃based analysis of[18F]DCFPyL PET radiomics for risk stratification in primary prostate cancer[J].Eur J Nucl Med Mol Imaging,2021,48(2):340-349

    • [62] OSMAN S S,LEIJENAAR R T,COLE A J,et al.Computed tomography-based radiomics for risk stratification in prostate cancer[J].Int J Radiat Oncol Biol Phys,2019,105(2):448-456

    • [63] ACAR E,LEBLEBICI A,ELLIDOKUZ B E,et al.Machine learning for differentiating metastatic and completely responded sclerotic bone lesion in prostate cancer:a retrospective radiomics study[J].Br J Radiol,2019,92(1101):20190286

    • [64] PEEKEN J C,SHOUMAN M A,KROENKE M,et al.A CT-based radiomics model to detect prostate cancer lymph node metastases in PSMA radioguided surgery patients[J].Eur J Nucl Med Mol Imaging,2020,47(13):2968-2977

    • [65] MOSTAFAEI S,ABDOLLAHI H,KAZEMPOUR DEH-KORDI S,et al.CT imaging markers to improve radiation toxicity prediction in prostate cancer radiotherapy by stacking regression algorithm[J].Radiol Med,2020,125(1):87-97

    • [66] OU W,LEI J H,LI M H,et al.Ultrasound-based ra-diomics score for pre⁃biopsy prediction of prostate cancer to reduce unnecessary biopsies[J].Prostate,2023,83(1):109-118

    • [67] LIANG L,ZHI X,SUN Y,et al.A nomogram based on a multiparametric ultrasound radiomics model for discrimination between malignant and benign prostate lesions[J].Front Oncol,2021,11:610785

    • [68] WILDEBOER R R,MANNAERTS C K,VAN SLOUN R J G,et al.Automated multiparametric localization of prostate cancer based on B ⁃mode,shear ⁃wave elastography,and contrast ⁃enhanced ultrasound radiomics[J].Eur Ra-diol,2020,30(2):806⁃815

    • [69] ZHANG Q,XIONG J Y,CAI Y H,et al.Multimodal feature learning and fusion on B ⁃mode ultrasonography and sonoelastography using point-wise gated deep networks for prostate cancer diagnosis[J].Biomed Tech,2020,65(1):87-98