Advances in CT Functional Imaging and Artificial Intelligence for Assessing Esophageal Varices in Cirrhosis
-
摘要:
肝硬化相关门脉高压引起的食管静脉曲张(EV),并由此导致的上消化道出血是最常见的并发症之一。内窥镜检查是诊断静脉曲张的“金标准”,但其属于侵入性检查,患者依从性较差,并且不便在短期内随访复查,在一定程度上限制了低出血风险患者中的应用。因此,临床上需寻找一种无创影像学方法,以准确诊断食管静脉曲张、评估其曲张程度,并预测潜在的出血风险。近年来,CT功能成像及人工智能新技术在食管静脉曲张的应用研究逐渐成为热点,二者相结合可能为有效诊断门静脉高压和食管静脉曲张提供新的策略。本文旨在对CT在食管静脉曲张诊断中的研究现状和进展进行综述,以期为临床诊断和治疗提供帮助。
Abstract:Esophageal varices (EV), which result from portal hypertension associated with cirrhosis, along with the consequent upper gastrointestinal bleeding, represent one of the most prevalent complications. Endoscopy is regarded as the "gold standard" for diagnosing varices; however, its invasive nature, coupled with poor patient compliance and the inconvenience of short-term follow-up, restricts its use in patients at low risk of bleeding. Consequently, there is a pressing need to identify a noninvasive imaging method that can accurately diagnose EV, evaluate their severity, and predict the risk of potential bleeding. In recent years, research focusing on the application of computed tomography (CT) functional imaging and novel artificial intelligence techniques in the context of EV has gained significant attention. The integration of these approaches may offer a new strategy for the effective diagnosis of portal hypertension and esophageal varices. This article aims to review the current research status and advancements in the use of CT for diagnosing EV, with the goal of assisting clinical diagnosis and treatment.
-
肝硬化是肝脏慢性炎症的结果,主要是由于慢性炎症性肝损伤导致肝肌成纤维细胞和巨噬细胞活化,从而造成细胞外基质异常积聚。这个过程逐渐进展为弥漫性肝脏纤维化,改变肝脏结构,形成再生结节并阻碍门静脉血流,导致门静脉高压(portal hypertension,PHT)[1]。每年约有200万人死于肝脏疾病,其中肝硬化相关门脉高压引起食管静脉曲张(esophageal varices,EV)及其破裂出血尤为常见[2],且风险随病情加重而上升[3]。另外值得注意的是,食管静脉曲张破裂出血(esophageal variceal bleeding,EVB)发病后6周的死亡率约为20%[4]。因此,准确诊断食管静脉曲张并评估其出血风险,对于肝硬化患者的治疗和预后极为重要。
内窥镜检查是诊断静脉曲张的“金标准”。根据Baveno VI指南,无静脉曲张的代偿性肝硬化患者被建议每3年接受一次内镜检查以监测病情发展[5]。另外,肝静脉压力梯度(hepatic vein pressure gradient,HVPG)不仅是诊断PHT的“金标准”,还能为肝硬化患者的风险分层提供重要依据[6]。通常情况下,HVPG达到或超过6 mmHg时被认为存在门静脉高压,当HVPG≥10 mmHg时患者可能存在食管静脉曲张[7]。HVPG在预测肝硬化患者并发症如EV和静脉曲张出血的发生方面优于肝活检[8]。但这两者均属于侵入性检查,患者依从性较差,并且不便在短期内随访复查,在一定程度上限制了低出血风险患者中的应用。因此,临床上需寻找一种无创影像学方法,以准确诊断食管静脉曲张、评估其曲张程度,并预测潜在的出血风险。
CT能够无创性地评估门静脉高压和食管静脉曲张,且与内镜检查结果有良好的一致性[9]。近年来,CT功能成像及人工智能新技术在食管静脉曲张的应用研究逐渐成为热点,二者相结合可能为有效诊断门静脉高压和食管静脉曲张提供新的策略。本文旨在对CT在食管静脉曲张诊断中的研究现状和进展进行综述,以期为临床诊断和治疗提供帮助。
1. 常规CT在EV中的研究进展
CT门静脉成像对肝硬化诊断有较高的敏感性和特异性。与内镜检查相比,多排螺旋CT可以通过快速扫描和多层图像重建获得高质量图像,具有更好的时间和空间分辨率。借助先进的CT图像后处理技术,如最大密度投影与多平面重建等,增强CT得以清晰地显示门静脉系统及其侧枝循环。这些技术不仅有助于我们观察门静脉的走行、分支以及与其他血管的关系,还可以准确判断门静脉系统内是否存在血栓、癌栓等,以及评估病变范围和程度[10]。
Barbara等[11]研究发现当曲张食管静脉内径≥4 mm时,可被识别为高风险静脉曲张,以便进行早期内镜下干预治疗。Xie等[12]通过研究发现食管静脉曲张的总面积作为出血预测的潜在新指标具有更高的准确性。有相关研究发现肝硬化食管胃底静脉曲张破裂出血的患者,其胃左静脉都有不同程度的扩张,说明胃左静脉预测EVB具有一定的价值。Caraiani等[13]与Faeze等[14]研究结果相似,证实特定侧支(胃左静脉、胃短静脉和食管旁静脉丛)的存在(和/或大小)与高危食管静脉曲张和食管静脉曲张出血均有显著相关性。Shang等[15]在研究中发现肝叶体积(liver lobe volume,LLV)可预测EV的严重程度和急性首次静脉曲张出血(first variceal haerrhage,FVH)的风险。此外,结合CV/TV和腹水构建的预测模型为识别严重EV提供了可靠的诊断性能。因此CT衍生的LLV值可以作为一种简单的测量方法,来协助临床进行诊疗方案的制定,并且可以作为上消化道内窥镜检查的补充工具,具有良好的临床实用性。以上研究不仅为预测出血提供了新的指标,更强调了CT扫描在精确评估食管静脉曲张中的关键作用,为临床针对高危患者制定预防措施提供科学依据。
由此可见,门静脉CT成像能直观展示其形态学的变化,在初步评估PHT及EV方面有一定临床价值。但鉴于其无法反映血流动力学变化,在病情诊断及预后评估方面,仍需要结合其他检查手段来弥补其局限性。
2. 能谱CT在EV中的研究进展
能谱CT物质分离算法可以基于不同元素在不同能级下的吸收特性差异,定量测量碘浓度,生成的碘彩图可以清晰显示肝脏病变碘浓聚,进一步反映血流动力学改变[16]。虚拟平扫(virtual non-contrast,VNC)成像[17]是能谱CT研究最多的应用之一。它是基于对能谱数据经后处理算法后获得的抑碘图,图像质量与常规CT平扫相当,VNC有望替代常规平扫(true non-contrast,TNC)从而减少辐射剂量。虚拟单能图像(virtual monoenergetic image,VMI或MonoE)相当于单一能量射线的成像,通过重建可获得高对比度、低噪声的图像,包括40 ~ 200 keV,从而可以更清晰的显示病灶[18]。
Wang等[19]研究认为肝实质标准化碘浓度是鉴别有临床意义的门静脉高压症(≥10 mmHg)、食管静脉曲张和高危食管静脉曲张的有效指标,同时也是无创评估肝硬化患者门静脉压的可靠参数。既往研究表明,基于增强CT计算的细胞外体积(extracellular volume,ECV)是评估肝纤维化的重要工具。ECV表示血管内和血管外细胞外间隙的总和[20],有助于评估肝病严重程度及预测进展。Seongjun等[21]研究发现通过双能CT测量的fECV结合白蛋白水平可预测肝硬化相关事件,提供客观的无创预后信息。然而需要注意的是,fECV与食管静脉曲张分级受多种因素影响,如药物治疗等[22]。
脾大是肝硬化门脉高压最早出现的体征之一[23],由脾静脉回流受阻致淤血性肿大,其变化能够间接反映食管胃底静脉曲张情况。Han等[24]通过研究发现高危EV伴随着脾脏碘容量(iodine weight of spleen,IW-S)的显著增加,以
1087 mg为临界值,IW-S检测高危EV的AUC为0.87,敏感度为84.9%,特异度为84.2%,说明IW-S可以作为高危食管胃底静脉曲张的筛查方法。并且IW-S结合了形态学和功能学信息,为EV的无创预测提供了新思路。总体而言,能谱CT以高分辨率成像、低辐射剂量以及多样化的定量参数,从形态学及功能学角度反映肝脏血流动力学改变,在肝硬化EV的评估中优势明显,为预防和治疗提供重要的理论依据和参考价值。但有研究表明不同CT扫描仪类型测得的能谱参数存在差异,其中肝脏的差异性最大,这可能会影响对病变微小特征和治疗效果的准确评估,需使用更稳健的标准化方法,例如将目标碘浓度标准化为多个参考点(NICALL),作为体内总体碘负荷的代用指标,从而进一步降低其差异性[25]。
3. CT灌注成像在EV中的研究进展
灌注(perfusion)是血液向组织输送氧气和营养物质的过程,对于反映器官和组织的血流动力学状态至关重要。CT灌注成像(CTP)可获得多个灌注参数,如血容量(blood volume,BV)、血流量(blood flow,BF)、平均通过时间(mean transit time,MTT)、达峰时间(time to peak, TTP)等,进而评估组织和器官的灌注状态[26]。在肝脏疾病的诊疗过程中,这些参数的变化能够较为准确的体现肝硬化门静脉高压时肝脏血流动力学所发生的变化[27],为临床决策提供重要影像学依据。
Yan等[28]通过研究表明用低剂量灌注CT测得的脾脏血流动力学参数BV和BF可以反映EV的严重程度,肝硬化EV患者的表面通透性(permeability surface,PS)、脾脏体积(volume of spleen,V-S)、脾脏总血容量(total blood volume of the spleen,BV-S)均高于非肝硬化和非EV患者,并且V-S的增加伴随着BV的减少,这与Motosugi[29]和Sauter[30]的以往研究结果一致。另外,周等[31]发现肾脏峰值模型中的门静脉灌注量(portal venous perfusion,PVP)可预测EGVB,临界值为59.69 mL/(min·100 mL)时,预测肝硬化食管胃底静脉曲张破裂出血(esophagogastric variceal bleeding,EGVB)的敏感度及特异度较高。既往也有研究表明CTP可直观定量监测术前术后的肝、脾实质血流量变化情况,DJ等[32]研究显示手术前后的灌注参数变化虽不显著,但可作为TIPS术后并发症的潜在预测因子,有助于在手术前进行全面评估甚至采取预防措施。Wang等[33]前瞻性收集28例EVB患者,根据HVPG(>12 mmHg和≤12 mmHg)分为中度组与重度组,研究结果显示肝血容量(liver blood volume,LBV)和肝血流量(liver blood flow,LBF)与HVPG和Child-Pugh评分呈负相关,且中度组的Child-Pugh评分、HVPG、LBF和LBV均显著高于重度组。
CTP作为功能成像技术,具有可重复性、连续追踪观察等优点,通过评估肝脾的血流动力学参数,为EV患者提供全面信息,帮助临床医生制定个体化诊疗方案,提高生存率。但国内外有关EV的大样本前瞻性研究仍较少,准确性有待进一步验证。总之,在实际应用中,CTP可以结合多种影像学指标来提高诊断性能,以做出正确的诊断评估和临床决策。
4. CT人工智能在EV中的研究进展
随着计算机技术和医学影像技术的迅速发展,越来越多无创检测EV的方法涌现出来,人工智能(artificial intelligence,AI)在提高诊断准确性和效率方面的优势逐渐显现。关于肝硬化及并发症的AI模型的研究报道,无论是传统的影像组学还是深度学习都具有非常高的诊断性能[34][35]。
影像组学可以自动从图像中提取的大量定量特征,弥补了传统影像学肉眼评估图像的不足,更精确的对疾病进行诊断、鉴别、分级及评价疗效等,从而实现个性化精准治疗[36−37]。Yan等[38]构建的影像组学模型在诊断高出血风险食管静脉曲张(high bleeding risk esophageal varices,HREV)上表现卓越,AUC较高。并且与Baveno VI和扩展Baveno VI标准相比,组学模型在诊断HREV方面的精度分别提高了49.0%和32.8%。但该研究仅纳入患有乙型肝炎的肝硬化患者,对于其他病因的肝硬化患者中HREV的识别能力尚需进一步验证。Liu等[39]也同样证实影像组学无创模型在诊断高风险EV中的有效性。
2015年Baveno VI共识研讨会提出当肝脏硬度<20 kPa,血小板计数大于150×109/L时,患者发生出血且需要治疗的风险很低,可以避免内镜筛查[5]。Lin等[40]结合影像组学特征与血小板计数构建列线图,在训练集、内部验证集和外部验证集的AUC分别为0.987、0.973和0.947,显著减少内镜检查需求并降低EV漏诊率。深度学习是机器学习的子集,基于多层人工神经网络学习特征,与传统的机器学习的区别在于其使用神经网络算法自动提取特征,不需要感兴趣区分割[41]。Li等[42]应用影像组学和机器学习方法,通过线性组合融合肝、脾和食管三个器官的特征构建模型来预测EV风险水平,结果表明该模型较以往影像组学模型的分类性能更好。此外,与肝、脾相比,通过食管构建的模型性能最佳。Lee等[43]基于深度学习建立的肝脾体积与临床因素指数,有效预测乙肝肝硬化高危静脉曲张,发现当脾体积与血小板的比值<1.63时,可以排除高危静脉曲张,并避免内窥镜检查。
尽管影像组学和深度学习目前还面临着诸多问题,但是众多研究结果表明在肝脏疾病的诊断、预后预测和疗效评估中仍具有广阔的前景,特别是在研究食管胃底静脉曲张的无创评估中影像组学显示出巨大潜力,有望在临床工作中成为肝硬化及并发症的重要诊断工具。
5. 临床-影像综合模型在EV中的研究进展
CT现已作为一种有价值的非侵入手段被采用,同时,国内外也有许多血清学指标建立的无创模型用于EV的诊断,如肝纤维化4因子(fibrosis 4 index,FIB4)、天门冬氨酸转氨酶-血小板比值(aspartate aminotransferase-to-platelet ratio index,APRI)、Lok评分和King评分等[44−45]。结合临床信息和CT影像特征,能够更全面、准确地评估肝硬化患者的食管静脉曲张风险。
汤等[46]研究发现脾脏体积、门静脉内径、胃左静脉内径、FIB-4和血管内皮生长因子(vascular endothelial growth factor,VEGF)五项指标联合诊断EV的AUC最大,其预测效能最高。肝硬化时,肝细胞广泛变形坏死,肝功能减退,对脂肪的代谢和转运能力下降,导致肝脏脂肪沉积[47]。张成孟等[48]采用定量CT(quantitative computed Tomography,QCT)技术测定EV患者的肝脏脂肪含量,并纳入血液生化指标,结果表明肝脏脂肪含量增加和血红蛋白减少都会增加EGVB发生的风险,以此构建的综合模型AUC为0.798,表现出较高的预测效能和较好的校准度。Luo等[49]研究发现门静脉血栓形成、天冬氨酸氨基转移酶、白蛋白、纤维蛋白原和脾脏厚度是EGVB的独立预测因素,同时提取肝脏和脾脏CT影像组学特征,建立的临床-影像综合模型在训练集和验证集中都表现出良好的预测性能,并且与APRI和FIB4等无创评分模型相比有更好的预测精度。Liu等[50]的研究结果与之相似,综合模型在预测EVB方面优于单一的肝脾影像组学模型和临床信息模型。
临床-影像综合模型在肝硬化EV的预测中已显示出良好的应用前景。通过早期识别高风险患者,可以对其进行密切监测或预防性治疗,提高患者生活质量。此外,随着技术的不断进步和模型的持续优化,预测模型的准确性和实用性将进一步提高,为临床决策提供有力支持。
6. 总结与展望
运用非侵入方法评估肝硬化及其并发症是未来医学发展的重要方向,特别是CT的前沿技术以及与人工智能的结合,在肝硬化食管静脉曲张的识别、诊断、分级、风险预测及预后等方面展现巨大潜力。CT因其高效、高分辨率、定性诊断能力强等优势在临床广泛应用。同时,多模态成像技术的融合和人工智能辅助诊断的引入进一步改善和提高了诊断准确性和疗效评估能力,在一定程度上减少对内镜检查及肝组织病理活检的依赖。然而CT诊断仍存在诸多挑战。例如CT扫描参数易受扫描时间、对比剂用量和注射流速等多种因素的影响,因此可以使同一类型的扫描仪器、统一扫描参数并结合临床信息等。其次,目前研究样本有限且多集中于单一机构,未来需扩大样本量、多中心合作来提升模型的普适性和准确性。展望未来,随着我国医疗科技的发展,CT技术将在肝硬化及食管静脉曲张的临床应用中展现出更广阔的前景。
-
[1] 徐小元, 丁惠国, 李文刚等. 肝硬化诊治指南[J]. 实用肝脏病杂志, 2019, 22(6): 770−786. DOI: 10.3969/j.issn.1672-5069.2019.06.004. XU X Y, DING H G, LI W G, et al. Chinese guidelines on the management of liver cirrhosis[J]. Journal of Practical Hepatology, 2019, 22(6): 770−786. DOI: 10.3969/j.issn.1672-5069.2019.06.004.
[2] SHEENA B S, HIEBERT L, HAN H, et al. Global, regional, and national burden of hepatitis B, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019[J]. The lancet Gastroenterology & hepatology, 2022, 7(9): 796−829. DOI: 10.1016/S2468-1253(22)00124-8.
[3] LI Y, LI L, WENG H L, et al. Computed tomography vs liver stiffness measurement and magnetic resonance imaging in evaluating esophageal varices in cirrhotic patients: A systematic review and meta-analysis[J]. World Journal of Gastroenterology, 2020, 26(18): 2247−2267. DOI: 10.3748/wjg.v26.i18.2247.
[4] GINèS P, KRAG A, ABRALDES J G, et al. Liver cirrhosis[J]. Lancet, 2021, 398(10308): 1359−1376. DOI: 10.1016/S0140-6736(21)01374-X.
[5] de Franchis R, Baveno VI Faculty. Expanding consensus in portal hypertension: Report of the Baveno VI Consensus Workshop: Stratifying risk and individualizing care for portal hypertension[J]. Journal of Hepatology, 2015, 63(3): 743−752. DOI: 10.1016/j.jhep.2015.05.022.
[6] BOSCH J. Portal Hypertension and Cirrhosis: From Evolving Concepts to Better Therapies[J]. Clinical liver disease(Hoboken), 2020, 15(Suppl 1): S8−S12. DOI: 10.1002/cld.844.
[7] REIBERGER T, SCHWABL P, TRAUNER M, et al. Measurement of the hepatic venous pressure gradient and transjugular liver biopsy[J]. Journal of visualized experiments: JoVE, 2020, (160). DOI: 10.3791/58819.
[8] GARCIA-TSAO G, ABRALDES J G, BERZIGOTTI A, et al. Portal hypertensive bleeding in cirrhosis: Risk stratification, diagnosis, and management: 2016 practice guidance by the American Association for the study of liver diseases[J]. Hepatology, 2017, 65(1): 310−335. DOI: 10.1002/hep.28906.
[9] 马关华, 陈小宇, 梁宇豪, 等. 多排螺旋CT门静脉成像技术在食管胃底静脉曲张中的应用[J]. 中外医学研究, 2021, 19(3): 63−65. DOI: 10.14033/j.cnki.cfmr.2021.03.024. MA G H, CHEN X Y, LIANG Y H, et al. Application of Multidetector Computed Tomography Portal Venography in Esophageal and Gastric Varices[J]. Chinese and Foreign Medical Research, 2021, 19(3): 63−65. DOI: 10.14033/j.cnki.cfmr.2021.03.024.
[10] ZHOU H Y, CHEN T W, ZHANG X M, et al. The diameter of the originating vein determines esophageal and gastric fundic varices in portal hypertension secondary to posthepatitic cirrhosis[J]. Clinics (Sao Paulo, Brazil), 2012, 67(6): 609−614. DOI: 10.6061/clinics/2012(06)11.
[11] MANCHEC B, PHAM E, NOOR M, et al. Contrast-Enhanced CT May Identify High-Risk Esophageal Varices in Patients With Cirrhosis[J]. American journal of roentgenology, 2020, 215(3): 617−623. DOI: 10.2214/AJR.19.22474.
[12] XIE W, CHEN F X, ZHU L Y, et al. Risk assessment of first upper gastrointestinal bleeding using computerized tomoscanning in esophageal varices patients with cirrhosis and portal hypertension[J]. Medicine, 2020, 99(5): e18923. DOI: 10.1097/MD.0000000000018923.
[13] CARAIANI C , PETRESC B , POP A , et al. Can the computed tomographic aspect of porto-systemic circulation in cirrhotic patients be associated with the presence of variceal hemorrhage?[J]. Medicina, 2020, 56(6): 301. DOI: 10.3390/medicina56060301.
[14] SALAHSHOUR F, MEHRABINEJAD M M, RASHIDI SHAHPASANDI M H, et al. Esophageal variceal hemorrhage: The role of MDCT characteristics in predicting the presence of varices and bleeding risk[J]. Abdominal radiology (New York), 2020, 45(8): 2305−2314. DOI: 10.1007/s00261-020-02585-5.
[15] WaN S, WEI Y, ZHANG X, et al. CT-derived quantitative liver volumetric parameters for prediction of severe esophageal varices and the risk of first variceal hemorrhage[J]. European journal of radiology, 2021, 144: 109984. DOI: 10.1016/j.ejrad.2021.109984.
[16] MAHMOUDI S, BERNATZ S, ALTHOFF F C, et al. Dual-energy CT based material decomposition to differentiate intrahepatic cholangiocarcinoma from hepatocellular carcinoma[J]. European Journal of Radiology, 2022, 156: 110556. DOI: 10.1016/j.ejrad.2022.110556.
[17] LEE M H, PARK H J, KIM J N, et al. Virtual non-contrast images from dual-energy CT angiography of the abdominal aorta and femoral arteries: comparison with true non-contrast CT images[J]. The British Journal of Radiology, 2022, 95(1138): 20220378. DOI: 10.1259/bjr.20220378.
[18] SIEGEL M J, RAMIREZ-GIRALDO J C. Dual-Energy CT in Children: Imaging Algorithms and Clinical Applications[J]. Radiology, 2019, 291(2): 286−297. DOI: 10.1148/radiol.2019182289.
[19] WANG J, ZHANG L, CHENG S M, et al. The evaluation of portal hypertension in cirrhotic patients with spectral computed tomography[J]. Acta Radiologica, 2023, 64(3): 918−925. DOI: 10.1177/02841851221101356.
[20] GUO S L, SU L N, ZHAI Y N, et al. The clinical value of hepatic extracellular volume fraction using routine multiphasic contrast-enhanced liver CT for staging liver fibrosis[J]. Clinical Radiology, 2017, 72(3): 242−246. DOI: 10.1016/j.crad.2016.10.003.
[21] BAK S, KIM JE, BAE K, et al. Quantification of liver extracellular volume using dual-energy CT: utility for prediction of liver-related events in cirrhosis[J]. European Radiology, 2020, 30(10): 5317−5326. DOI: 10.1007/s00330-020-06876-9.
[22] CICERO G, MAZZIOTTI S, SILIPIGNI S, et al. Dual-energy CT quantification of fractional extracellular space in cirrhotic patients: comparison between early and delayed equilibrium phases and correlation with oesophageal varices[J]. La Radiologia Medica, 2021, 126(6): 761−767. DOI: 10.1007/s11547-021-01341-z.
[23] Berzigotti A. Advances and challenges in cirrhosis and portal hypertension[J]. BMC Medicine, 2017, 15(1): 200. DOI: 10.1186/s12916-017-0966-6.
[24] HAN X, AN W, CAO Q , et al. Noninvasive evaluation of esophageal varices in cirrhotic patients based on spleen hemodynamics: a dual-energy CT study[J]. European Radiology, 2020, 30(6): 3210-3216. DOI: 10.1007/s00330-020-06680-5.
[25] LENNARTZ S, CAO J, PISUCHPEN N, et al. Intra-patient variability of iodine quantification across different dual-energy CT platforms: assessment of normalization techniques[J]. European Radiology, 2024, 34(8): 5131−5141. DOI: 10.1007/s00330-023-10560-z.
[26] 赵光明, 韩丹. CT灌注成像原理与技术[J]. 中国医学影像技术, 2003, 19(5): 636−638. DOI: 10.13929/j.1003-3289.2003.05.062. [27] TALAKIĆ E, SCHAFFELLNER S, KNIEPEISS D, et al. CT perfusion imaging of the liver and the spleen in patients with cirrhosis: Is there a correlation between perfusion and portal venous hypertension?[J]. European Radiology, 2017, 27(10): 4173−4180. DOI: 10.1007/s00330-017-4788-x.
[28] YAN C, HAN X, LIANG X, et al. Non-invasive evaluation of esophageal varices in patients with liver cirrhosis using low-dose splenic perfusion CT[J]. European Journal of Radiology, 2022, 152: 110326. DOI: 10.1016/j.ejrad.2022.110326.
[29] MOTOSUGI U, ICHIKAWA T, SOU H, et al. Multi-organ perfusion CT in the abdomen using a 320-detector row CT scanner: preliminary results of perfusion changes in the liver, spleen, and pancreas of cirrhotic patients[J]. European journal of radiology, 2012, 81(10): 2533−2537. DOI: 10.1016/j.ejrad.2011.11.054.
[30] SAUTER AW, SPIRA D, SCHULZE M, et al. Explanations for the heterogeneity of splenic enhancement derived from blood flow kinetic measurements using dynamic contrast-enhanced CT (DCE-CT)[J]. Acta radiologica, 2014, 55(6): 645−653. DOI: 10.1177/0284185113503322.
[31] 周欢. CT灌注肾峰值模型评估食管胃底静脉曲张破裂出血风险的应用价值[D]. 南华大学, 2022. DOI: 10.27234/d.cnki.gnhuu.2022.000666. [32] DONG J, ZHANG Y, WU Y F, et al. Computed tomography perfusion in differentiating portal hypertension: A correlation study with hepatic venous pressure gradient[J]. World Journal of Gastrointestinal Surgery, 2023, 15(4): 664−673. DOI: 10.4240/wjgs.v15.i4.664.
[33] WANG L, ZHANG Y, WU Y F, et al. Computed tomography perfusion in liver and spleen for hepatitis B virus-related portal hypertension: A correlation study with hepatic venous pressure gradient[J]. World Journal of Gastroenterology, 2022, 28(42): 6068−6077. DOI: 10.3748/wjg.v28.i42.6068.
[34] LIU Y, NING Z, ÖRMECI N, et al. Deep Convolutional Neural Network-Aided Detection of Portal Hypertension in Patients With Cirrhosis[J]. Clinical Gastroenterology and Hepatology, 2020, 18(13): 2998-3007. e5. DOI: 10.1016/j.cgh.2020.03.034.
[35] MENG D, WEI Y, FENG X, et al. CT-Based Radiomics Score Can Accurately Predict Esophageal Variceal Rebleeding in Cirrhotic Patients[J]. Frontiers in Medicine, 2021, 8: 745931. DOI: 10.3389/fmed.2021.745931.
[36] LAMBIN P, RIOS-VELAZQUEZ E, LEIJENAAR R, et al. Radiomics: extracting more information from medical images using advanced feature analysis[J]. European Journal of Cancer, 2012, 48(4): 441−446. DOI: 10.1016/j.ejca.2011.11.036.
[37] PARK H J, PARK B, LEE S S. Radiomics and Deep Learning: Hepatic Applications[J]. Korean journal of Radiology, 2020, 21(4): 387−401. DOI: 10.3348/kjr.2019.0752.
[38] YAN Y, LI Y, FAN C, et al. A novel machine learning-based radiomic model for diagnosing high bleeding risk esophageal varices in cirrhotic patients[J]. Hepatology International, 2022, 16(2): 423−432. DOI: 10.1007/s12072-021-10292-6.
[39] LIU H, SUN J, LIU G, et al. Establishment of a non-invasive prediction model for the risk of oesophageal variceal bleeding using radiomics based on CT[J]. Clinical Radiology, 2022, 77(5): 368−376. DOI: 10.1016/j.crad.2022.01.046.
[40] LIN Y, LI L, YU D, et al. A novel radiomics-platelet nomogram for the prediction of gastroesophageal varices needing treatment in cirrhotic patients[J]. Hepatology international, 2021, 15(4): 995−1005. DOI: 10.1007/s12072-021-10208-4.
[41] LIU Z, XU L, QIN N, et al. Prediction of esophageal and gastric varices rebleeding for cirrhotic patients based on deep learning[J]. Biomedical Signal Processing and Control, 2023, 81: 104420. DOI: 10.1016/J.BSPC.2022.104420.
[42] LI LJ, LIN Y K, YU D X, et al. A Multi-Organ Fusion and LightGBM Based Radiomics Algorithm for High-Risk Esophageal Varices Prediction in Cirrhotic Patients[J]. Ieee Access, 2021, 9: 15041−15052. DOI: 10.1109/Access.2021.3052776.
[43] LEE M, LEE S S, CHOI W M, et al. An index based on deep learning-measured spleen volume on CT for the assessment of high-risk varix in B-viral compensated cirrhosis[J]. European Radiology, 2021, 31(5): 3355−3365. DOI: 10.1007/s00330-020-07430-3.
[44] DENG H, QI X, GUO X. Diagnostic Accuracy of APRI, AAR, FIB-4, FI, king, lok, forns, and fibroindex scores in predicting the presence of esophageal varices in liver cirrhosis: A systematic review and meta-analysis[J]. Medicine, 2015, 94(42): e1795. DOI: 10.1097/MD.0000000000001795.
[45] 李欣忆, 李娇娇, 李莹莹, 等. 基于APRI和PALBI构建的列线图对肝硬化并发食管胃底静脉曲张破裂出血的预测价值[J]. 临床肝胆病杂志, 2024, 40(3): 521−526. DOI: 10.12449/JCH240314. LI X Y, LI J J, LI Y Y, et al. Establishment of a nomogram model for predicting liver cirrhosis with esophagogastric variceal bleeding based on aspartate aminotransferase-to-platelet ratio index and platelet-albumin-bilirubin score[J]. Journal of Clinical Hepatology, 2024, 40(3): 521−526. DOI: 10.12449/JCH240314.
[46] 汤泽宇, 顾菲, 周健文, 等. CT参数联合FIB-4和血清VEGF水平预测乙型肝炎肝硬化患者并发食管静脉曲张效能分析[J]. 实用肝脏病杂志, 2024, 27(4): 579−582. DOI: 10.3969/j.issn.1672-5069.2024.04.022. TANG Z Y, GU F, ZHOU J W, et al. Diagnostic efficacy of CT parameters and FIB-4 and serum VEGF levels in predicting esophageal varices in patients with hepatitis B-induced liver cirrhosis[J]. Journal of Practical Hepatology, 2024, 27(4): 579−582. DOI: 10.3969/j.issn.1672-5069.2024.04.022.
[47] KUMAR S, DUAN Q, WU R, et al. Pathophysiological communication between hepatocytes and non-parenchymal cells in liver injury from NAFLD to liver fibrosis[J]. Advanced Drug Delivery Reviews, 2021, 176: 113869. DOI: 10.1016/j.addr.2021.113869.
[48] 张成孟, 丁治民, 孙宵宇, 等. 定量CT肝脏脂肪含量联合临床指标预测肝硬化患者食管胃底静脉曲张破裂出血风险的初步研究[J]. 放射学实践, 2024, 39(7): 902−906. DOI: 10.13609/j.cnki.1000-0313.2024.07.008. ZHANG C M, DING Z M, SUN X Y, et al. Quantitative CT liver fat content combined with clinical indicators in the prediction of the risk of esophagogastric variceal bleeding in patients with liver cirrhosis[J]. Radiologic Practice, 2024, 39(7): 902−906. DOI: 10.13609/j.cnki.1000-0313.2024.07.008.
[49] LUO R, GAO J, GAN W, et al. Clinical-radiomics nomogram for predicting esophagogastric variceal bleeding risk noninvasively in patients with cirrhosis[J]. World Journal of Gastroenterology, 2023, 29(6): 1076−1089. DOI: 10.3748/wjg.v29.i6.1076.
[50] LIU H, SUN J, LIU G, et al. Establishment of a non-invasive prediction model for the risk of oesophageal variceal bleeding using radiomics based on CT[J]. Clinical Radiology, 2022, 77(5): 368−376. DOI: 10.1016/j.crad.2022.01.046.
计量
- 文章访问数: 70
- HTML全文浏览量: 26
- PDF下载量: 8