DenseNet网络深度学习分析CT图像鉴别肺结节良恶性的价值

  目的  探究DenseNet网络深度学习分析CT图像鉴别肺结节良恶性的价值。  方法  选取2017年2月~2019年5月我院收治的疑似肺结节患者80例,患者均进行CT扫描和DenseNet网络深度学习的人工智能系统诊断其良恶性,以病理结果作为金标准。分析CT图像、DenseNet网络深度学习分析联合CT图像对肺结节良恶性的诊断价值。  结果  CT图像表现肺密度增高影,有云雾状阴影,可清晰显示支气管内血管情况,评估结节良恶性准确率为88.75%,敏感度为76.92%,特异性为94.44%,与病理诊断的Kappa值为0.736（P < 0.001）;DenseNet网络深度学习联合CT评估结节良恶性的敏感度为96.15%,特异性为88.89%,DenseNet网络深度学习联合CT评估准确率高于单纯CT评估准确率（91.25% vs 88.75%）,且与病理诊断一致性较好（Kappa= 0.810,P < 0.001）。  结论  DenseNet网络深度学习分析CT图像鉴别肺结节良恶性准确性较高,且与病理结果具有较好的一致性。      

64层CT在肺内孤立性结节或肿块良恶性鉴别诊断中的应用

目的:探讨64层CT容积扫描(VCT)在肺内结节或肿块诊断与鉴别诊断中的应用价值。材料与方法:收集经临床证实的肺内结节或肿块患者68例,所有病例均行64层CT增强扫描或双层CT引导下穿刺活检的,组织本科室两名高级职称医师,采用"双盲法"分析CT图像。结果:CT诊断良性33例,病理诊断良性28例,误诊5例,病理符合率84.8%;CT诊断恶性35例,病理诊断恶性33例,误诊2例,病理符合率92.3%。结论:64层CT容积扫描及相应后处理功能更有利于显示良恶性结节或肿块的特征性表现,为其良恶性的诊断与鉴别诊断提供可靠依据,故64层VCT对肺孤立性结节或肿块的定性诊断有着重要意义。     

动态CT增强扫描对孤立性肺结节的诊断价值分析

目的:分析动态CT增强扫描对孤立性肺结节的诊断价值。材料与方法:对我院2010年1月至2013年12月间收治的经病理学检查后确诊的54例孤立性肺结节患者临床资料进行回顾性分析,所有患者均给予动态CT增强扫描,将CT扫描结果与病理检查结果进行对比分析。结果:与病理检查结果比较,动态CT增强扫描诊断良性肺结节符合率93.33%,恶性肺结节符合率95.83%;良性肺结节与恶性肺结节增强扫描后与增强前相比有明显升高(P0.05);恶性肺结节的CT值变化情况与良性相比差异具有显著性(P0.05)。结论:动态CT增强扫描在孤立性肺结节诊断及鉴别诊断中均有较高的应用价值,值得临床推广应用。     

增强CT扫描鉴别对于提高诊断肺内孤立性结节的准确率探讨

目的:研究探讨增强CT扫描对于提高肺内肺内孤立性结节诊断鉴别准确率的应用价值。方法:选取我院收治的肺内孤立性结节患者92例作为研究对象,所有患者均在常规CT平扫的基础上接受增强CT扫描,将增强CT扫描的结果与与穿刺活检或手术获得的病理诊断结果进行对照,计算其鉴别良恶性病变的准确度,比较增强CT不同扫描时间良性结节、炎性结节、恶性结节的CT值变化情况及其各自的特征性参数(强化峰值、肺内孤立性结节强化峰值与主动脉强化值之比)。结果:经最终病理诊断,良恶性病变分别有42例、50例,将增强CT扫描结果与病理诊断结果对照,则良性病变、恶性病变的诊断准确率分别为94.24%、90.00%。不同类型结节患者随着CT增强扫描的延长,CT值均呈先升高后下降的趋势,而PH、SPH/PPH计算结果的比较则有炎性结节高于恶性结节高于良性结节的情况,比较均有统计学差异,P0.05。结论:增强CT扫描可以对肺内孤立性结节的良恶性进行充分鉴别,鉴别的准确率高,可以为治疗提供可靠依据,值得临床推广应用。     

DCE-MRI定量渗透性参数联合ADC值在肺部良恶性病变中鉴别诊断价值

目的探讨利用动态增强磁共振成像(dynamic contrast-enhanced MRI,DCE-MRI)的血流动力学双室模型渗透性参数联合ADC值鉴别诊断肺部良恶性病变的应用价值。材料与方法搜集2014年7月至2015年3月在我院行肺部CT增强检查、常规MRI扫描、DWI成像及DCE-MRI扫描的患者共49例(恶性病变29例,良性病变20例),MRI动态增强扫描采用三维快速容积扫描技术,通过MRI后处理工作站计算病灶的ADC值,血流动力学定量分析软件Omni-Kinetics计算病灶的微血管渗透性参数Ktrans、Kep等。结果 CT联合ADC值、DCE-MRI定量渗透性参数鉴别诊断肺良恶性结节准确率为93.9%;CT联合ADC值鉴别诊断肺良恶性结节准确率为85.7%;CT鉴别肺良恶性结节准确率为75.5%,CT联合ADC值、DCE-MRI定量渗透性参数鉴别肺良恶性结节准确性与CT有显著性差异(P0.05)。结论 DCE-MRI定量分析微血管渗透性参数Ktrans、Kep联合ADC值对肺部良恶性病变诊断效能高于CT,而且真正实现定量分析鉴别诊断肺部结节良恶性,值得广泛应用于临床工作。     

增强CT直方图分析鉴别甲状腺良恶性结节

目的观察基于CT平扫和增强图像直方图特征鉴别诊断甲状腺良恶性结节的价值。方法回顾性分析经手术后病理证实的甲状腺结节患者132例,共140个结节。选取轴位病灶最大层面,采用Mazda软件沿病灶边缘勾画ROI并进行直方图分析,获取9个参数,比较良恶性结节的差异,并以ROC曲线分析差异有显著统计学意义的灰度直方图参数对甲状腺良恶性结节的鉴别诊断效能。结果CT平扫恶性结节均值及第10、50、90百分位数高于良性结节(P均<0.05);增强后恶性结节均值及第1、10、50、90百分位数高于良性结节(P均<0.05),良性结节方差高于恶性结节(P<0.05)。两者偏度、峰度、第99百分位数在CT平扫和增强中差异均无显著统计学意义(P均>0.05)。CT平扫和增强直方图参数中,第10百分位数AUC最高,为0.68,鉴别甲状腺良恶性结节的敏感度和特异度分别为74.32%和62.12%。结论CT直方图分析可作为鉴别甲状腺良恶性结节的重要辅助手段。     

单源双能CT鉴别诊断甲状腺结节良恶性

目的 探讨单源双能CT在体碘含量定量分析技术鉴别诊断甲状腺结节良恶性的价值。方法 55例甲状腺结节患者接受GE单源双能量CT平扫及增强扫描,选择甲状腺结节的实性部分、未受累正常腺体作为ROI,在专用的图像分析软件上测量其在不同单能量时平扫及增强CT值、平扫碘含量,比较甲状腺良恶性结节不同单能量平扫CT值、强化程度及平扫碘含量、不同单能量时CT值衰减幅度的差异。结果 平扫时甲状腺恶性结节碘含量明显低于良性结节(P< 0.001),ROC曲线下面积(AUC)为0.93。甲状腺良恶性结节平扫40 keV和140 keV单能量时CT值衰减幅度差异有统计学意义(P<0.001),ROC的AUC为0.92。结论 甲状腺良恶性结节的碘含量、不同单能量时CT值衰减幅度均存在差异,单源双能量CT技术可能为鉴别诊断甲状腺良恶性结节提供有价值信息。     

超声造影对甲状腺单发实性结节良恶性的诊断价值

目的探讨超声造影对甲状腺单发实性结节良恶性的诊断价值。方法 75例经病理证实为甲状腺单发实性结节,先行常规超声检查,后超声造影观察其增强特征和时间-强度曲线定量参数:峰值强度(Peak)、达峰时间(Tp)、曲线下面积(AUC)及造影剂平均通过时间(MTT),判断结节的良恶性。结果甲状腺良性结节造影后形态多规则,边界清晰,明显增强,分布均匀且无灌注缺损;恶性结节表现为形态不规则,边界不清,无明显增强,分布不均匀,可见灌注缺损;两者差异有统计学意义(P0.05)。与常规超声比较,超声造影测量甲状腺良性结节大小差异无统计学意义,测量恶性结节直径较大(P0.05)。与良性结节比较,恶性结节AUC和Peak降低,Tp延迟(P0.05);两组MTT差异无统计学意义。以病理结果为金标准,超声造影诊断甲状腺良恶性结节的符合率为89.3%。结论甲状腺良恶性结节造影增强特征明显不同,定量参数Peak、AUC及Tp可作为甲状腺良恶性结节鉴别诊断中的参考指标。     

64排螺旋CT低剂量扫描对肺孤立性小结节良恶性病变的鉴别诊断价值分析

目的:探讨64排螺旋CT低剂量扫描鉴别诊断肺孤立性小结节性良恶性病变的临床价值。方法:选取2017年5月～2019年5月进行检查的70例肺孤立性小结节患者作为研究对象,所有患者均接受64排螺旋CT低剂量扫描。以病理学诊断为"金标准",分析64排螺旋CT低剂量扫描在肺孤立性小结节良恶性病变鉴别诊断中的应用价值,比较良性、恶性肺孤立性小结节的影像学特征及良、恶性肺孤立性小结节不同时段的CT值。结果:64排螺旋CT低剂量扫描诊断肺孤立性小结节的灵敏度为94.74%,特异度为84.38%,准确度为90.00%;肺孤立性小结节良性病变中边缘毛刺征、分叶状或不规则状检出率及内部结构卫星灶、均匀检出率均低于恶性病变,边缘清晰检出率、内部结构钙化检出率均高于恶性病变,差异有统计学意义(P0.05);肺孤立性小结节良性病变30 s、90 s及180 s CT值均低于恶性病变,差异有统计学意义(P0.05)。结论:64排螺旋CT低剂量扫描可有效鉴别肺孤立性小结节良、恶性,为临床明确病情、指导治疗及预后评估提供了参考依据。     

肺结节诊断中16层螺旋CT的应用研究

目的观察16层螺旋CT在肺结节诊断中的应用效果。方法选择某院80例肺结节患者实施此次研究,时间为2018年1月至12月,所有患者均实施16层螺旋CT平扫及增强扫描,观察患者临床效果。结果对比良性、恶性结节患者不同序列强化值检测结果及CT平扫、增强扫描CT值,差异无统计学意义(P0.05),在增强扫描状态下,不同时间,组间对比差异明显,良性结节组CT值明显低于恶性结节组(P0.05)。所有患者经CT扫描,良性结节为25.00%,恶性结节为75.00%。良、恶性结节可划分为不同类型。对80例肺结节患者的16层螺旋CT扫描结果与病理结果对比,准确率为85.00%,其中,20例良性结节患者中误诊患者1例,60例恶性结节患者中误诊11例。结论将16层螺旋CT诊断用于肺结节临床诊断中,可有效提升诊断准确率。     

基于增强双期CT成像的肺亚厘米结节良恶性预测模型

  目的  分析基于增强双期CT成像的肺亚厘米结节良恶性预测模型。  方法  选择我院2019年1月~2021年3月收治的98例肺亚厘米结节患者作为研究对象，依照病理诊断结果分为良性病变组（n=64）和恶性病变组（n=34）。所有受试者行基于增强双期CT成像，采用Logistic回归模型分析增强双期CT成像预测结节良恶性预测模型，绘制ROC曲线分析增强双期CT成像的肺亚厘米结节良恶性预测模型的应用价值。  结果  良性病变组患者毛刺、结节边界清楚、上叶、分叶征、空泡征、胸膜凹陷征、血管集束征、磨玻璃密度发生率与恶性病变组的差异有统计学意义（P < 0.05）；增强双期CT成像预测肺亚厘米结节良恶性预测模型为Log(P)=1.211×毛刺+2.843×分叶+1.981×磨玻璃+0.793×边界不清+1.326；增强双期CT成像预测肺亚厘米结节良恶性预测模型预测患者肺亚厘米结节良恶性的曲线下面积为0.930（P < 0.05）。  结论  基于增强双期CT成像预测肺亚厘米结节良恶性模型临床价值较高，具有较高的预测价值。      

Management of solitary pulmonary nodules

The solitary pulmonary nodule (SPN), a single intrapulmonary spherical lesion that is fairly well circumscribed, is a common clinical problem. About half of SPNs seen in clinical practice are malignant, usually bronchogenic carcinomas. Some nodules are primary tumors of other kinds or metastatic. Virtually all benign SPNs are tuberculous or fungal granulomas. The standard management of the SPN of unknown cause is prompt surgical removal unless benignity is established by prior chest roentgenograms showing that the nodule has been stable (i.e., showing no growth) for 2 years or by the presence of a "benign" pattern of calcification. Less universally accepted criteria for benignity include (1) transthoracic needle aspiration biopsy (TNAB) showing a specific benign process, and (2) patient's age under 30 to 35 years. Bronchoscopy has a low diagnostic yield, particularly for benign nodules. SPNs usually grow at constant rates, expressed as the "doubling time" (DT). A nodule with a DT between 20 and 400 days is usually malignant. Benign nodules usually have a DT greater than 400 days. The prospective determination of DT by serial chest roentgenograms (the "wait and watch" strategy) is widely criticized but has clinical utility in special circumstances, particularly if the likelihood of malignancy is low and/or the anticipated surgical mortality is high. The presence and pattern of calcification are best shown by high-resolution thin-section computed tomography (CT). Diffuse, laminated, central or "popcorn" patterns of calcification indicate benignity. An eccentric calcium deposit or a stippled pattern does not rule out malignancy. CT densitometry will often show "occult" calcification in nodules that show no direct visual evidence of calcium deposition. The characteristics of the edge of the nodule correlate with the likelihood of malignancy. Nodules with irregular or spiculated margins are almost always malignant. The probability that the nodule is malignant (pCA) is related to the age of the patient, the diameter of the nodule, the amount of tobacco smoke inhalation, the overall prevalence of malignancy in SPNs, the nature of the edge of the lesion, and the presence or absence of occult calcification. It is possible by Bayesian techniques to combine these factors to calculate a more precise and comprehensive prediction of pCA in any given nodule. The 5-year survival after nodule resection depends on the size of the nodule at the time of surgery; it may be as high as 80% with nodules that are 1 cm in diameter. Lymph node involvement is uncommon with small tumors, and many authorities question the need for CT staging in such cases.(ABSTRACT TRUNCATED AT 400 WORDS)     

超声检测血流动力学参数对甲状腺良恶性结节的诊断价值

  目的  分析超声通过血流动力学参数评估甲状腺结节良恶性及其诊断的敏感度和特异性。  方法  选择2020年6月~2022年1月收治的122例甲状腺结节患者,依照病理检测结果将患者分为良性结节组（n=78）和恶性结节组（n=44）,对患者进行彩色超声扫描检查,术后切除的组织样品行常规病理学检查。以病理诊断结果为金标准,绘制ROC曲线分析各指标单独应用及联合应用评估甲状腺结节良恶性的敏感度和特异性。  结果  良性结节组患者病灶区收缩期峰值流速（PSV）及阻力指数（RI）值均明显低于恶性结节组,舒张末期血流速度（EDV）水平高于恶性结节组（P < 0.05）;PSV、EDV及RI联合应用预测甲状腺结节良恶性预测价值为Log(P)=-0.654×PSV+0.627×EDV-0.608×RI+0.613;PSV、EDV及RI联合应用预测甲状腺结节良恶性敏感度、特异度及曲线下面积均明显高于各指标单独应用（P < 0.05）。  结论  通过超声扫描检查血流动力学参数评估甲状腺结节良恶性具有较高的敏感度和特异性。      

AmCAD-UT Detection在甲状腺结节超声诊断中的应用

目的 评估计算机辅助诊断（CAD）系统AmCAD-UT Detection（安克侦）用于甲状腺超声的诊断效能及临床价值。方法 采集171例甲状腺结节患者的甲状腺超声图像，分别由安克侦及4名超声科医师（A、B、C、D，分别具有10年、5年、1年及1个月以上甲状腺超声诊断经验）单独及以安克侦辅助医师分析图像，并根据美国放射学会甲状腺影像报告和数据系统（ACR-TIRADS）指南进行分类；以病理结果为金标准，绘制安克侦及4名医师辅以安克侦前后根据ACR-TIRADS指南对结节进行分类的ROC曲线，计算ACR-TIRADS指南诊断良恶性结节的最佳截断值及AUC，评价其诊断效能。结果 共纳入205个甲状腺结节，89个良性、116个恶性病变。ACR-TIRADS指南诊断良恶性结节的最佳截断值为TR5级。安克侦诊断甲状腺恶性结节的敏感度与医师B差异无统计学意义（P=1.00），特异度则低于医师A及B（P均<0.05），其AUC与医师A、B、D差异均有统计学意义（Z=4.34、3.71、2.76，P均<0.05）。辅以安克侦后，4名医师诊断甲状腺结节的敏感度（93.10%、90.52%、85.34%、75.00%）及AUC值（0.95、0.93、0.86、0.86）均较前提高（P均<0.05），特异度则仅医师C、D较前改善（P均<0.05）。结论 安克侦对诊断甲状腺结节具有一定价值，敏感度与具有5年诊断经验的超声科医师相似，用以辅助可提高超声科医师、尤其是低年资医师对于甲状腺结节的诊断效能。     

Quantitative Analysis of Dynamic Power Doppler Sonograms for Patients with Thyroid Nodules

To clarify and determine whether power Doppler sonograms are useful for the detection of malignant thyroid nodules, a computerized quantification method was used to evaluate the vascular density of a thyroid nodule in a prospective setting. Sonographic power Doppler images were collected in consecutive frames (45 frames of images), and a proprietary program (AmCAD-UV) was implemented using methods proposed in this article automatically calculated a quantified power Doppler vascular index (PDVI). The minimum PDVI value (PDVI_min) was suggested as a measure of the vascular density of the nodule. The vascular densities of the peripheral and central areas of the nodule, referred to as central PDVI_min and Ring PDVI_min, respectively, were also evaluated. For 238 tumors (79 malignant and 159 benign) from 208 patients, all of the proposed indices of benign lesions were significantly higher than those of the malignant lesions. The area under the receiver operating characteristic curve (AUC) reaches 71% with the PDVI_min. When the vascular patterns were further classified into intra-nodular and peripheral vascularity types, no vascularity type was observed significantly more frequently in malignant nodules than in benign nodules. These proposed computerized vascular indices provide a quantification method to objectively evaluate thyroid nodules and have potential as predictors of thyroid malignancy. The conventional vascular characterizations of malign nodules, that is, more vessels are observed in malignant nodules than in benign nodules, are shown to be unreliable in our study. Instead, a higher value of the quantified power Doppler vascular density was observed in benign nodules.     

Accurate diagnosis of pulmonary nodules using a noninvasive DNA methylation test

BACKGROUNDCurrent clinical management of patients with pulmonary nodules involves either repeated low-dose CT (LDCT)/CT scans or invasive procedures, yet causes significant patient misclassification. An accurate noninvasive test is needed to identify malignant nodules and reduce unnecessary invasive tests.METHODWe developed a diagnostic model based on targeted DNA methylation sequencing of 389 pulmonary nodule patients’ plasma samples and then validation in 140 plasma samples independently. We tested the model in different stages and subtypes of pulmonary nodules.RESULTSA 100-feature model was developed and validated for pulmonary nodule diagnosis; the model achieved a receiver operating characteristic curve–AUC (ROC-AUC) of 0.843 on 140 independent validation samples, with an accuracy of 0.800. The performance was well maintained in (a) a 6 to 20 mm size subgroup (n = 100), with a sensitivity of 1.000 and adjusted negative predictive value (NPV) of 1.000 at 10% prevalence; (b) stage I malignancy (n = 90), with a sensitivity of 0.971; (c) different nodule types: solid nodules (n = 78) with a sensitivity of 1.000 and adjusted NPV of 1.000, part-solid nodules (n = 75) with a sensitivity of 0.947 and adjusted NPV of 0.983, and ground-glass nodules (n = 67) with a sensitivity of 0.964 and adjusted NPV of 0.989 at 10% prevalence. This methylation test, called PulmoSeek, outperformed PET-CT and 2 clinical prediction models (Mayo Clinic and Veterans Affairs) in discriminating malignant pulmonary nodules from benign ones.CONCLUSIONThis study suggests that the blood-based DNA methylation model may provide a better test for classifying pulmonary nodules, which could help facilitate the accurate diagnosis of early stage lung cancer from pulmonary nodule patients and guide clinical decisions.FUNDINGThe National Key Research and Development Program of China; Science and Technology Planning Project of Guangdong Province; The National Natural Science Foundation of China National.     

Can Color Doppler Sonography Aid in the Prediction of Malignancy of Thyroid Nodules?

OBJECTIVE: To determine whether color Doppler interrogation of a thyroid nodule can aid in the prediction of malignancy. METHODS: We obtained color Doppler images of thyroid nodules undergoing sonographically guided fine-needle aspiration. The color Doppler appearance of each nodule was graded from 0 for no visible flow through 4 for extensive internal flow. The size, sonographic appearance, results of fine-needle aspiration, and surgical pathologic findings, if available, were recorded for each nodule. RESULTS: There were 254 nodules sampled, of which 32 were malignant (all confirmed at surgery) and 177 were benign. Fourteen (43.8%) of the 32 malignant nodules were color type 4, compared with only 26 (14.7%) of the 177 benign nodules (P = .0004, Fisher exact test). Thirteen (40.1%) of the 32 malignant nodules were solid, as were 18 (10.2%) of the 177 benign nodules (P = .006, Fisher exact test). Among solid nodules, the prevalence of malignancy was greater when the nodule was hypervascular (13 [41.9%] of 31) than when the color type was less than 4 (11 [14.7%] of 77; P = .004, Fisher exact test). CONCLUSIONS: Solid hypervascular thyroid nodules have a high likelihood of malignancy (nearly 42% in our series). The color characteristics of a thyroid nodule, however, cannot be used to exclude malignancy, because 14% of solid nonhypervascular nodules were malignant.     

应用计算机人工神经网络系统结合CT与PET鉴别良、恶性肺结节

目的 利用肺结节CT、PET特征,开发计算机人工神经网络(ANN)辅助诊断系统,评价其对肺结节良恶性的鉴别能力。方法 连续收集112例肺内单发小结节(<3.0 cm)患者,均接受PET/CT及胸部CT检查,二者间隔小于1个月。112例患者中恶性肺结节52例,良性60例,均经组织学或临床随诊证实。利用结节的CT特征及PET特征开发计算机ANN辅助诊断系统。计算机ANN的训练及测试采用Round-Robin方法。采用ROC方法评价计算机ANN输出结果并进行统计学分析。结果 CT计算机ANN程序采用20个输入单元,包括4个临床特征及16个CT特征,ROC曲线下面积(Az)为0.83;PET计算机ANN程序采用4个临床特征及1个PET特征作为5个输入单元,Az值为0.91;CT+PET计算机ANN程序采用临床特征CT及PET所有21个输入单元,Az值为0.95。与CT计算机ANN程序、PET计算机ANN程序相比,CT+PET计算机ANN程序输出结果明显提高(P＝0.015、0.037)。 结论 CT+PET ANN计算机辅助诊断程序输出结果优于单纯PET或CT计算机ANN结果。当PET对肺结节诊断有困难时,结节的CT特征有助于鉴别诊断。     

MSCS-DeepLN: Evaluating lung nodule malignancy using multi-scale cost-sensitive neural networks

The accurate identification of malignant lung nodules using computed tomography (CT) screening images is vital for the early detection of lung cancer. It also offers patients the best chance of cure, because non-invasive CT imaging has the ability to capture intra-tumoral heterogeneity. Deep learning methods have obtained promising results for the malignancy identification problem; however, two substantial challenges still remain. First, small datasets cannot insufficiently train the model and tend to overfit it. Second, category imbalance in the data is a problem. In this paper, we propose a method called MSCS-DeepLN that evaluates lung nodule malignancy and simultaneously solves these two problems. Three light models are trained and combined to evaluate the malignancy of a lung nodule. Three-dimensional convolutional neural networks (CNNs) are employed as the backbone of each light model to extract the lung nodule features from CT images and preserve lung nodule spatial heterogeneity. Multi-scale input cropped from CT images enables the sub-networks to learn the multi-level contextual features and preserve diverse. To tackle the imbalance problem, our proposed method employs an AUC approximation as the penalty term. During training, the error in this penalty term is generated from each major and minor class pair, so that negatives and positives can contribute equally to updating this model. Based on these methods, we obtain state-of-the-art results on the LIDC-IDRI dataset. Furthermore, we constructed a new dataset collected from a grade-A tertiary hospital and annotated using biopsy-based cytological analysis to verify the performance of our method in clinical practice.     

肺腺癌患者CT影像预测纯磨玻璃结节浸润程度及对肺结节病理性质的判定价值

  目的  研究肺腺癌患者CT影像对纯磨玻璃结节(PGGN)浸润程度及对肺结节病理性质的预测价值。  方法  采用回顾性研究方法,以我院2017年1月~2021年1月住院治疗的122例PGGN患者作为研究对象。根据肺腺癌的种类分组,其中侵袭前组患者82例原位腺癌(AIS)组患者39例,微浸润性腺癌(MIA)组患者43例,侵袭组患者40例; 根据肺腺癌良恶性分组,其中良性组80例,恶性组42例。比较侵袭前组与侵袭组、AIS组与MIA组以及良性组与恶性组患者的病灶直径、CT值、形状、瘤肺界面、毛刺征、空泡征、三维形状、分叶征、空气支气管征、胸膜凹陷症情况之间的差异。  结果  侵袭前组以及侵袭组患者的病灶直径、CT值、形状、毛刺征、分叶征、空气支气管征情况之间的差异有统计学意义(P < 0.05);AIS以及MIA组患者的分叶征、空泡征、CT值之间的差异有统计学意义(P < 0.05);良性组以及恶性组患者的瘤肺界面、病灶直径、CT值、形状、毛刺征、分叶征、空气支气管征情况之间的差异有统计学意义(P < 0.05);病灶直径、CT值、形状、毛刺征、分叶征、空气支气管征联合检测对肺癌的侵袭以及恶性组患者的诊断敏感度高于单独检测; ROC曲线分析示,病灶直径、CT值、形状、毛刺征、分叶征、空气支气管征联合检测对浸润性以及恶性肿瘤的曲线下面积高于单独检测。  结论  肺腺癌患者CT影像对于PGGN浸润程度及对肺结节病理性质具有显著的预测价值。