CT透视引导下胸部小病灶穿刺活检的临床价值

目的探讨CT引导胸部小病灶穿刺活检技术及诊断价值。方法 50例胸部小病灶(<3 cm)在CT引导下采用自动切割式活检针穿刺活检。结果50例胸部肿块穿刺活检病灶取材成功100%。胸部肿块穿刺活检的诊断准确率96%(48/50),对恶性病变诊断准确率97.5%(39/40),良性病变为90%(9/10)。并发气胸2例,少量出血1例。结论 C T引导下穿刺活检是安全、有效的诊断手段,对于胸部病变诊断的准确率高,CT引导下胸部小病灶穿刺活检对胸部病变的诊断和治疗具有重要意义。     

Computer-aided differentiation of malignant from benign solitary pulmonary nodules imaged by high-resolution CT

We investigated the possibility of using computer analysis of high-resolution CT images to radiologically classify the shape of pulmonary nodules. From a total of 107 HRCT images of solid, solitary pulmonary nodules with prior differentiation as benign (n=55) or malignant (n=52), we extracted the desired pulmonary nodules and calculated two quantitative parameters for characterizing nodules: circularity and second central moment. Using discriminant analysis for two thresholds in differentiating malignant from benign states resulted in a sensitivity of 76.9%, a specificity of 80%, a positive predictive value of 78.4%, and a negative predictive value of 78.6%.     

Automated detection of small pulmonary nodules in whole lung CT scans

RATIONALE AND OBJECTIVES: The objective of this work was to develop and evaluate a robust algorithm that automatically detects small solid pulmonary nodules in whole lung helical CT scans from a lung cancer screening study. MATERIALS AND METHODS: We developed a three-stage detection algorithm for both isolated and attached nodules. The algorithm consisted of nodule search space demarcation, nodule candidates' generation, and a sequential elimination of false positives. Isolated nodules are nodules that are surrounded by lung parenchyma, whereas attached nodules are connected to large, dense structures such as pleural and/or mediastinal surface. Two large well-documented whole lung CT scan databases (Databases A and B) were created to train and test the detection algorithm. Database A contains 250 sequentially selected scans with 2.5-mm slice thickness that were obtained at Weill Medical College of Cornell University. With equipment upgrade at this college, a second database, Database B, was created containing 250 scans with a 1.25-mm slice thickness. A total of 395 and 482 nodules were identified in Databases A and B, respectively. In both databases, the majority of the nodules were isolated, comprising 72.1% and 82.3% of nodules in Databases A and B, respectively. RESULTS: The detection algorithm was trained and tested on both Databases A and B. For isolated nodules with sizes 4 mm or larger, the algorithm achieved 94.0% sensitivity and 7.1 false positives per case (FPPC) for Database A (2.5 mm). Similarly, the algorithm achieved 91% sensitivity and 6.9 FPPC for Database B (1.25 mm). The algorithm achieved 92% sensitivity with 17.4 FPPC and 89% sensitivity with 5.5 FFPC for attached nodules with sizes 3 mm or larger in the Database A (2.5 mm) and Database B (1.25 mm), respectively. CONCLUSION: The developed algorithm achieved practical performance for automated detection of both isolated and the more challenging attached nodules. The automated system will be a useful tool to assist radiologists in identifying nodules from whole lung CT scans in a clinical setting.     

CT灌注成像在肺部良恶性结节鉴别诊断中的潜能

目的确定CT灌注成像在肺部良恶性结节鉴别诊断中的潜能。方法 32例经临床和病理证实为孤立性肺结节患者,其中,24例恶性和8例良性均经CT灌注成像,随后,对两组的灌注参数、强化峰值及达峰时间进行了测量与统计学比较。结果恶性组血流量(BF)为33.35±10.23mL/(100 g·min),良性组BF为23.58±8.65mL/(100 g·min),两组间有显著性差异(P<0.05);恶性组血容量(BV)为11.36±3.89mL/100 g,良性组BV为6.32±3.21 mL/100 g,两组间有显著性差异(P<0.01);恶性组达峰时间(TTP)为32.8±2.9 s,良性组为24.6±2.3 s,两组间有显著性差异(P<0.05);良性与恶性两组间强化峰值无统计学差异(P>0.05)。结论 CT灌注成像不仅是评价全部结节灌注情况的一种可行方法,而且有助于肺部良恶性结节的鉴别。     

基于分形维数对肝脏CT图像的纹理特征研究

 目的用分形维数刻画肝脏图像的纹理特征.方法以差分盒计算方法计算肝脏CT图像表面灰度的分维数值.结果正常肝组织CT图像表面灰度的分维数值小于癌变肝脏组织CT图像表面的分维数值.结论(1)正常肝脏软组织的分维数值大约在2.35左右,而肝癌软组织的分维数大约在2.40;(2)同一脏器软组织的分维数值与它们所处的位置无关,只与组织所表现出来的性质有关,即:是正常组织还是癌变组织;(3)分维数值从某种意义上代表了组织的一些纹理特征.     

基于放疗过程中腮腺图像纹理特征的变化早期预测头颈部肿瘤放射性口干症

目的 分析头颈部肿瘤患者放疗过程中腮腺图像纹理特征的变化,研究其与急性放射性口干症（级别）的关系。建立数学模型,早期预测放射性口干严重性。方法 观察23例头颈部肿瘤放疗的患者,根据放射治疗肿瘤协作组（RTOG）标准评价患者每周口干程度。采集这些患者放疗中每周的验证CT图像,传至MIM系统,勾画出腮腺的轮廓,在MATLAB（R2013a）中开发内部分析程序。分析放疗过程中每周腮腺CT图像的纹理特征的变化,包括平均CT值（MCTN）、标准差（STD）、偏斜度（skewness）、峰度（kurtosis）和熵（entropy）,以及体积的变化。建立数学模型,并利用KNN方法对所建模型进行优化,预测口干级别。结果 平均CT值和体积的变化与口干程度无明显相关性（P>0.05）,但根据二者每周的变化建立模型,预测口干级别,准确度为99%。结论 同时基于平均CT值和相对体积变化建立模型可早期预测口干严重程度。     

Diagnostic performance of PET/CT in differentiation of malignant and benign non-solid solitary pulmonary nodules

OBJECTIVE: To evaluate whether [F-18] fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) can distinguish benign from malignant solitary pulmonary nodules (SPNs) with non-solid components. METHODS: [F-18] FDG-PET/CT scans were performed on 53 consecutive patients (30 men, 23 women; mean age 65 years) who had SPNs with non-solid components identified by CT screening for lung cancer. All patients underwent surgical resection, and all lesions were pathologically proved. Visual score, maximal, and mean standardized uptake value (SUV), and maximal and mean lesion-to-normal tissue count density ratio (LNR) were calculated in all lesions. In addition, clinical characteristics, laboratory test results, and CT findings were assessed. RESULTS: Benign SPNs with non-solid components had a higher uptake on [F-18] FDG-PET/CT. Visual score, maximal and mean SUV, and maximal and mean LNR were significantly higher in the benign when compared with the malignant SPNs (P < 0.001). When the cutoff of 1.5 was assigned for maximal SUV, the diagnostic performance of [F-18] FDG-PET/CT in predicting benign SPN revealed 100.0% sensitivity, 96.4% specificity, and 100.0% accuracy. CONCLUSIONS: [F-18] FDG-PET/CT is useful for the differential diagnosis of SPNs with non-solid components.     

Medical image analysis of 3D CT images based on extension of Haralick texture features

PURPOSE: A new approach to the segmentation of 3D CT images is proposed in an attempt to provide texture-based segmentation of organs or disease diagnosis. 3D extension of Haralick texture features was studied calculating co-occurrences of all voxels in a small cubic region around the voxel. RESULTS: For verification, the proposed method was tested on a set of abdominal 3D volumes of patients. Statistically, the improvement in segmentation was significant for most of the organs considered herein. CONCLUSIONS: The proposed method has potential application in medical image segmentation, including diagnosis of diseases.     

CT征象联合肿瘤标志物对孤立性肺结节良恶性的鉴别诊断分析

目的 探讨CT征象联合肿瘤标志物对孤立性肺结节良恶性的鉴别诊断价值.方法 选取70例孤立性肺结节患者,行高分辨率CT、肿瘤标志物检查及病理检查.以病理组织学检查结果分为恶性结节组39例,良性结节组31例,对比两组CT征象和肿瘤标志物水平差异,运用受试者工作特征曲线(ROC)分析CT联合肿瘤标志物检测鉴别孤立性肺结节良恶...     

Role of baseline nodule density and changes in density and nodule features in the discrimination between benign and malignant solid indeterminate pulmonary nodules

Purpose

To retrospectively evaluate whether baseline nodule density or changes in density or nodule features could be used to discriminate between benign and malignant solid indeterminate nodules.

Materials and methods

Solid indeterminate nodules between 50 and 500 mm³ (4.6–9.8 mm) were assessed at 3 and 12 months after baseline lung cancer screening (NELSON study). Nodules were classified based on morphology (spherical or non-spherical), shape (round, polygonal or irregular) and margin (smooth, lobulated, spiculated or irregular). The mean CT density of the nodule was automatically generated in Hounsfield units (HU) by the Lungcare^© software.

Results

From April 2004 to July 2006, 7310 participants underwent baseline screening. In 312 participants 372 solid purely intra-parenchymal nodules were found. Of them, 16 (4%) were malignant. Benign nodules were 82.8 mm³ (5.4 mm) and malignant nodules 274.5 mm³ (8.1 mm) (p = 0.000). Baseline CT density for benign nodules was 42.7 HU and for malignant nodules −2.2 HU (p = ns). The median change in density for benign nodules was −0.1 HU and for malignant nodules 12.8 HU (p < 0.05). Compared to benign nodules, malignant nodules were more often non-spherical, irregular, lobulated or spiculated at baseline, 3-month and 1-year follow-up (p < 0.0001). In the majority of the benign and malignant nodules there was no change in morphology, shape and margin during 1 year of follow-up (p = ns).

Conclusion

Baseline nodule density and changes in nodule features cannot be used to discriminate between benign and malignant solid indeterminate pulmonary nodules, but an increase in density is suggestive for malignancy and requires a shorter follow-up or a biopsy.     

基于CT和18F-FDG PET/CT的肺癌风险预测模型对肺结节恶性风险的验证研究

目的 比较基于CT的Brock模型、VA模型和基于18F-FDG PET/CT的Herder模型预测肺结节恶性风险的效能,验证模型的预测准确率。 方法 回顾性分析2009年7月至2016年7月行CT检查并经病理确诊或随访确诊的120例肺结节患者,其中可能为恶性的59例患者接受了18F-FDG PET/CT检查。绘制3种模型的受试者工作特征（ROC）曲线并计算曲线下面积（AUC）。测算肺结节（直径4~30 mm）的恶性风险。根据每个模型基于排除标准的队列和所有患者总队列的AUC,验证模型的准确性。采用MedCalc软件进行相关性分析,DeLong方法进行两条ROC的比较。 结果 120例肺结节患者中,49例患者（40.8％）患有恶性结节（31.6％原发性肺癌、8.2％转移性疾病）。在受排除标准限制的队列中,Brock和VA模型的AUC分别为 0.887和0.758,两者间的差异有统计学意义（Z=6.483,P=0.006）。 在接受18F-FDG PET/CT检查的患者中,Herder模型的AUC为0.937。当对队列中的所有患者（即包括原模型纳入标准之外的患者）测试模型时,每个模型的AUC值均有所降低,但两类队列比较差异无统计学意义（Z=21.357,P=0.121）。 对于≤10 mm的结节,Brock和VA 模型的AUC值分别为0.846和0.536,Brock模型明显优于VA 模型（Z=8.768,P=0.0026）。 结论 Brock模型可预测CT扫描中检测到的肺结节恶变的可能性,在接受18F-FDG PET/CT进行肺结节评估的患者中,Herder模型的预测效能最高。      

基于超声影像构建机器学习模型预测甲状腺良恶性结节

目的 构建基于超声影像特征的机器学习模型预测甲状腺结节的良恶性,选择最佳模型以准确预测甲状腺结节的良恶性。 方法 回顾性分析有明确病理结果的甲状腺结节病人2 410例共2 516个结节的超声影像特征。使用SPSS Modeler18.0统计软件,将结节随机分为训练队列和验证队列,训练队列包括1 992个结节（80%）,验证队列包括524个结节（20%）。在训练队列和验证队列中,分别使用支持向量机（SVM）、Logistc回归分析、分类回归树（C&R）、决策树（C5.0）、贝叶斯网络和类神经网络6个分类器构建机器学习模型。采用受试者操作特征（ROC）曲线下面积（AUC）分析模型的原始倾向评分,以评估6种模型的预测能力;并使用DeLong检验比较6种模型的预测能力。选择预测能力最好的机器学习模型,筛选预测重要变量。使用R软件,基于训练队列数据绘制列线图,并基于训练队列及验证队列数据绘制校准曲线对列线图进行验证。 结果 在训练队列和验证队列中,SVM相比其他模型预测甲状腺结节良恶性的能力最好,AUC分别为0.983和0.973（均P<0.05）。选取SVM筛选的6个预测重要变量绘制的列线图显示纵横比>1、微钙化、包膜外侵犯评分最高,其次为边缘、桥本氏甲状腺炎及回声水平。训练队列和验证队列的校准曲线均显示,列线图的预测结果与实际结果有良好的一致性。 结论 基于超声影像特征构建的机器学习模型可以准确预测甲状腺结节的性质,其中SVM的预测能力最高。     

The effects of co-occurrence matrix based texture parameters on the classification of solitary pulmonary nodules imaged on computed tomography

In this project, patients with a solitary pulmonary nodule, were imaged using high resolution computed tomography. Quantitative measures of texture were extracted from these images using co-occurrence matrices. These matrices were formed with different combinations of gray level quantization, distance between pixels and angles. The derived measures were input to a linear discriminant classifier to predict the classification (benign or malignant) of each nodule. Using a relative quantization scheme with eight levels, four features yielded an area under the ROC curve (A_z) of 0.992; 93.8% (30/32) of cases were correctly classified when training and testing on the same cases; while 90.6% (29/32) were correctly classified when jackknifing was used.     

动态增强CT扫描在孤立性肺结节临床诊断中的应用现状和进展

对孤立性肺结节的诊断仍是当前临床工作中的难点之一。借助于动态增强CT扫描，可以了解结节增强形态、增强幅度以及结节的密度随时间变化情况，进而判断结节的性质。就动态增强CT扫描技术具体操作环节及其对结节的诊断价值进行综述。     

孤立性肺结节CT灌注成像技术研究现状与进展

多层螺旋CT灌注成像技术(multi-slice spiral CT perfusion imaging,CTPI)对孤立性肺结节(solitary pulmonary noudules,SPN)的诊断及鉴别诊断已成为近年来研究的热点,CTPI既能提供结节的形态学信息又能提供结节内部血流参数及强化的时间-密度曲线等多种生理学信息,是一种功能成像,目前研究揭示其在SPN的诊断及鉴别诊断中有重要的应用价值.本文着重综述CTPI鉴别SPN的研究进展.     

CT的形态学征象联合影像组学特征诊断模型在肺结节诊断中的应用

目的 结合肺结节CT形态学征像与影像组学特征建立联合诊断模型,期望提高孤立性肺结节的诊断准确性.方法 选取2012年2月～2019年1月206例肺实性结节(直径≤2cm)的CT图像.由两名医师独立进行形态学征像评估,并诊断结节良恶性,使用PyRadiomics开源软件提取、筛选定量影像组学特征,采用逻辑回归建立影像组学...     

基于乳腺X线图像纹理特征建立机器学习模型在鉴别良恶性乳腺肿块中的价值

目的:探讨基于乳腺X线图像的纹理分析建立机器学习模型在鉴别乳腺肿块良恶性中的价值.方法:回顾性搜集经病理证实的124个乳腺良性肿块和139个乳腺恶性肿块的乳腺X线图像.并按照7﹕3的比例划将所有病灶随即分为训练集和验证集.使用MaZda软件,在X线图像上于乳腺病灶内手动勾画ROI,提取6类共133个纹理特征,经降维处理...     

Limited value of shape, margin and CT density in the discrimination between benign and malignant screen detected solid pulmonary nodules of the NELSON trial

Purpose

To evaluate prospectively the value of size, shape, margin and density in discriminating between benign and malignant CT screen detected solid non-calcified pulmonary nodules.

Material and methods

This study was institutional review board approved. For this study 405 participants of the NELSON lung cancer screening trial with 469 indeterminate or potentially malignant solid pulmonary nodules (>50 mm³) were selected. The nodules were classified based on size, shape (round, polygonal, irregular) and margin (smooth, lobulated, spiculated). Mean nodule density and nodule volume were automatically generated by software. Analyses were performed by univariate and multivariate logistic regression. Results were presented as likelihood ratios (LR) with 95% confidence intervals (CI). Receiver operating characteristic analysis was performed for mean density as predictor for lung cancer.

Results

Of the 469 nodules, 387 (83%) were between 50 and 500 mm³, 82 (17%) >500 mm³, 59 (13%) malignant, 410 (87%) benign. The median size of the nodules was 103 mm³ (range 50–5486 mm³). In multivariate analysis lobulated nodules had LR of 11 compared to smooth; spiculated nodules a LR of 7 compared to smooth; irregular nodules a LR of 6 compared to round and polygonal; volume a LR of 3. The mean nodule CT density did not predict the presence of lung cancer (AUC 0.37, 95% CI 0.32–0.43).

Conclusion

In solid non-calcified nodules larger than 50 mm³, size and to a lesser extent a lobulated or spiculated margin and irregular shape increased the likelihood that a nodule was malignant. Nodule density had no discriminative power.     

腺癌性孤立性肺结节的^18F—FDG PET／CT表现

目的探讨腺癌性孤立性肺结节（ASPN）的^18F—FDG PET／CT显像特点。方法回顾分析35例ASPN的^18F-FDG PET／CT显像形态学和代谢特点,计算SUVmax,以公式[（延迟显像SUVmax-早期显像SUVmax）／早期SUVmax×100％]计算△SUVmax。以SPSS11．5软件对数据分别行t检验、方差分析和Fisher确切概率法检验。结果（1）42．86％（15／35）ASPN呈典型的癌性肺结节表现（结节状FDG摄取增高）,另有57．14％（20／35）ASPN FDG摄取呈片状、云雾状、肉眼无法辨认;结节状、云雾状、片状、肉眼无法辨认ASPN的SUVmax大小顺序递减,不同FDG摄取形态的ASPN早期和晚期SUVmax差异均有统计学意义,F=30．696和24．758,P均〈0．001。（2）54．29％（19／35）ASPN SUVmax≥2．5,45．71％（16／35）ASPN SUVmax〈2．5。（3）68．57％（24／35）ASPN呈实性密度结节,31．43％（11／35）ASPN呈“磨玻璃”密度结节;早期SUVmax分别为4．54±2．69、1．30±0．87,t=-5．234,P〈0．001。（4）延迟显像ASPN的SUVmax为422±3．52,高于早期显像的3．49±2．72（t=-4021,P〈0．1301）;延迟显像SUVmax是否增高与早期显像SUVmax的高低相关：94．74％（18／19）SUVmax≥2．5ASPN的△SUVmax为正值,仅56．25％（9／16）SUVmax〈2．5ASPN的△SUVmax为正值,P=0．013。（5）高分化ASPN SUVmax为1．70±1．51,低于中低分化ASPN的4．91±2．69,t=-3．951,P〈0．001,且△SUVmax〉0的比例（10／17）也低于中低分化ASPN（13／14）,P=0．045。结论ASPNFDG摄取形态、代谢活性差异大,SUVmax〈2．5ASPN比例较高,△SUVmax对这类结节良恶性的鉴别有一定帮助。     

ADC值联合基于T2WI脂肪抑制图像的纹理特征预测软组织良恶性肿瘤的价值

目的探讨DWI的ADC值联合纹理特征鉴别良恶性软组织肿瘤的价值。方法回顾性分析中国科学技术大学附属第一医院西区经病理证实的94例软组织肿瘤(恶性44例,良性50例)MRI及DWI图像。在GE ADW4.6工作站测量肿块的实性成分ADC值。在T₂WI脂肪抑制图像上的肿瘤最大层面手动勾画ROI并提取纹理特征;采用独立样本t检验对良恶性软组织肿瘤的ADC值及纹理参数进行统计学分析,并多因素logistic回归分析建模,计算诊断效能。结果良恶性软组织肿瘤的ADC值分别为(1.6±0.3)×10^-3 mm²/s、(1.2±0.5)×10^-3 mm²/s,差异有统计学意义(t=-5.382,P<0.05),以1.28×10^-3 mm²/s为诊断良恶性软组织肿瘤临界值,AUC为0.783,灵敏度为92.00%,特异度为65.91%。纹理特征中直方图特征(frequency size、skewness),灰度共生矩阵特征(Inertia_All Direction_offset7、Inverse Difference Moment_angle0_offset1、Inverse Difference Moment_angle0_offset7)及Haralick特征(Haralick Correlation_All Direction_offset4_SD)鉴别良恶性软组织肿瘤的曲线下面积分别为AUC 0.825、0.739、0.826、0.816、0.820、0.783。多因素logistic回归分析最佳预测模型鉴别良恶性软组织肿瘤的曲线下面积、灵敏度、特异度分别为0.930、88.00%、86.36%。结论ADC值联合纹理特征对术前预测软组织良恶性肿瘤有较高的应用价值。