高分辨率CT定性、定量技术评估慢性阻塞性肺疾病的临床应用价值
发布时间：2018-07-25 19:08
【摘要】：第一部分高分辨率CT定量技术评估慢性阻塞性肺疾病的临床应用价值目的:利用高分辨率 CT(High-resolution computed tomography,HRCT)容积定量技术自动测量每个肺叶的低密度区容积百分比(low attenuation areas volume percentage,LA A%),所得定量参数与肺功能(pulmonary function test,PFT)各项参数进行相关性研究,从而综合评估慢性阻塞性肺疾病(Chronic obstructive pulmonary disease,COPD)患者肺功能损伤的严重程度,对临床制定诊疗方案提供有价值的影像支持。材料和方法:收集2015年12月至2016年12月期间经临床及肺功能检查确诊的COPD患者83例(所有患者均于7天内完成HRCT检查及肺功能检查)。所有病例都采用GE 64排Lightspeed VCT于深吸气末进行扫描,把HRCT原始数据传到后处理工作站GEadw4.6,系统设定吸气末CT阈值-950HU为肺气肿区,运用CT后处理Parenchyma analysis软件自动得出以下参数:包括左肺下叶LAA%、左肺上叶LAA%、右肺上叶LAA%、右肺下叶LAA%、右肺中叶LAA%、双肺LAA%、右肺 LAA%、左肺 LAA%,总肺气肿体积(the total emphysema volume,TEV),总肺体积(Total Lung Volume,TLV)。本文主要采用的肺功能参数如下:第一秒用力呼气容积(forced expiratory volume in one second,FEV1)、第一秒用力呼气容积实测值和预计值的比值(FEV1%预测值)、用力肺活量(forced vital capacity,FVC)、FEV1与FVC的比值(FEV1/FVC)、呼气流量峰值(peak expiratory flow,PEF)、用力呼出 25%肺活量时呼气流量(forced expiratory flow at 25%of FVC exhaled,FEF25)、用力呼出 50%肺活量时呼气流量(forced expiratory flow at 50%of FVC exhaled,FEF50)、用力呼出 75%肺活量时呼气流量(forced expiratory flow at 75%of FVC exhaled,FEF75)、肺一氧化碳弥散量(diffusion capacity of carbon monoxide in the lung,DLCO)实测值占预计值百分比(DLCO%)、残总比即残气容积与肺总量的比值(RV/TLC)。本研究把肺功能参数和之前提到的CT定量化参数进行Spearman相关性研究。COPD不同分级间CT定量化参数的差异性分析应用单因素方差分析法(One-Way ANOVA),组间比较采用LSD检验。采用Kruskal-Wallis检验对COPD不同分级间FEV1/FVC、FEV1%预测值、BMI等进行比较分析。本研究统计学用SPSS20.0软件分析,p0.05认为差异有统计学意义,p0.01认为差异有显著统计学意义。结果:83例患者,年龄47～85岁,平均年龄66岁,性别均为男性,BMI指数13.3～28.1,吸烟者71例,不吸烟者12例,根据2017年COPD分级标准,该组病例中:GOLD1级8 例;GOLD2 级 33 例;GOLD3 级 27 例;GOLD 4 级 15 例。左肺下叶LAA%、左肺LAA%、右肺下叶LAA%、右肺LAA%、总肺LAA%与FEV1/FVC、FEV1、FEV1%预计值、FEF25、FEF50均有相关性,两肺下叶与PEF、FEF75也有相关性,两肺下叶LAA%与肺功能气流受限参数(FEV1、FEV1%预测值、FEV1/FVC、PEF、FEF50)呈显著相关。FEV1/FVC与TEV显著相关(r=0.759**,P0.001)。TLV 与 TLC 显著相关(r=-0.355,**P=0.001)。DLCO%与两肺上叶LAA%有相关性(r=-0.473,P=0.026)。两肺下叶LAA%在GOLD1级与GOLD3级间差异有统计学意义,左肺下叶LAA%在GOLD 1和GOLD4级之间差异有统计学意义。除了右肺中叶LAA%外,其他各肺叶LAA%在GOLD2级与GOLD3级间差异均有统计学意义。总肺LAA%、左肺下叶LAA%、右肺上LAA%在GOLD2级与GOLD4级间差异有统计学意义。FEV1/FVC、FEV1%预测值在GOLD各组间差异有统计学意义。结论:两肺下叶LAA%与肺功能气流受限参数(FEV1、FEV1%预测值、FEV1/FVC、FEF50、PEF)均具有显著相关性,与RV/TLC、DLCO%无明显相关性。两肺上叶与DLCO%有相关性。因此提示不同肺叶LAA%可评估COPD病人肺功能的损伤部位和损伤严重情况,从而为临床制定不同治疗方案提供相关依据。第二部分高分辨率CT图像利用计算机后处理技术定性肺气肿亚型目的:通过计算机后处理技术为临床提供一种可重复、无偏倚、更精准的肺气肿亚型的自动识别方法,以深入认识肺气肿不同形态及不同病变程度,从而为临床医生在制定COPD患者的个性化治疗方案上提供了一种新的思路。材料和方法:采用GE64排LightspeedVCT于深吸气末进行全肺扫描,采集病例组(肺气肿典型病例)6例(每种亚型为主型各2例)及正常组6例,并将重建后数据导入ITK-SNAP软件,然后利用不同的颜色代表不同的肺气肿亚型及正常肺组织进而进行图像标注:正常的肺组织(Normaltissue,NT)使用红色标注,小叶中心型肺气肿(Centrilobular Emphysema,CLE)使用绿色标注,全小叶型肺气肿(Panlobular Emphysema,PLE)利用蓝色标注,间隔旁型肺气肿(Paraseptal Emphysema,PSE)利用黄色进行标注。在全部标注结束的病例组中每个病例随机取1000个不重叠的异常ROI(Region of interest),每个异常ROI都有一个对应的标签(label):包括CLE、PLE、PSE,采集总共6000个异常ROI。病例组中另外再取1000个不重叠的正常肺组织(Normaltissue,NT)ROI。正常组中随机取1000个不重叠的正常组织(NT)的ROI。病例组中从6000个异常ROI中每类肺气肿亚型(CLE,PSE,PLE)随机选1000个ROI,共3000个,加上正常组的1000个ROI作为训练样本(training samples),从而训练出可以自动识别肺气肿亚型的分类器。然后在病例组中剩下的3000个异常ROI及1000个正常组织ROI中,每次每类肺气肿亚型(CLE,PSE,PLE)及正常肺组织(NT)随机选200个ROI作为测试样本(test samples)。利用计算机后处理 Intensity(INT)、Rotation invariant Local Binary Patterns(RILBPs)、INT+RILBPs三种方法分别自动识别各种肺气肿亚型,测试结果与人工标注结果进行对比并计算分类精度,以上测试实验重复5次,以5次分类精度的平均值作为最终分类精度。结果:利用计算机后处理INT、RILBPs、INT+RILBPs三种方法分别自动识别肺气肿亚型的分类精度如下:只提取CLE、PLE、PSE三种肺气肿亚型进行测试,测得精度 INT 法为 88.28%,RILBPs 法为 86.46%,INT+RILBPs 法 94.62%;三种方法中均以小叶中心型肺气肿(CLE)的分类精度最高。加入正常组织(NT)后分类精度下降,INT 法为 72.09%,RILBPs 法为 67.34%,INT+RILBPs 法 84.29%。该组中除了 INT法中以全小叶型肺气肿(PLE)分类精度最高外,其他两种方法中均为CLE分类精度最高。由此可见INT+RILBPs这一方法分类精度最高,在肺气肿亚型中以CLE的分类精度较高。结论:利用计算机后处理INT、RILBPs、INT+RILBPs三种方法对比分析计算机自动识别肺气肿亚型分类精度,INT+RILBPs法分类精度明显高于其他两种方法;INT+RILBPs法仅提取CLE、PLE、PSE三种肺气肿亚型测试结果明显高于加入正常肺组织后的分类精度。INT+RILBPs这一方法在自动识别肺气肿亚型上可以为临床提供诊断依据。
[Abstract]:The first part of the high resolution CT quantitative technique to evaluate the clinical application of chronic obstructive pulmonary disease: using the high resolution CT (High-resolution computed tomography, HRCT) volume quantitative technique to automatically measure the volume percentage of the low density region of each lobe (low attenuation areas volume percentage, LA), and the quantitative parameters Correlation study with pulmonary function test (PFT) parameters to assess the severity of pulmonary function damage in patients with chronic obstructive pulmonary disease (Chronic, obstructive pulmonary disease, COPD), and provide valuable imaging support for clinical diagnosis and treatment programs. Materials and methods: from December 2015 to 2016 In December, 83 patients with COPD were diagnosed with clinical and pulmonary function tests (all patients completed HRCT examination and pulmonary function examination within 7 days). All cases were scanned by GE 64 row Lightspeed VCT at the deep inhalation terminal. HRCT original data were transmitted to the post processing workstation GEadw4.6, and the system set the terminal CT threshold -950HU to emphysema. The following parameters were obtained by using CT after processing Parenchyma analysis software, including left lower lobe LAA%, left upper lobe LAA%, right upper lobe LAA%, right lower lobe LAA%, right pulmonary middle lobe LAA%, double lung LAA%, right lung LAA%, left lung LAA%, total emphysema volume, total lung volume. The parameters of the lung function are as follows: the first second forced expiratory volume (forced expiratory volume in one second, FEV1), the ratio of the measured value of the forced expiratory volume to the predicted value (FEV1% prediction), the forced vital capacity (forced vital capacity, FVC), the ratio of the expiratory volume, the peak expiratory flow, and the force of the expiratory flow, and the peak expiratory flow peak. Expiratory flow (forced expiratory flow at 25%of FVC exhaled, FEF25), expiratory expiratory flow of 50% vital capacity (forced expiratory flow at), expiratory exhalation of 75% vital capacity, lung carbon monoxide dispersion Usion capacity of carbon monoxide in the lung, DLCO) measured value as a percentage of expected value (DLCO%), the ratio of residual volume to total volume of lung (RV/TLC). Factor variance analysis (One-Way ANOVA), LSD test was used among groups. Kruskal-Wallis test was used to compare FEV1/FVC, FEV1% prediction value, BMI and so on. This study was analyzed by SPSS20.0 software, P0.05 thought the difference was statistically significant, P0.01 thought the difference was significant statistically. Results: 83 cases People aged 47~85, age 47~85, average age 66 years, sex are male, BMI index 13.3 ~ 28.1, smokers 71 cases, non smokers 12 cases, according to the 2017 COPD grading standard, the group cases: GOLD1 class 8 cases, GOLD2 class 33 cases, GOLD3 class 27 cases, GOLD 4 level 15 cases, left lung LAA%, right lower lobe LAA%, right lung LAA%, LAA% lung and FEV1/F. VC, FEV1, FEV1% predicted values, FEF25 and FEF50 were correlated. Two the lower lobe of the lung was also associated with PEF and FEF75. Two lower lobe LAA% and lung function limited parameters (FEV1, FEV1% predictive value, FEV1/FVC, PEF) were significantly correlated with the two lung The leaf LAA% was correlated (r=-0.473, P=0.026). Two the difference between LAA% in the lower lobe of the lung in GOLD1 and GOLD3 was statistically significant. The difference between the LAA% in the lower lobe of the left lung was statistically significant between GOLD 1 and GOLD4. Except for the middle lobe LAA% of the right lung, the difference between the other lung lobes was statistically significant between the GOLD2 level and the grade. There was a significant difference between the GOLD2 and GOLD4 levels on the right lung LAA%, and the difference between the FEV1% prediction values in the GOLD group was statistically significant. Conclusion: two the LAA% in the lower lobe of the lung has a significant correlation with the limited parameters of the pulmonary function (FEV1, FEV1% prediction, FEV1/FVC, FEF50, PEF), and there is no significant correlation with the lung function. Two the upper lobe of the lung is not significantly correlated. It is pertinence. Therefore, it is suggested that different pulmonary lobectomy LAA% can evaluate the damage location and serious damage of lung function in COPD patients, and provide the relevant basis for the clinical formulation of different treatment schemes. The second part of high resolution CT image is used by computer post-processing technique to determine the subtype of emphysema. An automatic identification of reduplicated, unbiased, and more accurate emphysema subtypes to deeply understand the different morphologic and pathological changes of emphysema, thus providing a new idea for clinicians to develop individualized treatment schemes for COPD patients. Materials and methods: GE64 row LightspeedVCT at the end of deep inhalation for the whole lung 6 cases (typical cases of emphysema) (2 cases of emphysema) and 6 cases of normal group were collected, and the reconstructed data were introduced into ITK-SNAP software, and then different colors representing different emphysema subtypes and normal lung tissues were used for image tagging: normal lung tissue (Normaltissue, NT) used red label, lobule. Centrilobular Emphysema (CLE) used green tagging, full lobular emphysema (Panlobular Emphysema, PLE) using blue tagging, paranatal emphysema (Paraseptal Emphysema, PSE) using yellow for tagging. 1000 unoverlapped abnormal ROI (Region of) at the end of all the cases were randomly selected. Interest), each abnormal ROI has a corresponding label (label): including CLE, PLE, PSE, and collect 1000 unoverlapped normal lung tissues (Normaltissue, NT) ROI. in a total of 6000 non overlapping normal groups (Normaltissue, NT) ROI., and take 1000 non overlapping normal tissues (NT) in the ROI. case group from 6000 abnormal lungs in each type of emphysema. The subtype (CLE, PSE, PLE) randomly selected 1000 ROI, a total of 3000, and 1000 ROI of the normal group as a training sample (training samples), and then trained the classifier that could automatically identify the emphysema subtype. Then the 3000 remaining ROI and 1000 normal tissue ROI were left in the case group, and each type of emphysema subtype (CLE, PSE, PLE) and positive was in each class. NT (NT) was randomly selected as a test sample (test samples). Using a computer to process Intensity (INT), Rotation invariant Local Binary Patterns (RILBPs), three methods were used to automatically identify the various emphysema subtypes. The test results were compared with the artificial annotation results and the classification accuracy was calculated. The above tests were tested. The experiment was repeated 5 times and the average value of 5 classification accuracy was used as the final classification accuracy. Results: the classification accuracy of the subtype of emphysema was automatically identified by three methods of computer post-processing INT, RILBPs and INT+RILBPs, respectively: only CLE, PLE, PSE, three emphysema subtypes were tested, the accuracy INT method was 88.28%, RILBPs method was 86.46%, I. NT+RILBPs method 94.62%; among the three methods, the classification accuracy of the lobular central emphysema (CLE) was the highest. The classification accuracy decreased after adding the normal tissue (NT), the INT method was 72.09%, the RILBPs method was 67.34%, and the INT+RILBPs method 84.29%. had the highest classification accuracy except for the total lobular emphysema (PLE) in the INT method, and the other two methods were all. The classification precision of CLE is the highest. This shows that INT+RILBPs has the highest classification precision, and the classification precision of CLE is higher in emphysema subtype. Conclusion: computer after processing INT, RILBPs, INT+RILBPs three methods to compare and analyze the classification accuracy of emphysema subtype by computer, the precision of INT+RILBPs method is obviously higher than that of other two. The INT+RILBPs method only extracts CLE, PLE, and PSE three emphysema subtypes, which is significantly higher than the classification precision.INT+RILBPs after adding normal lung tissue. This method can provide diagnostic basis for the automatic identification of emphysema subtype.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R563.9;R816.41
本文编号：2144793
[Abstract]:The first part of the high resolution CT quantitative technique to evaluate the clinical application of chronic obstructive pulmonary disease: using the high resolution CT (High-resolution computed tomography, HRCT) volume quantitative technique to automatically measure the volume percentage of the low density region of each lobe (low attenuation areas volume percentage, LA), and the quantitative parameters Correlation study with pulmonary function test (PFT) parameters to assess the severity of pulmonary function damage in patients with chronic obstructive pulmonary disease (Chronic, obstructive pulmonary disease, COPD), and provide valuable imaging support for clinical diagnosis and treatment programs. Materials and methods: from December 2015 to 2016 In December, 83 patients with COPD were diagnosed with clinical and pulmonary function tests (all patients completed HRCT examination and pulmonary function examination within 7 days). All cases were scanned by GE 64 row Lightspeed VCT at the deep inhalation terminal. HRCT original data were transmitted to the post processing workstation GEadw4.6, and the system set the terminal CT threshold -950HU to emphysema. The following parameters were obtained by using CT after processing Parenchyma analysis software, including left lower lobe LAA%, left upper lobe LAA%, right upper lobe LAA%, right lower lobe LAA%, right pulmonary middle lobe LAA%, double lung LAA%, right lung LAA%, left lung LAA%, total emphysema volume, total lung volume. The parameters of the lung function are as follows: the first second forced expiratory volume (forced expiratory volume in one second, FEV1), the ratio of the measured value of the forced expiratory volume to the predicted value (FEV1% prediction), the forced vital capacity (forced vital capacity, FVC), the ratio of the expiratory volume, the peak expiratory flow, and the force of the expiratory flow, and the peak expiratory flow peak. Expiratory flow (forced expiratory flow at 25%of FVC exhaled, FEF25), expiratory expiratory flow of 50% vital capacity (forced expiratory flow at), expiratory exhalation of 75% vital capacity, lung carbon monoxide dispersion Usion capacity of carbon monoxide in the lung, DLCO) measured value as a percentage of expected value (DLCO%), the ratio of residual volume to total volume of lung (RV/TLC). Factor variance analysis (One-Way ANOVA), LSD test was used among groups. Kruskal-Wallis test was used to compare FEV1/FVC, FEV1% prediction value, BMI and so on. This study was analyzed by SPSS20.0 software, P0.05 thought the difference was statistically significant, P0.01 thought the difference was significant statistically. Results: 83 cases People aged 47~85, age 47~85, average age 66 years, sex are male, BMI index 13.3 ~ 28.1, smokers 71 cases, non smokers 12 cases, according to the 2017 COPD grading standard, the group cases: GOLD1 class 8 cases, GOLD2 class 33 cases, GOLD3 class 27 cases, GOLD 4 level 15 cases, left lung LAA%, right lower lobe LAA%, right lung LAA%, LAA% lung and FEV1/F. VC, FEV1, FEV1% predicted values, FEF25 and FEF50 were correlated. Two the lower lobe of the lung was also associated with PEF and FEF75. Two lower lobe LAA% and lung function limited parameters (FEV1, FEV1% predictive value, FEV1/FVC, PEF) were significantly correlated with the two lung The leaf LAA% was correlated (r=-0.473, P=0.026). Two the difference between LAA% in the lower lobe of the lung in GOLD1 and GOLD3 was statistically significant. The difference between the LAA% in the lower lobe of the left lung was statistically significant between GOLD 1 and GOLD4. Except for the middle lobe LAA% of the right lung, the difference between the other lung lobes was statistically significant between the GOLD2 level and the grade. There was a significant difference between the GOLD2 and GOLD4 levels on the right lung LAA%, and the difference between the FEV1% prediction values in the GOLD group was statistically significant. Conclusion: two the LAA% in the lower lobe of the lung has a significant correlation with the limited parameters of the pulmonary function (FEV1, FEV1% prediction, FEV1/FVC, FEF50, PEF), and there is no significant correlation with the lung function. Two the upper lobe of the lung is not significantly correlated. It is pertinence. Therefore, it is suggested that different pulmonary lobectomy LAA% can evaluate the damage location and serious damage of lung function in COPD patients, and provide the relevant basis for the clinical formulation of different treatment schemes. The second part of high resolution CT image is used by computer post-processing technique to determine the subtype of emphysema. An automatic identification of reduplicated, unbiased, and more accurate emphysema subtypes to deeply understand the different morphologic and pathological changes of emphysema, thus providing a new idea for clinicians to develop individualized treatment schemes for COPD patients. Materials and methods: GE64 row LightspeedVCT at the end of deep inhalation for the whole lung 6 cases (typical cases of emphysema) (2 cases of emphysema) and 6 cases of normal group were collected, and the reconstructed data were introduced into ITK-SNAP software, and then different colors representing different emphysema subtypes and normal lung tissues were used for image tagging: normal lung tissue (Normaltissue, NT) used red label, lobule. Centrilobular Emphysema (CLE) used green tagging, full lobular emphysema (Panlobular Emphysema, PLE) using blue tagging, paranatal emphysema (Paraseptal Emphysema, PSE) using yellow for tagging. 1000 unoverlapped abnormal ROI (Region of) at the end of all the cases were randomly selected. Interest), each abnormal ROI has a corresponding label (label): including CLE, PLE, PSE, and collect 1000 unoverlapped normal lung tissues (Normaltissue, NT) ROI. in a total of 6000 non overlapping normal groups (Normaltissue, NT) ROI., and take 1000 non overlapping normal tissues (NT) in the ROI. case group from 6000 abnormal lungs in each type of emphysema. The subtype (CLE, PSE, PLE) randomly selected 1000 ROI, a total of 3000, and 1000 ROI of the normal group as a training sample (training samples), and then trained the classifier that could automatically identify the emphysema subtype. Then the 3000 remaining ROI and 1000 normal tissue ROI were left in the case group, and each type of emphysema subtype (CLE, PSE, PLE) and positive was in each class. NT (NT) was randomly selected as a test sample (test samples). Using a computer to process Intensity (INT), Rotation invariant Local Binary Patterns (RILBPs), three methods were used to automatically identify the various emphysema subtypes. The test results were compared with the artificial annotation results and the classification accuracy was calculated. The above tests were tested. The experiment was repeated 5 times and the average value of 5 classification accuracy was used as the final classification accuracy. Results: the classification accuracy of the subtype of emphysema was automatically identified by three methods of computer post-processing INT, RILBPs and INT+RILBPs, respectively: only CLE, PLE, PSE, three emphysema subtypes were tested, the accuracy INT method was 88.28%, RILBPs method was 86.46%, I. NT+RILBPs method 94.62%; among the three methods, the classification accuracy of the lobular central emphysema (CLE) was the highest. The classification accuracy decreased after adding the normal tissue (NT), the INT method was 72.09%, the RILBPs method was 67.34%, and the INT+RILBPs method 84.29%. had the highest classification accuracy except for the total lobular emphysema (PLE) in the INT method, and the other two methods were all. The classification precision of CLE is the highest. This shows that INT+RILBPs has the highest classification precision, and the classification precision of CLE is higher in emphysema subtype. Conclusion: computer after processing INT, RILBPs, INT+RILBPs three methods to compare and analyze the classification accuracy of emphysema subtype by computer, the precision of INT+RILBPs method is obviously higher than that of other two. The INT+RILBPs method only extracts CLE, PLE, and PSE three emphysema subtypes, which is significantly higher than the classification precision.INT+RILBPs after adding normal lung tissue. This method can provide diagnostic basis for the automatic identification of emphysema subtype.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R563.9;R816.41
【相似文献】
相关期刊论文 前10条
1 于静,文姝,唐立;人体正常菌群微生物定量技术的研究进展[J];中国微生态学杂志;2004年01期
2 赵军宁;温志坚;;形态定量技术新进展与形态药理学[J];四川生理科学杂志;1991年Z1期
3 王建文;声学定量技术的应用[J];中国医学影像技术;1997年02期
4 杨丽型,夏稻子,礼广森;声学密度定量技术检测不同年龄健康人颈动脉内膜[J];中国超声医学杂志;2004年06期
5 郭方明,宫国建,宋秀媛,范民忠,周建民;声学密度定量及声学定量技术评价心肌梗塞患者的左心功能[J];中国超声诊断杂志;2003年11期
6 胡智安;韩鄂辉;张频;朱剑芳;;声触诊组织定量技术在甲状腺弥漫性病变诊断中的应用[J];南昌大学学报(医学版);2011年12期
7 王现,林其珊,陈伟达;声学定量技术评价小儿左心功能(附69例报告)[J];心肺血管病杂志;1999年02期
8 杨丽型,夏稻子,礼广森;声学密度定量技术对颈动脉斑块致脑梗死的危险性的评价[J];中国超声医学杂志;2005年02期
9 林子梅;房世保;王建红;孙咏梅;;声触诊组织定量技术对甲状腺弥漫性疾病的诊断价值[J];中华临床医师杂志(电子版);2013年01期
10 杨国春;宫建;王学梅;王坤;王云忠;姜镔;;组织结构声学定量技术评估正常成年人肝脏组织声学结构的初步研究[J];中国医学工程;2013年07期
相关会议论文 前10条
1 郭薇;陈斌;卢荔红;浦晓东;;实时三维超声心动图定量局部心肌重量的研究[A];第九届全国超声心动图学术会议论文集[C];2007年
2 周琦;李容;姜珏;尚旭;王华;刘婷;冯晓蕾;;声学组织定量技术定量诊断肝脏弥漫性病变的初步探讨[A];中国超声医学工程学会第九届全国腹部超声医学学术会议论文汇编[C];2012年
3 何婉媛;金峗杰;周盛;季正标;马娇娇;王文平;;声触诊组织定量技术在移植肾急性排斥诊断中的临床价值[A];中华医学会第十三次全国超声医学学术会议论文汇编[C];2013年
4 张志民;赵洪岩;朱来宽;;采用非标记定量技术对变形链球菌耐氟菌株的差异蛋白质组学研究[A];全国第八次牙体牙髓病学学术会议论文汇编[C];2011年
5 张芸;郭建;何俐;吴杞柱;张体江;龚启勇;;弥散张量成像定量技术在急性缺血性脑卒中的临床应用[A];第十一届全国神经病学学术会议论文汇编[C];2008年
6 郑孝志;吴菁;季平;茅红卫;;模拟声触诊组织定量技术评价健康男性阴茎勃起硬度[A];中华医学会第十三次全国超声医学学术会议论文汇编[C];2013年
7 曹纬倩;张伟;黄江铭;杨們原;;基于质谱的N-糖链双同位素标记定量技术及其在定量糖组学中的应用[A];中国化学会第29届学术年会摘要集——第38分会：质谱分析[C];2014年
8 刘莹莹;谢明星;项飞翔;张艳容;李薇玢;陈玉媛;;彩色多普勒感兴趣区定量技术评价原发性高血压患者肾脏末梢血流灌注[A];庆祝中国超声诊断50年暨第十届全国超声医学学术会议论文汇编[C];2008年
9 张丽明;郑冬珠;蒋晓马;钱晓刚;;C-13NMR定量技术测定HM-PAO的非对映立体异构体的异构纯度[A];第七届全国波谱学学术会议论文摘要集[C];1992年
10 张梅;张运;张园园;高月花;李秀昌;戴晓华;;超声声学密度定量技术对兔动脉粥样硬化消退治疗的评价[A];2003年全国医学影像技术学术会议论文汇编[C];2003年
相关博士学位论文 前2条
1 赵洪岩;采用非标记定量技术对变形链球菌耐氟菌株的差异蛋白质组学研究[D];吉林大学;2011年
2 桑志红;建立SILAC定量技术并发现CD40受体激活后招募的复合物新组分[D];中国人民解放军军事医学科学院;2011年
相关硕士学位论文 前5条
1 章生龙;声触诊组织定量技术在桥本甲状腺炎诊断中的应用价值[D];安徽医科大学;2015年
2 王丹阳;声触诊组织定量技术在正常人和2型糖尿病人胰腺中的临床应用[D];河北医科大学;2017年
3 凌小莉;高分辨率CT定性、定量技术评估慢性阻塞性肺疾病的临床应用价值[D];浙江大学;2017年
4 文曙;荧光PCR定量技术在遗传病诊断中的应用[D];中南大学;2003年
5 杨丽型;声学密度定量技术检测颈动脉粥样硬化的临床研究[D];大连医科大学;2005年
，本文编号：2144793
本文链接：https://www.wllwen.com/yixuelunwen/huxijib/2144793.html
最近更新
教材专著
