CT纹理分析与肝癌病理分化程度的相关性研究
|
摘要
目的研究增强CT纹理分析与肝细胞癌(HCC)病理分化程度的相关性。方法回顾性分析经病理证实的307例HCC患者临床及影像资料,提取患者增强CT动脉期及门静脉期图像中肿瘤部位形态、灰度直方图、灰度共生矩阵、灰度游程矩阵等纹理特征。根据训练集动脉期、门静脉期纹理特征及两者间差异,分别建立线性支持向量机(SVM)预测模型。将所得模型应用于测试集,采用ROC曲线分析评估各模型在测试集中预测HCC高分化或中低分化的效能。结果根据动脉期、门静脉期纹理特征及两者间差异所建立SVM预测模型的ROC曲线AUC分别为0. 75、0. 59、0. 57,动脉期SVM预测模型AUC最高,其精确性、敏感性、特异性分别为78%、81%、66%。动脉期SVM预测模型对于肝癌病理分化程度的预测效果最佳,明显优于门静脉期及两组间差异。结论根据动脉期CT纹理特征所获得的SVM预测模型能有效预测HCC病理分化程度。
Objective To study the correlation of texture analysis based on contrast-enhanced CT and the pathological differentiation degrees in hepatocellular carcinoma. Methods 307 patients with clinicopathologically confirmed HCC were retrospectively analyzed. Texture features,such as formfactor,histogram features,GLCM features and GLRLM features,was extracted from the arterial-phase and portal-phase CT images of HCC. Linear SVM predictive models were developed in training set according to texture features of arterial-phase,porta-phase and the difference between them respectively. Receiver operating characteristic curve was used to determine the performance of each predicting model in distinguishing between well and moderate/poorly differentiated HCC in testing set. Results The predicting model yielded an AUC of 0. 75 for arterial-phase,0. 59 for portal-phase,and 0. 57 for the difference. The highest AUC values fell in the arterial-phase SVM predicting model,with accuracy,sensitivity,and specificity of 78%,81%,and 66%,respectively. The arterial-phase SVM predicting model had the best predictive performance on the differentiation of HCC pathological grade,and it is obviously better than the portal-phase in assessing the difference between them. Conclusion The SVM predictive model developed by arterial-phase CT texture feature can effectively predict pathological grade of HCC.
Objective To study the correlation of texture analysis based on contrast-enhanced CT and the pathological differentiation degrees in hepatocellular carcinoma. Methods 307 patients with clinicopathologically confirmed HCC were retrospectively analyzed. Texture features,such as formfactor,histogram features,GLCM features and GLRLM features,was extracted from the arterial-phase and portal-phase CT images of HCC. Linear SVM predictive models were developed in training set according to texture features of arterial-phase,porta-phase and the difference between them respectively. Receiver operating characteristic curve was used to determine the performance of each predicting model in distinguishing between well and moderate/poorly differentiated HCC in testing set. Results The predicting model yielded an AUC of 0. 75 for arterial-phase,0. 59 for portal-phase,and 0. 57 for the difference. The highest AUC values fell in the arterial-phase SVM predicting model,with accuracy,sensitivity,and specificity of 78%,81%,and 66%,respectively. The arterial-phase SVM predicting model had the best predictive performance on the differentiation of HCC pathological grade,and it is obviously better than the portal-phase in assessing the difference between them. Conclusion The SVM predictive model developed by arterial-phase CT texture feature can effectively predict pathological grade of HCC.
引文
1 Torre LA,Siegel RL,Ward EM,et al. Global Cancer Incidence and Mortality Rates and Trends-An Update[J]. Cancer epidemiology,biomarkers&prevention:a publication of the American Association for Cancer Research,cosponsored by the American Society of Preventive Oncology,2016,25:16-27.
2 Zhang Y,Ren JS,Shi JF,et al. International trends in primary liver cancer incidence from 1973 to 2007[J]. BMC Cancer,15,1(2015-03-04),2015,15:94.
3 Skogen K,Schulz A,Dormagen JB,et al. Diagnostic performance of texture analysis on MRI in grading cerebral gliomas[J]. European Journal of Radiology,2016,85:824-829.
4 Teruel JR,Heldahl MG,Goa PE,et al. Dynamic contrast-enhanced MRI texture analysis for pretreatment prediction of clinical and pathological response to neoadjuvant chemotherapy in patients with locally advanced breast cancer[J]. NMR in Biomedicine,2014,27:887-896.
5 Bayanati H,Thornhill RE,Souza CA,et al. Quantitative CT texture and shape analysis:Can it differentiate benign and malignant mediastinal lymph nodes in patients with primary lung cancer?[J]. European Radiology,2015,25:480-487.
6 Han F,Wang H,Zhang G,et al. Texture Feature Analysis for Computer-Aided Diagnosis on Pulmonary Nodules[J]. Journal of Digital Imaging,2015,28:99-115.
7 Wibmer A,Hricak H,Gondo T,et al. Haralick texture analysis of prostate MRI:utility for differentiating non-cancerous prostate from prostate cancer and differentiating prostate cancers with different Gleason scores[J]. European Radiology,2015,25:2840-2850.
8 Sotiropoulos GC,Molmenti EP,Losch C,et al. Meta-analysis of tumor recurrence after liver transplantation for hepatocellular carcinoma based on 1,198 cases[J]. European Journal of Medical Research,2007,12:527-534.
9 Jing J,Yoen JH,Ho LK,et al. Nuclear Expression of Hepatitis B Virus X Protein Is Associated with Recurrence of Early-Stage Hepatocellular Carcinomas:Role of Viral Protein in Tumor Recurrence[J].Journal of Pathology&Translational Medicine,2016,50:181-189.
10 Zhou L,Rui JA,Ye DX,et al. Edmondson-Steiner grading increases the predictive efficiency of TNM staging for long-term survival of patients with hepatocellular carcinoma after curative resection[J].World Journal of Surgery,2008,32:1748-1756.
11 Huang S,Huang GQ,Zhu GQ,et al. Establishment and Validation of SSCLIP Scoring System to Estimate Survival in Hepatocellular Carcinoma Patients Who Received Curative Liver Resection[J]. Plo S One,2015,10:e0129000.
12 涂海斌,郭佳,林健玲,等. CEUS定量分析鉴别小肝癌分级[J].中国介入影像与治疗学,2017,14:543-547.
13 Tarhan NC,Hatipogˇlu T,Ercan E,et al. Correlation of dynamic multidetector CT findings with pathological grades of hepatocellular carcinoma[J]. Diagnostic&Interventional Radiology,2012,18:236;author reply
14 梁长华,梁盼,刘甲,等.原发性肝癌320排容积CT全肿瘤灌注参数值与微血管密度及其与病理分级的相关性研究[J].临床放射学杂志,2017,36:517-522.
15 Chen CY,Chen J,Xia CC,et al. T1mapping combined with GdEOB-DTPA-enhanced magnetic resonance imaging in predicting the pathologic grading of hepatocellular carcinoma[J]. Journal of Biological Regulators and Homeostatic Agents,2017,31:1029-1036.
16 Peng Z,Jiang M,Cai H,et al. Gd-EOB-DTPA-enhanced magnetic resonance imaging combined with T1mapping predicts the degree of differentiation in hepatocellular carcinoma[J]. BMC Cancer,2016,16:625.
17 Li X,Zhang K,Shi Y,et al. Correlations between the minimum and mean apparent diffusion coefficient values of hepatocellular carcinoma and tumor grade[J]. Journal of Magnetic Resonance Imaging,2016,44:1442-1447.
18 Bai DS,Zhang C,Chen P,et al. The prognostic correlation of AFP level at diagnosis with pathological grade,progression,and survival of patients with hepatocellular carcinoma[J]. Scientific Reports,2017,7:12870.
19 Marrero JA,Kulik LM,Sirlin C,et al. Diagnosis,Staging and Management of Hepatocellular Carcinoma:2018 Practice Guidance by the American Association for the Study of Liver Diseases[J]. Hepatology(Baltimore,Md),2018.
20 Liver EAFTSOT. EASL-EORTC clinical practice guidelines:management of hepatocellular carcinoma[J]. Journal of Hepatology,2012,56:908-943.
21 Ganeshan B,Miles KA. Quantifying tumour heterogeneity with CT[J]. Cancer Imaging:the Official Publication of the International Cancer Imaging Society,2013,13:140-149.
22 Ng F,Kozarski R,Ganeshan B,et al. Assessment of tumor heterogeneity by CT texture analysis:can the largest cross-sectional area be used as an alternative to whole tumor analysis?[J]. European Journal of Radiology,2013,82:342-348.
23 Raman SP,Schroeder JL,Huang P,et al. Preliminary data using computed tomography texture analysis for the classification of hypervascular liver lesions:generation of a predictive model on the basis of quantitative spatial frequency measurements-a work in progress[J].Journal of Computer Assisted Tomography,2015,39:383-395.
24 Li M,Fu S,Zhu Y,et al. Computed tomography texture analysis to facilitate therapeutic decision making in hepatocellular carcinoma[J]. Oncotarget,2016,7:13248-13259.
25 Simpson AL,Adams LB,Allen PJ,et al. Texture analysis of preoperative CT images for prediction of postoperative hepatic insufficiency:a preliminary study[J]. Journal of the American College of Surgeons,2015,220:339-346.
26 Chen S,Zhu Y,Liu Z,et al. Texture analysis of baseline multiphasic hepatic computed tomography images for the prognosis of single hepatocellular carcinoma after hepatectomy:A retrospective pilot study[J]. European Journal of Radiology,2017,90:198-204.
27 Bauer S,Nolte LP,Reyes M. Fully automatic segmentation of brain tumor images using support vector machine classification in combination with hierarchical conditional random field regularization[J]. International Conference on Medical Image Computing and ComputerAssisted Intervention,2011,14:354-361.
28 Huber MB,Nagarajan MB,Leinsinger G,et al. Performance of topological texture features to classify fibrotic interstitial lung disease patterns[J]. Medical Physics,2011,38:2035-2044.
2 Zhang Y,Ren JS,Shi JF,et al. International trends in primary liver cancer incidence from 1973 to 2007[J]. BMC Cancer,15,1(2015-03-04),2015,15:94.
3 Skogen K,Schulz A,Dormagen JB,et al. Diagnostic performance of texture analysis on MRI in grading cerebral gliomas[J]. European Journal of Radiology,2016,85:824-829.
4 Teruel JR,Heldahl MG,Goa PE,et al. Dynamic contrast-enhanced MRI texture analysis for pretreatment prediction of clinical and pathological response to neoadjuvant chemotherapy in patients with locally advanced breast cancer[J]. NMR in Biomedicine,2014,27:887-896.
5 Bayanati H,Thornhill RE,Souza CA,et al. Quantitative CT texture and shape analysis:Can it differentiate benign and malignant mediastinal lymph nodes in patients with primary lung cancer?[J]. European Radiology,2015,25:480-487.
6 Han F,Wang H,Zhang G,et al. Texture Feature Analysis for Computer-Aided Diagnosis on Pulmonary Nodules[J]. Journal of Digital Imaging,2015,28:99-115.
7 Wibmer A,Hricak H,Gondo T,et al. Haralick texture analysis of prostate MRI:utility for differentiating non-cancerous prostate from prostate cancer and differentiating prostate cancers with different Gleason scores[J]. European Radiology,2015,25:2840-2850.
8 Sotiropoulos GC,Molmenti EP,Losch C,et al. Meta-analysis of tumor recurrence after liver transplantation for hepatocellular carcinoma based on 1,198 cases[J]. European Journal of Medical Research,2007,12:527-534.
9 Jing J,Yoen JH,Ho LK,et al. Nuclear Expression of Hepatitis B Virus X Protein Is Associated with Recurrence of Early-Stage Hepatocellular Carcinomas:Role of Viral Protein in Tumor Recurrence[J].Journal of Pathology&Translational Medicine,2016,50:181-189.
10 Zhou L,Rui JA,Ye DX,et al. Edmondson-Steiner grading increases the predictive efficiency of TNM staging for long-term survival of patients with hepatocellular carcinoma after curative resection[J].World Journal of Surgery,2008,32:1748-1756.
11 Huang S,Huang GQ,Zhu GQ,et al. Establishment and Validation of SSCLIP Scoring System to Estimate Survival in Hepatocellular Carcinoma Patients Who Received Curative Liver Resection[J]. Plo S One,2015,10:e0129000.
12 涂海斌,郭佳,林健玲,等. CEUS定量分析鉴别小肝癌分级[J].中国介入影像与治疗学,2017,14:543-547.
13 Tarhan NC,Hatipogˇlu T,Ercan E,et al. Correlation of dynamic multidetector CT findings with pathological grades of hepatocellular carcinoma[J]. Diagnostic&Interventional Radiology,2012,18:236;author reply
14 梁长华,梁盼,刘甲,等.原发性肝癌320排容积CT全肿瘤灌注参数值与微血管密度及其与病理分级的相关性研究[J].临床放射学杂志,2017,36:517-522.
15 Chen CY,Chen J,Xia CC,et al. T1mapping combined with GdEOB-DTPA-enhanced magnetic resonance imaging in predicting the pathologic grading of hepatocellular carcinoma[J]. Journal of Biological Regulators and Homeostatic Agents,2017,31:1029-1036.
16 Peng Z,Jiang M,Cai H,et al. Gd-EOB-DTPA-enhanced magnetic resonance imaging combined with T1mapping predicts the degree of differentiation in hepatocellular carcinoma[J]. BMC Cancer,2016,16:625.
17 Li X,Zhang K,Shi Y,et al. Correlations between the minimum and mean apparent diffusion coefficient values of hepatocellular carcinoma and tumor grade[J]. Journal of Magnetic Resonance Imaging,2016,44:1442-1447.
18 Bai DS,Zhang C,Chen P,et al. The prognostic correlation of AFP level at diagnosis with pathological grade,progression,and survival of patients with hepatocellular carcinoma[J]. Scientific Reports,2017,7:12870.
19 Marrero JA,Kulik LM,Sirlin C,et al. Diagnosis,Staging and Management of Hepatocellular Carcinoma:2018 Practice Guidance by the American Association for the Study of Liver Diseases[J]. Hepatology(Baltimore,Md),2018.
20 Liver EAFTSOT. EASL-EORTC clinical practice guidelines:management of hepatocellular carcinoma[J]. Journal of Hepatology,2012,56:908-943.
21 Ganeshan B,Miles KA. Quantifying tumour heterogeneity with CT[J]. Cancer Imaging:the Official Publication of the International Cancer Imaging Society,2013,13:140-149.
22 Ng F,Kozarski R,Ganeshan B,et al. Assessment of tumor heterogeneity by CT texture analysis:can the largest cross-sectional area be used as an alternative to whole tumor analysis?[J]. European Journal of Radiology,2013,82:342-348.
23 Raman SP,Schroeder JL,Huang P,et al. Preliminary data using computed tomography texture analysis for the classification of hypervascular liver lesions:generation of a predictive model on the basis of quantitative spatial frequency measurements-a work in progress[J].Journal of Computer Assisted Tomography,2015,39:383-395.
24 Li M,Fu S,Zhu Y,et al. Computed tomography texture analysis to facilitate therapeutic decision making in hepatocellular carcinoma[J]. Oncotarget,2016,7:13248-13259.
25 Simpson AL,Adams LB,Allen PJ,et al. Texture analysis of preoperative CT images for prediction of postoperative hepatic insufficiency:a preliminary study[J]. Journal of the American College of Surgeons,2015,220:339-346.
26 Chen S,Zhu Y,Liu Z,et al. Texture analysis of baseline multiphasic hepatic computed tomography images for the prognosis of single hepatocellular carcinoma after hepatectomy:A retrospective pilot study[J]. European Journal of Radiology,2017,90:198-204.
27 Bauer S,Nolte LP,Reyes M. Fully automatic segmentation of brain tumor images using support vector machine classification in combination with hierarchical conditional random field regularization[J]. International Conference on Medical Image Computing and ComputerAssisted Intervention,2011,14:354-361.
28 Huber MB,Nagarajan MB,Leinsinger G,et al. Performance of topological texture features to classify fibrotic interstitial lung disease patterns[J]. Medical Physics,2011,38:2035-2044.