RESUMO
Objective:To investigate the correlation between three-dimensional histogram analysis of dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)and Gleason score(GS)in prostate cancer(Pca)from two hospital, and its diagnostic efficacy for discriminating low-grade from high-grade Pca.Methods:A total of 102 pathologically confirmed Pca patients in the First Affiliated Hospital of Zhejiang Chinese Medical University and Hangzhou Traditional Chinese Medical Hospital(TCM Hospital)Affiliated to Zhejiang Chinese Medical University from January 2017 to October 2020 were retrospectively analyzed.The quantitative parameters of Pca, including transport constant(K trans), rate constant(K ep), percent volume of the extravascular extracellular space(V e)and fraction of the Intraplasmic contrast volume(V p), were obtained by manually layer by layer delineating of interested regions of all lesions on the original DCE-MRI imaging.Then the three-dimensional histogram analysis of the above parameters were performed to obtain the minimum, maximum, median, mean, area, 10 thpercentile, 25 thpercentile, 75 thpercentile and 90 thpercentile.The correlations between quantitative parameters and GS, and diagnostic efficiencies were analyzed. Results:102 Pca patients were divided into low-grade prostate cancer group(GS≤3+ 4)(n=44)and high-grade Pca group(GS≥4+ 3)(n=58). There were no statistically significant differences in age and location of lesions between the two groups( P>0.05), but there were statistically significant differences in Gleason score, PSA level and lesion diameter between the two groups( U=0.000, 730.000, 711.000, all P<0.05). The median, mean, 10 thpercentile, 25 thpercentile, 75 thpercentile, 90 thpercentile derived from K trans, and K ep(median, mean, 10%, 25%, 75%, 90%)together with maximum of K transand mean for V e were positively correlated with GS( r=0.405 to 0.583, P<0.05), in which mean of K transhad the highest positive correlation( r=0.583, P=0.000). The histogram parameters derived from V pwere negatively correlated with GS( r=-0.301 to 0.341, P<0.05). The area under ROC of 75th percentile derived from K transwas the highest(0.832). When the cut-off value of 75 thpercentile derived from K transwas ≥0.680/min, its Youden index, sensitivity, and specificity were 0.594, 0.776, 0.818, respectively. Conclusions:The three-dimensional histogram of DCE-MRI quantitative parameters has correlation with GS in Pca patients, can be used to discriminate low-grade from high-grade Pca.
RESUMO
Objective:To investigate the value of radiomics based on unenhanced CT texture analysis in predicting the WHO/International Society of Urological Pathology (ISUP) grading of clear cell renal cell carcinoma (ccRCC).Methods:Postoperative pathology-confirmed ccRCC subjects ( n=90) who received CT scanning and had a definite pathological grading in Cancer Hospital of the University of Chinese Academy of Sciences were collected retrospectively from December 2016 to May 2019. The cases were randomly divided into training group ( n=63) and test group ( n=27) as a ratio of 7∶3. All cases were classified into low grade (grades Ⅰ and Ⅱ, n=57) and high grade (grades Ⅲ and Ⅳ, n=37) according to the new pathological grading (WHO/ISUP grading, version 2016) of renal carcinoma. 3D-ROI segmentation was performed on unenhanced CT images and 93 texture features were extracted. The least absolute shrinkage and selection operator (LASSO) regression was used to reduct dimension of texture parameters and then the radiomics score (Rad-score) was established. The logistic regression was used to develop the prediction model with the pathological grading as the gold standard. The ROC curve and calibration curve were used to evaluate the predictive performance of the model, and the area under the curve (AUC), accuracy, sensitivity and specificity were calculated. The Hosmer-Lemeshow test was used to evaluate calibration degree of the model. Results:The 10 non-zero coefficient texture features were screened out through dimension reduction steps. The Rad-score was formed according to the linear combination of these ten features and corresponding coefficients, and then the prediction model was developed. The AUC of the model in training group was 0.933 (95%CI 0.862-1.000), the sensitivity was 92.3%, the specificity was 89.2%, and the model accuracy was 90.5%. The calibration curve showed the good calibration ( P=0.257). The AUC value in test group was 0.875 (95%CI 0.734-1.000), the sensitivity, specificity and accuracy were 72.7%, 87.5% and 81.5%. The calibration curve showed the good calibration ( P=0.125). Conclusion:The radiomics prediction model based on unenhanced CT texture analysis have application potential for the evaluation of WHO/ISUP grading of ccRCC.
RESUMO
Objective@#To investigate the prognostic value of radiomics analysis in predicting axillary lymph nodes (ALN) metastasis of breast cancer based on dynamic contrast-enhanced MR imaging (DCE-MRI).@*Methods@#One hundred and ninety-six patients with suspected breast cancer were prospectively collected for dynamic breast DCE-MRI. Enhanced MR imaging data of 72 axillary lymph nodes were evaluated separately by a chief radiologist and a resident, and the consistency analysis was performed. Lymph nodes were dichotomized according to the pathology results derived from operation or biopsy under real-time virtual sonography based on MRI data. Clinical and imaging data were also divided into corresponding groups. (Imaging) Data from both groups were respectively classified as training set and testing set by stratified sampling in proportion with 3∶1. AK software was applied to extract 6 major categories of 385 features (including histogram, morphology, texture parameters, gray level co-occurrence matrix, run-length matrix and grey level zone size matrix from imaging), and a set of statistically significant features were subsequently obtained by dimension reduction. The prediction model was established through binary classification logistic regression and employed to externally test the validation set by the method of confusion matrix. Meanwhile, ROC analysis was applied to assess the diagnostic performance of the model.@*Results@#Of the 72 axillary lymph nodes, 35 were metastatic negative and 37 were positive. The consistency of enhanced MRI radiomics features was good, between 0.841 and 0.980. Uniformity, ClusterProminence_AllDirection_offset1_SD, Correlation_AllDirection_offset1, LongRunEmphasis_angle90_offset7 and SurfaceVolumeRatio were statistically significant differences (P<0.01), the area under the ROC between 0.747 and 0.931. In the training and testing group, the areas under the ROC, sensitivity, specificity and accuracy of the model were 0.953, 0.893, 0.926, 92.6% (50/54) and 0.944, 0.900, 1.000, 88.9% (16/18) respectively.@*Conclusion@#The prediction model based on radiomic features may provide a non-invasive and effective approach to the assessment of the risk of ALN metastasis of breast cancer.
RESUMO
Objective@#To explore the risk factors of predicting white matter hyperintensities progression based on radiomics of MRI of whole-brain white matter.@*Methods@#The imaging and clinical data of 152 patients with white matter hyperintensities admitted to Zhejiang People′s Hospital from March 2014 to October 2018 were retrospectively analyzed. The whole brain white matter on baseline T1WI images of each patient were segmented by SPM12 software package, and images of white matter were imported into AK software for texture feature extraction and dimensionality reduction. At last, least absolute shrinkage and selection operator(LASSO) was used to calculate the score of radiomics signature of each patient. According to the improved Fazekas scale, patients with WMH progression were divided into three groups: any white matter hyperintensities (AWMH), periventricular white matter hyperintensities (PWMH) and deep white matter hyperintensities (DWMH). Statistical differences of clinical factors and radiomics signature between WMH progression subgroups and non-progression subgroups were compared with independent sample t test or Mann-Whitney U test, and Univariate logistic regression was used to select independent clinical risk factors and multivariate logistic regression along with imaging omics tags were used to construct predictive models, which was evaluated by ROC curve. And the correlation between the clinical indicators of independent risk factors and the texture feature of radiomics signature was analyzed.@*Results@#The efficacy of the model for the detection for AWMH progression, PWMH progression and DWMH progression was 0.818,0.778 and 0.824, respectively. Age is an independent risk factor for AWMH progression and DWMH progression[OR(95%CI), 4.776(2.152-10.601) vs. 3.851(1.101-8.245); P<0.05]. BMI is an independent risk factor for DWMH[OR(95%CI), 3.004(1.204-7.370); P<0.05], and hyperlipidemia is an independent risk factor for AWMH and PWMH[OR(95%CI), 4.062(1.834-8.998) vs. 3.549(1.666-7.563); P<0.05]. In AWMH subgroup, Surface Area was negatively correlated with age and low density lipoprotein(LDL) (r=-0.401, -0.312), and Inverse Difference Moment_ALLDirection_offset 1_SD was negatively correlated with age(r=-0.412). In PWMH subgroup, Compactness 1 was negatively correlated with LDL(r=-0.198), and Inverse Difference Moment_angle 0_offset 7 was positively correlated with LDL(r=0.252). In DWMH subgroup, LongRunEmphasis_ALLDirection_offset 7 was negatively correlated with age(r=-0.322), and GLCMEntropy_angle0_offset 4 was negatively correlated with age(r=-0.278). GLCMEntropy_AllDirection_offset4 was negatively correlated with body mass index(r=-0.514).@*Conclusion@#Radiomics based on whole-brain white matter MR imaging can predict WMH progression and identify the risk factors in high-risk populations, thus providing early additional information to conventional magnetic resonance imaging to predict WMH progression.
RESUMO
Objective To investigate the prognostic value of radiomics analysis in predicting axillary lymph nodes (ALN) metastasis of breast cancer based on dynamic contrast-enhanced MR imaging (DCE-MRI). Methods One hundred and ninety-six patients with suspected breast cancer were prospectively collected for dynamic breast DCE-MRI. Enhanced MR imaging data of 72 axillary lymph nodes were evaluated separately by a chief radiologist and a resident, and the consistency analysis was performed. Lymph nodes were dichotomized according to the pathology results derived from operation or biopsy under real-time virtual sonography based on MRI data. Clinical and imaging data were also divided into corresponding groups. (Imaging) Data from both groups were respectively classified as training set and testing set by stratified sampling in proportion with 3∶1. AK software was applied to extract 6 major categories of 385 features (including histogram, morphology, texture parameters, gray level co-occurrence matrix, run-length matrix and grey level zone size matrix from imaging), and a set of statistically significant features were subsequently obtained by dimension reduction. The prediction model was established through binary classification logistic regression and employed to externally test the validation set by the method of confusion matrix. Meanwhile, ROC analysis was applied to assess the diagnostic performance of the model. Results Of the 72 axillary lymph nodes, 35 were metastatic negative and 37 were positive. The consistency of enhanced MRI radiomics features was good, between 0.841 and 0.980. Uniformity, ClusterProminence_AllDirection_offset1_SD, Correlation_AllDirection_offset1, LongRunEmphasis_angle90_offset7 and SurfaceVolumeRatio were statistically significant differences (P<0.01), the area under the ROC between 0.747 and 0.931. In the training and testing group, the areas under the ROC, sensitivity, specificity and accuracy of the model were 0.953, 0.893, 0.926, 92.6% (50/54) and 0.944, 0.900, 1.000, 88.9% (16/18) respectively. Conclusion The prediction model based on radiomic features may provide a non-invasive and effective approach to the assessment of the risk of ALN metastasis of breast cancer.
RESUMO
Objective To investigate the value of support vector machine based MRI-radiomics in the differential diagnosis of primary hepatic carcinomas (PHCs). Methods PHCs patients were retrospectively collected from July 2013 to February 2017 in the First Affiliated Hospital of Zhejiang University.All patients underwent unenhanced and enhanced MRI liver scan before surgery,and confirmed by pathology. A total of 294 PHCs patients (305 lesions), including 96 cases (97 lesions) of massive type cholangiocarcinoma (mCC), 107(107 lesions)of hepatocellular carcinoma (HCC), and 91 (101 lesions) of mixed hepatocellular and cholangiocellular carcinomas(HCC-CC).All patients underwent unenhanced and dynamic enhanced MRI liver scan including arterial, portal venous and equilibrium phases. Two hundred and three lesions (65 mCC, 71 HCC, 67 HCC-CC) were assigned into the training set, the remaining 102 lesions(32 mCC,36 HCC,34 HCC-CC)into the validation set,according to a ratio of 2:1.The entire lesions were delineated manually using a region of interest on equilibrium phase of enhanced MRI by using a home-made dedicated software(Analysis Kit,AK,General Electrics,US).Then the least absolute shrinkage and selection operator (LASSO) regression was used to select image features with a method of 10 fold cross-validation, and to reduce the dimensionality. The spearman method was used afterwards to condense the image features by removing redundant.A predictive model of diagnosing the different types of PHCs was established based on support vector machines(SVM),and the accuracy of applying the model in the data sets was used to evaluate the diagnostic efficacy of the model. Results A total of 280 quantitative imaging features were extracted in the training set.Thirty one imaging features were selected after LASSO regression and dimensionality reduction,and 21 features were remained after redundancy removing.The SVM showed the best generalization ability when the first 11 imaging features were used due to the Hughes effect.The 11 imaging features include 4 parameters of histogram,2 of textures,4 of gray-level co-occurrence matrix and 1 of gray-level run length matrix. A predictive model for PHCs was established after the study of the 11 imaging features and a regression analysis using SVM.The accuracy of the predictive model was 80.3% (163/203) in differentiating PHCs in the training set. The accuracy of the model was 75.5% (77/102) after applying it in the validation set. The diagnostic accuracy for HCC-CC, HCC and mCC was 85.3% (29/34), 77.8% (28/36) and 62.5% (20/32), respectively, in the validation set. For HCC-CC, 3 cases were misdiagnosed as mCC and 2 cases as HCC.For HCC,3 cases were misdiagnosed as HCC-CC and 5 cases as mCC.For mCC,9 cases were misdiagnosed as HCC-CC and 3 cases as HCC.The model showed the highest accuracy in predicting HCC-CC.Conclusion Radiomics method based on SVM may have a high accuracy in predicting different pathologic types of PHC,with the highest accuracy for HCC-CC.
RESUMO
Objective To explore the value of CT radiomics model in predicting three-year survival time in patients with primary hepatocellular carcinoma (HCC). Methods Eighty one patients pathologically or clinically confirmed HCC and B stageof Barcelona clinical liver cancer before transcatheter arterial chemoembolization (TACE) in Zhejiang Cancer Hospitalwere retrospectively enrolled from January 2010 to June 2014.A primary cohort consisted of 64 patients and an independent validation cohort consisted of 17 patients. The patients were divided into survival group of 39 cases and death groupof 42 cases duringthree-year follow-up. All the patients underwentnon-enhanced and contrast-enhanced CTimages scan before TACE. Three hundered and seventy six quantization radiomics features were extracted from the arterial phase and portal phase CTimages of target lesion. LASSO regression model was used for data dimension reduction. Logistic regression was used to develop the prediction model. The predictive ability of the model was validated using the area under the curve (AUC) of receiver operating characteristic(ROC) analysis. Results The radiomics features selected from the arterial and portal phase were 8 and 5, respectively. The arterial prediction model showed AUC=0.833, sensitivity=83.9%(26/31), specificity=81.8%(27/33), accuracy=82.8%(53/64)in primary datasetand AUC=0.861, sensitivity=75.0%(6/8), specificity=100.0%(9/9), accuracy=88.2%(15/17)in independent validation dataset.The portal prediction model showed AUC=0.858, sensitivity=83.3%(25/30), specificity=85.3%(29/34), accuracy=84.4%(54/64)in primary dataset and AUC=0.750, sensitivity=75.0%(6/8), specificity=100.0%(9/9), accuracy=88.2(15/17)in independent validation dataset. Conclusion This study shows CT radiomics model can be conveniently used to facilitate the preoperative individualized prediction of three-year survival time in patients with HCC.
RESUMO
<p><b>OBJECTIVE</b>To explore the application value of texture analysis of magnetic resonance images (MRI) in predicting the efficacy of neoadjuvant chemoradiotherapy(nCRT) for rectal cancer.</p><p><b>METHODS</b>A total of 34 rectal cancer patients who were hospitalized at Zhejiang Provincial People's Hospital from February 2015 to April 2017 were prospectively enrolled and received 3.0T MRI examination at pre-nCRT (1 day before nCRT), early stage (at 10-day after nCRT) and middle stage (at 20-day after nCRT).</p><p><b>INCLUSION CRITERIA</b>distance from tumor lower margin to anal edge was less than 12 cm under rectoscope; rectal cancer was confirmed by preoperative pathology; clinical stage was T3 or above; lymph node metastasis existed but without distant metastasis; functions of liver, kidney and heart present no contraindications of operation.</p><p><b>EXCLUSION CRITERIA</b>unfinished nCRT, surgery and three examinations of MRI; image motion artifacts; lack of postoperative pathological results. All the patients underwent rectal cancer long-term three-dimensional radiotherapy and chemotherapy combined with nCRT (oxaliplatin plus capecitabine). The tumor regression grading (TRG) was divided into TRG 0 to 4 grade after nCRT, and TRG 4 was classified as pathological complete remission (pCR); TRG 2 to 3 was classified as partial remission (PR); the rest was no remission (NR). Extraction and analysis of texture features in T2-weighted MR-defined tumor region were performed using Omni Kinetics texture software. The texture values of each time point were statistically analyzed, and the differences of texture values and change differences between pCR and PR+NR, and NR and pCR+PR were compared respectively. Statistically significant texture values were screened and were used in receiver operating characteristic (ROC) curve to assess the prediction of the efficacy of nCRT.</p><p><b>RESULTS</b>Of 34 patients, 21 were males and 13 were females with median age of 49.3 years. Nineteen (55.9%) patients were low rectal adenocarcinoma and 15 (44.1%) patients were middle rectal adenocarcinoma. Nine (26.5%) cases belonged to pCR, 13 (38.2%) belonged to PR, and 12 (35.3%) belonged to NR. Before nCRT, the entropy of tumor area in pCR patients was significantly higher than that in PR+NR patients (7.164±0.272 vs. 6.823±0.309, t=2.925, P=0.006). At the middle stage of nCRT, as compared with PR+NR patients for the texture features of tumor region, the variance (1566±281 vs. 2883±867, t=-4.435, P=0.000) and entropy(5.436±0.934 vs. 6.803±0.577, t=-4.118,P=0.002) of pCR patients were significantly lower; kurtosis(4.800±1.288 vs. 3.206±1.211, t=3.333, P=0.002) and energy (0.016±0.005 vs. 0.010±0.004, t=3.240, P=0.003) of pCR patients were significantly higher. As compared to pCR+PR patients, the kurtosis(2.461±0.931 vs. 4.264±1.205, t=-4.493, P=0.000) and energy (0.011±0.004 vs. 0.014±0.004, t=-3.453, P=0.000) of the NR patients were significantly lower. As for texture change values between early stage and middle stage, the entropy difference was significant between pCR and PR+NR, NR and pCR+PR (1.344±0.819 vs. 0.489±0.319, t=3.047, P=0.014; 0.446±0.213 vs. 0.917±0.677, t=-3.638, P=0.001, respectively). As for texture change values between pre-nCRT and middle stage, variance and entropy differences between pCR and PR+NR (1759±1226 vs. 977±842, t=2.113, P=0.042; 1.728±0.918 vs. 0.524±0.355, t=3.832, P=0.004), and the change values of entropy between NR and pCR+PR (0.475±0.349 vs. 1.044±0.860, t=-2.722, P=0.011) were statistically significant. The above indicators were included in the ROC curve. The results revealed that at the middle stage, entropy value >5.983 indicated the best efficacy for the diagnosis of pCR, with the area under the ROC curve (AUC) of 0.885, the sensitivity of 100%, and the specificity of 66.7%; the energy <0.010 indicated the best AUC for diagnosis of NR was 0.902, with the sensitivity of 91.7% and specificity of 81.8%.</p><p><b>CONCLUSIONS</b>Texture analysis based on T2 weighted images can predict the efficacy of nCRT for rectal cancer. The middle stage of nCRT is the best time of prediction. The entropy and energy of this period are texture parameters having higher predictive ability.</p>
Assuntos
Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Quimiorradioterapia , Espectroscopia de Ressonância Magnética , Terapia Neoadjuvante , Estadiamento de Neoplasias , Valor Preditivo dos Testes , Prognóstico , Neoplasias Retais , Diagnóstico por Imagem , Terapêutica , Resultado do TratamentoRESUMO
Objective To discuss the application of routine CT three-phase perfusion parameter,that is arterial enhancement fraction (AEF) value,in evaluating the curative effect of transcatheter arterial chemoembolization (TACE) for hepatocellular carcinoma (HCC).Methods The clinical data of a total of 30 patients with pathologically proved HCC were enrolled in this study.Routine CT three-phase perfusion scan was performed 1-3 days before as well as 30-40 days after TACE in all patients.AEF value was calculated by using CT Kinetics software (GE Healthcare).The formula for calculating AEF value was as follows:AEF value=(arterial phase CT value-plain scan CT value)÷(portal phase CT value-plain scan CT value).The results were statistically analyzed.Results Effective treatment group had 17 patients,and ineffective treatment group had 13 patients.The postoperative AEF values in the effective treatment group and the ineffective treatment group were (0.351±0.090) and (0.438±0.050) respectively,the difference between the two groups was statistically significant (P<0.05).Taking postoperative AEF value of 0.392 as the critical value to predict the postoperative effect of TACE,the sensitivity and specificity were 86.7% and 73.2% respectively,and the area under the curve was 0.876 (P<0.001).Conclusion The routine CT three-phase perfusion parameter (AEF) can quantitatively reflect the hemodynamic changes of HCC after TACE,which is helpful for making early evaluation of TACE effect,meanwhile,no additional radiation dose will be added.
RESUMO
Transcatheter arterial chemoembolization (TACE) has already been a mature and an effective treatment for advanced hepatocellular carcinoma (HCC).Clinically,it is very important to quickly and accurately evaluate the postoperative curative effect with minimally invasive technique so as to determine the next treatment options.At present,postoperative conventional CT and MRI are the main means to assess the curative effect of TACE,but it is a pity that after the treatment the functional changes of the tumor occur earlier than the morphological changes.In recent years,functional MRI techniques,such as diffusionweighted imaging (DWI),multi-b value DWI,dynamic contrast-enhanced (DCE) imaging,etc.have been more and more used for quantitative evaluation of the diffusion of water molecules and the blood microcirculation perfusion within the tumor tissue,and some progresses have been achieved in the evaluation of curative efficacy for tumor.This paper aims to make a comprehensive review about the research progress of the above mentioned functional imaging methods as well as their current application status in evaluation of the curative effect of TACE.
RESUMO
Objective To study the value of transfer constant(Ktrans)derived from dynamic contrast-enhanced MRI (DCE-MRI) for quantitative evaluation of Ki-67 labeling index (Ki-67 LI) in glioma. Methods Twenty patients with glioma who underwent DCE-MRI and operation were retrospectively reviewed. The Ktrans value and Ki-67 LI were acquired and correlated using the Spearman correlation test. Also, the Ktrans values were compared between high(larger than 10%)and low(no more than 10%)Ki-67 LI group with Mann-Whitney U test, receiver operating characteristic curves was performed to evaluate the diagnostic value. Results The Ktrans value(0.0165 to 0.8048, median 0.1252)was significantly associated with Ki-67 LI(5%to 50%, median 20%) (r=0.721,P<0.001), and the Ktrans value was significantly higher in high Ki-67 group(0.0810 to 0.8048, median 0.1810)than that in low Ki-67 LI group(0.0165 to 0.1456, median 0.0697)(Z=-3.209, P=0.001). The most predictive Ktrans value differentiated high Ki-67 LI and low Ki-67 LI with an area under the curve(AUC) of 0.945 at a sensitivity of 92.3% and specificity of 85.7%. Conclusion Ktrans value could be used for quantitative evaluation of Ki-67 LI in glioma.