Radiomics based on multimodal MRI for the differential diagnosis of benign and malignant lung nodule/mass / 中华放射学杂志

Objective:To develop a multimodal MRI-based radiomics model for the differential diagnosis of benign and malignant lung lesions, and to compare the discriminative abilities of different models.Methods:Totally 114 patients with 115 lesions (44 benign and 71 malignant) in Nantong First Peoples′s Hospital from January 2014 to October 2019 were included in the study. All patients underwent non-enhanced MR examination, and textural features from T 1WI,T 2WI and apparent diffusion coefficient (ADC) imaging were extracted. The feature selection methods included L1 based, mutual information, tree based, recursive feature elimination and F-test. Then we constructed a prediction model by using logistic regression (LR), support vector machine (SVM), random forest (RF) and k-nearest neighbor (KNN) respectively. In order to control the number of modeling features and reduce the ininterpretability of the model, the new model was obtained by manually modifying some parameters of the hyperparameter model. One hundred and fourteen cases were rotated as training and validation sets. The performance of each model was evaluated by confounding matrix and receiver operating characteristic (ROC) curve. Results:The area under the curve (AUC) of T 2WI based LR model for the differential diagnosis of benign and malignant pulmonary nodules/masses was 0.71 and the F1 score was 0.57. Based on T 1WI images, LR and SVM model could be used to identify benign and malignant pulmonary nodules, the AUC before parameter adjustment were 0.77 and 0.78, the accuracy after parameter adjustment (LR a,SVM a) was 0.67, 0.70, and both the AUC were 0.72. However, no matter which feature or classifier was selected, both the AUC and accuracy of ADC-based model were less than 0.70. Conclusion:Multimodal MRI-based radiomics model is valuable for the differential diagnosis of benign and malignant pulmonary nodules/masses, and T 1WI-based model shows the best discrimination.

The value of diffusion-weighted imaging based on monoexponential and biexponential model in predicting the response of chemotherapy in non-small cell lung cancer patients / 中华放射学杂志

Objective To investigate the ADC value and intravoxel incoherent motion (IVIM) parameters to predict the early response to chemotherapy in patients with locally advanced non-small cell lung cancer (NSCLC). Methods Twenty-six patients with pathologically confirmed NSCLC in our hospital from June 2015 to January 2017 were prospectively studied. MR routine scan, DWI and IVIM scans were performed before and at the end of the first cycle of chemotherapy. The ADC value, pure diffusion parameter (D), perfusion parameter (D*) and perfusion fraction (f) before and after chemotherapy were recorded respectively, and the maximum diameter of tumor (MDMRI) was measured on T2WI. The effective group and the ineffective group were divided according to RECIST 1.1. Intra- and interobserver reproducibility were assessed with intraclass correlation coefficient (ICC). The differences of the parameters and their change rate were compared pre-and posttreatment, in the effective and the ineffective groups by using t test. Receiver operating characteristic (ROC) curves were generated to explore the optimal parameters and thresholds for early prediction of the efficacy of chemotherapy. Z test was used to compare the differences in the areas under the ROC curves. Results The repeatability of D*was relatively poor (ICC 0.507-0.716). The ADC and D values of the effective group before chemotherapy were (1.28±0.21)×10-3 and (1.05±0.14)×10-3mm2/s, respectively, and those of the ineffective group were (1.55 ± 0.25) × 10-3 and (1.29 ± 0.18) × 10-3mm2/s, respectively. The differences between two groups were statistically significant (t=-2.989, -3.755, respectively, P all<0.05). While there was no significant difference in f between the effective and ineffective groups before chemotherapy (t=-1.034, P=0.312). The ADC, D and f values of the effective group before chemotherapy were lower than those after chemotherapy, and the differences were statistically significant (t=-3.969,-5.617 and-2.876, respectively, P all<0.05). While there was no significant difference in ADC and D values in ineffective group (t=-0.737, -0.866, P>0.05, respectively). The difference of f was statistically significant (t=-2.731, P<0.05).ΔADC％,ΔD％andΔMDMRI％in effective group were (23.90 ± 22.28)％, (32.58 ± 20.63)％ and (18.77 ± 10.36)％, respectively, and those in ineffective group were (1.19 ± 11.35)％, (6.22 ± 21.63)％ and (9.35 ± 8.26)％, respectively. The differences between two groups were statistically significant (t=3.393, 3.154 and 2.487, respectively, P all<0.05). There was no significant difference inΔf％between two groups (t=1.346, P=0.191). ROC curve analysis showed that the area under curve (AUC) of preADC, preD, ΔADC％, ΔD％ and ΔMDMRI％ was 0.788, 0.848, 0.800, 0.830 and 0.727, respectively. When setting the preD=1.20×10-3 mm2/s as the boundary under the largest AUC, however, there was no significant differences in AUC compared with preADC, ΔADC％, ΔD％ and ΔMDMRI％ (Z=1.336, 0.363, 0.136, 1.395, P>0.05). Conclusion The ADC value, D value before chemotherapy and the early change rate after chemotherapy can better predict the early response of chemotherapy in NSCLC patients.

Short-term reproducibility and impact factors of intravoxel incoherent motion parameters for lung cancer / 中国医学影像技术

Objective To prospectively evaluate the short-term test-retest reproducibility of intravoxel incoherent motion (IVIM) parameters for lung cancer,and to investigate its impact factors.Methods Thirty-eight lung cancer patients underwent twice free breathing IVIM scanning (0.5-1.0 h interval).Two radiologists independently analyzed IVIM images to obtain true diffusion coefficient (D),pseudo-diffusion coefficient (D*) and perfusion fraction (f).All the parameters were compared between repeated measurements and scans.Intra-and inter-observer,test-retest reproducibility was assessed with interclass correlation coefficients (ICCs).The impact of lung cancer type (peripheral and central),size (the largest diameter ≥2 cm and ＜2 cm) and location (upper,mid,and lower lung field) on short-term reproducibility were compared by using coefficient of variations (WCVs).Results There was no significant difference between repeated measurements for all parameters (all P＞ 0.05).All the parameters showed good intra-and inter-observer agreement.WCVsof D* and f (19.88％-36.83％) were higher than those of D (2.97％-4.62％).WCVs of D were higher for central lung cancer and the largest diameter ＜2 cm,and increasing from apical to lower zone.WCV of f was larger in the central lung cancer and the largest diameter ≥2 cm,while D* was greatly influenced by type,size and location of lung cancer.Conclusion The test retest reproducibility of D is good,while of D* and f are poor in lung cancer.The type,size and location of lung cancers are the impact factors of IVIM parameters.

Progresses of intravoxel incoherent motion DWI in evaluation of treatment response of tumors / 中国医学影像技术

Intravoxel incoherent motion (IVIM) can distinguish the molecular diffusion from the pseudo-diffusion of microcircular perfusion to obtain true diffusion coefficient (D),pseudo-diffusion coefficient (D*) and perfusion fraction (f),which are sensitive to detect alterations in cellularity and microcirculation perfusion.IVIM can also evaluate early therapeutic effects,optimize treatment plan,predict prognoses,and provide information for choosing appropriate treatment methods.In this paper,the principle of IVIM and its application in tumor response evaluation were reviewed.

A new ursane caffeoyl ester from the seeds of Impatiens balsamina L / 中国药科大学学报

A new triterpene ester,α-amyrin caffeate(1),was isolated from the seeds of Impatiens balsamina L.

Chemical constituents from Hedyotis diffusa / 中国中药杂志

<p><b>OBJECTIVE</b>To investigate the chemical constituents from Hedyotis diffusa.</p><p><b>METHOD</b>The compounds were isolated and purified by various chromatographic techniques and identified by their physicochemical properties and spectral data.</p><p><b>RESULT</b>Eight compounds had been reported in last paper, and this time eight more compounds were isolated and identified as 6-hydroxystigmasta-4,22-dien-3-one (1), 3-hydroxystigmasta-4,22-dien-7-one (2), 2-hydroxy-3-methylanthraquinone (3), 2,6-dihydroxy-3-methyl-4-methoxyanthraquinone (4), iso-scutellarein (5), isoetin (6), aesculetin (7), gypsogenic acid (8).</p><p><b>CONCLUSION</b>Compounds 1-3, 5-8 were obtained from the genus Hedyotis for the first time.</p>

Chemical constituents in leaves of Acanthopanax gracilistylus / 中草药

Objective To study chemical constituents in the leaves of Acanthopanax gracilistylus.Methods The chemical components were isolated and purified by silica gel,ODS C-18,and Sephadex LH-20 column chromatogram.The chemical structures were elucidated on the basis of physicochemical properties and spectral data.Results Seventeen compounds were isolated and identified as(-)-kaur-16-en-19-oic acid(Ⅰ),quercetin(Ⅱ),kaempferol(Ⅲ),protocatechuic acid(Ⅳ),acankoreoside A(Ⅴ),acantrifoside A(Ⅵ),3?,11?-dihydroxy-20(29)-lupene-23,28-dioic acid(Ⅶ),?-sitosterol(Ⅷ),daucosterol(Ⅸ),palmitic acid(Ⅹ),rutin(Ⅺ),stigmast-5,22-dien-3-O-?-D-glucopyranoside(ⅩⅡ),acankoreagenin(ⅩⅢ),3,11-dihydroxy-23-oxo-20(29)-lupen-28-oic acid(ⅩⅣ),3-hydroxy-23-oxo-20(29)-lupen-28-oic acid(ⅩⅤ),myristin(ⅩⅥ),and acanthopanaxgric acid(ⅩⅦ).Conclusion Compounds Ⅱ-Ⅳ,Ⅶ,Ⅺ,ⅩⅡ,ⅩⅣ,and ⅩⅤ are obtained from the leaves of the plant for the first time and compounds ⅩⅦ is a new proved compound named acanthopanaxgric acid.