Zaiyi Liu

Intravoxel incoherent motion (IVIM) in evaluation of breast lesions: Comparison with conventional DWI

OBJECTIVES To obtain perfusion as well as diffusion information in normal breast tissues and breast lesions from intravoxel incoherent motion (IVIM) imaging with biexponential analysis of multiple b-value diffusion-weighted imaging (DWI) and compare these parameters to apparent diffusion coefficient (ADC) obtained with monoexponential analysis in their ability to discriminate benign lesions and malignant tumors. MATERIALS AND METHODS In this prospective study, informed consent was acquired from all patients. Eighty-four patients with 40 malignant tumors, 41 benign lesions, 30 simple cysts and 39 normal breast tissues were imaged at 1.5 T utilizing contrast-enhanced magnetic resonance imaging (MRI) and DWI using 12 b values (range: 0-1000 s/mm(2)). Tissue diffusivity (D), perfusion fraction (f) and pseudo-diffusion coefficient (D*) were calculated using segmented biexponential analysis. ADC (b = 0 and 1000 s/mm(2)) was calculated with monoexponential fitting of the DWI data. D, f, D* and ADC values were obtained for normal breast tissues, simple cysts, benign lesions and malignant tumors. Receiver operating characteristic analysis was performed for all DWI parameters. RESULTS There was good interobserver agreement on the measurements between the 2 observers. D values were significantly different among malignant tumors, benign lesions, simple cysts and normal breast tissues (P = 0.000) and it was the same result for f, D* and ADC values. Further comparisons of these 4 parameters between every single pair were as the following. D and ADC values of malignant tumors were significantly smaller than those of benign lesions, simple cysts and normal tissues (P = 0.000, respectively). The f value of malignant tumors was significantly higher than that of benign lesions, simple cysts and normal breast tissues (P = 0.001, P = 0.000, and P = 0.000). D and ADC values demonstrated higher sensitivity and specificity in differentiating benign lesions and malignant tumors, with area under the curve (AUC) of 0.952 and 0.945, respectively, while f and D* with the lower AUC of 0.723 and 0.630, respectively. Combining f and D values had a sensitivity up to 98.75%. CONCLUSION DWI response curves in malignant tumors, benign lesions and normal fibroglandular tissues are found to be biexponential fit in comparison with the monoexponential fit for simple cysts. IVIM provides separate quantitative measurement of D for cellularity and f and D* for vascularity and is helpful for differentiation between benign and malignant breast lesions.

Liver Diffusion-weighted MR Imaging: Reproducibility Comparison of ADC Measurements Obtained with Multiple Breath-hold, Free-breathing, Respiratory-triggered, and Navigator-triggered Techniques

PURPOSE To prospectively compare the reproducibility of normal liver apparent diffusion coefficient (ADC) measurements by using different respiratory motion compensation techniques with multiple breath-hold (MBH), free-breathing (FB), respiratory-triggered (RT), and navigator-triggered (NT) diffusion-weighted (DW) imaging and to compare the ADCs at different liver anatomic locations. MATERIALS AND METHODS The study protocol was approved by the institutional review board, and written informed consent was obtained from each participant. Thirty-nine volunteers underwent liver DW imaging twice. Imaging was performed with a 1.5-T MR imager with MBH, FB, RT, and NT techniques (b = 0, 100, and 500 sec/mm(2)). Three representative sections--superior, central, and inferior--were selected on left and right liver lobes, respectively. On each selected section, three regions of interest were drawn, and ADCs were measured. Analysis of variance was used to assess ADCs among the four techniques and various anatomic locations. Reproducibility of ADCs was assessed with the Bland-Altman method. RESULTS ADCs obtained with MBH (range: right lobe, [1.641-1.662] × 10(-3)mm(2)/sec; left lobe, [2.034-2.054] ×10(-3)mm(2)/sec) were higher than those obtained with FB (right, [1.349-1.391] ×10(-3)mm(2)/sec; left, [1.630-1.700] ×10(-3)mm(2)/sec), RT (right, [1.439-1.455] ×10(-3)mm(2)/sec; left, [1.720-1.755] ×10(-3)mm(2)/sec), or NT (right, [1.387-1.400] ×10(-3)mm(2)/sec; left, [1.661-1.736] ×10(-3)mm(2)/sec) techniques (P < .001); however, no significant difference was observed between ADCs obtained with FB, RT, and NT techniques (P = .130 to P >.99). ADCs showed a trend to decrease moving from left to right. Reproducibility in the left liver lobe was inferior to that in the right, and the central middle segment in the right lobe had the most reproducible ADC. Statistical differences in ADCs were observed in the left-right direction in the right lobe (P < .001), but they were not observed in the superior-inferior direction (P = .144-.450). However, in the left liver lobe, statistical differences existed in both directions (P = .001 to P = .016 in the left-right direction, P < .001 in the superior-inferior direction). CONCLUSION Both anatomic location and DW imaging technique influence liver ADC measurements and their reproducibility. FB DW imaging is recommended for liver DW imaging because of its good reproducibility and shorter acquisition time compared with that of MBH, RT, and NT techniques.

MRI quantification of non-Gaussian water diffusion in normal human kidney: a diffusional kurtosis imaging study

Our aim was to prospectively evaluate the feasibility of diffusional kurtosis imaging (DKI) in normal human kidney and to report preliminary DKI measurements. Institutional review board approval and informed consent were obtained. Forty‐two healthy volunteers underwent diffusion‐weighted imaging (DWI) scans with a 3‐T MR scanner. b values of 0, 500 and 1000 s/mm2 were adopted. Maps of fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (D⊥), axial diffusivity (D||), mean kurtosis (MK), radial kurtosis (K⊥) and axial kurtosis (K||) were produced. Three representative axial slices in the upper pole, mid‐zone and lower pole were selected in the left and right kidney. On each selected slice, three regions of interest were drawn on the renal cortex and another three on the medulla. Statistical comparison was performed with t‐test and analysis of variance. Thirty‐seven volunteers successfully completed the scans. No statistically significant differences were observed between the left and right kidney for all metrics (p values in the cortex: FA, 0.114; MD, 0.531; D⊥, 0.576; D||, 0.691; MK, 0.934; K⊥, 0.722; K||, 0.891; p values in the medulla: FA, 0.348; MD, 0.732; D⊥, 0.470; D||, 0.289; MK, 0.959; K⊥, 0.780; K||, 0.287). Kurtosis metrics (MK, K||, K⊥) obtained in the renal medulla were significantly (p <0.001) higher than those in the cortex (0.552 ± 0.04, 0.637 ± 0.07 and 0.530 ± 0.08 in the medulla and 0.373 ± 0.04, 0.492 ± 0.06 and 0.295 ± 0.06 in the cortex, respectively). For the diffusivity measures, FA of the medulla (0.356 ± 0.03) was higher than that of the cortex (0.179 ± 0.03), whereas MD, D⊥ and D|| (mm2/ms) were lower in the medulla than in the cortex (3.88 ± 0.09, 3.50 ± 0.23 and 4.65 ± 0.29 in the cortex and 2.88 ± 0.11, 2.32 ± 0.20 and 3.47 ± 0.31 in the medulla, respectively). Our results indicate that DKI is feasible in the human kidney. We have reported the preliminary DKI measurements of normal human kidney that demonstrate well the non‐Gaussian behavior of water diffusion, especially in the renal medulla. Copyright

Angiomyolipoma with Minimal Fat: Differentiation From Clear Cell Renal Cell Carcinoma and Papillary Renal Cell Carcinoma by Texture Analysis on CT Images

RATIONALE AND OBJECTIVES To retrospectively evaluate the diagnostic performance of texture analysis (TA) for the discrimination of angiomyolipoma (AML) with minimal fat, clear cell renal cell cancer (ccRCC), and papillary renal cell cancer (pRCC) on computed tomography (CT) images and to determine the scanning phase, which contains the strongest discriminative power. MATERIALS AND METHODS Patients with pathologically proved AMLs (n = 18) lacking visible macroscopic fat at CT and patients with pathologically proved ccRCCs (n = 18) and pRCCs (n = 14) were included. All patients underwent CT scan with three phases (precontrast phase [PCP], corticomedullary phase [CMP], and nephrographic phase [NP]). The selected images were analyzed and classified with TA software (MaZda). Texture classification was performed for 1) minimal fat AML versus ccRCC, 2) minimal fat AML versus pRCC, and 3) ccRCC versus pRCC. The classification results were arbitrarily divided into several levels according to the misclassification rates: excellent (misclassification rates ≤10%), good (10%< misclassification rates ≤20%), moderate (20%< misclassification rates ≤30%), fair (30%< misclassification rates ≤40%), and poor (misclassification rates ≥40%). RESULTS Excellent classification results (error of 0.00%-9.30%) were obtained with nonlinear discriminant analysis for all the three groups, no matter which phase was used. On comparison of the three scanning phases, we observed a trend toward better lesion classification with PCP for minimal fat AML versus ccRCC, CMP, and NP images for ccRCC versus pRCC and found similar discriminative power for minimal fat AML versus pRCC. CONCLUSIONS TA might be a reliable quantitative method for the discrimination of minimal fat AML, ccRCC, and pRCC.

Effects of contrast-enhancement, reconstruction slice thickness and convolution kernel on the diagnostic performance of radiomics signature in solitary pulmonary nodule

The Effects of contrast-enhancement, reconstruction slice thickness and convolution kernel on the diagnostic performance of radiomics signature in solitary pulmonary nodule (SPN) remains unclear. 240 patients with SPNs (malignant, n = 180; benign, n = 60) underwent non-contrast CT (NECT) and contrast-enhanced CT (CECT) which were reconstructed with different slice thickness and convolution kernel. 150 radiomics features were extracted separately from each set of CT and diagnostic performance of each feature were assessed. After feature selection and radiomics signature construction, diagnostic performance of radiomics signature for discriminating benign and malignant SPN was also assessed with respect to the discrimination and classification and compared with net reclassification improvement (NRI). Our results showed NECT-based radiomics signature demonstrated better discrimination and classification capability than CECT in both primary (AUC: 0.862 vs. 0.829, p = 0.032; NRI = 0.578) and validation cohort (AUC: 0.750 vs. 0.735, p = 0.014; NRI = 0.023). Thin-slice (1.25 mm) CT-based radiomics signature had better diagnostic performance than thick-slice CT (5 mm) in both primary (AUC: 0.862 vs. 0.785, p = 0.015; NRI = 0.867) and validation cohort (AUC: 0.750 vs. 0.725, p = 0.025; NRI = 0.467). Standard convolution kernel-based radiomics signature had better diagnostic performance than lung convolution kernel-based CT in both primary (AUC: 0.785 vs. 0.770, p = 0.015; NRI = 0.156) and validation cohort (AUC: 0.725 vs.0.686, p = 0.039; NRI = 0.467). Therefore, this study indicates that the contrast-enhancement, reconstruction slice thickness and convolution kernel can affect the diagnostic performance of radiomics signature in SPN, of which non-contrast, thin-slice and standard convolution kernel-based CT is more informative.

The development and validation of a CT-based radiomics signature for the preoperative discrimination of stage I-II and stage III-IV colorectal cancer

Objectives To investigative the predictive ability of radiomics signature for preoperative staging (I-IIvs.III-IV) of primary colorectal cancer (CRC). Methods This study consisted of 494 consecutive patients (training dataset: n=286; validation cohort, n=208) with stage I–IV CRC. A radiomics signature was generated using LASSO logistic regression model. Association between radiomics signature and CRC staging was explored. The classification performance of the radiomics signature was explored with respect to the receiver operating characteristics(ROC) curve. Results The 16-feature-based radiomics signature was an independent predictor for staging of CRC, which could successfully categorize CRC into stage I-II and III-IV (p <0.0001) in training and validation dataset. The median of radiomics signature of stage III-IV was higher than stage I-II in the training and validation dataset. As for the classification performance of the radiomics signature in CRC staging, the AUC was 0.792(95%CI:0.741-0.853) with sensitivity of 0.629 and specificity of 0.874. The signature in the validation dataset obtained an AUC of 0.708(95%CI:0.698-0.718) with sensitivity of 0.611 and specificity of 0.680. Conclusions A radiomics signature was developed and validated to be a significant predictor for discrimination of stage I-II from III-IV CRC, which may serve as a complementary tool for the preoperative tumor staging in CRC.

In Vitro Labeling of Mesenchymal Stem Cells with Superparamagnetic Iron Oxide by Means of Microbubble-enhanced US Exposure: Initial Experience

PURPOSE To investigate the feasibility of magnetically labeling stem cells with superparamagnetic iron oxide (SPIO) by means of microbubble-enhanced ultrasonographic (US) exposure (MUE) and to study the effects of this approach--without secondary transfection agents--on the viability, proliferation activity, and differentiation capability of MUE-labeled stem cells. MATERIALS AND METHODS Institutional review board approval was obtained for this study. Human mesenchymal stem cells (MSCs) ([1 to 2] x 10(6)/mL) were studied in four experiment groups: sham exposure to US with microbubbles and SPIO (group A), exposure to US with SPIO but without microbubbles (group B), exposure to US with microbubbles and SPIO (group C), and sham exposure to US without SPIO or microbubbles (group D). Intracellular iron uptake was analyzed qualitatively at light and electron microscopy. The viability and proliferation activity of MSCs were evaluated. The adipogenic and osteogenic differentiation capability of the labeled MSCs was also evaluated. Ninety-five percent confidence intervals were derived for assessment of differences in cell viability and proliferation activity between groups C and D. RESULTS Light and electron microscopy revealed intracytoplasmic iron uptake and nearly 100% cell labeling efficiency. The MUE-labeled MSCs had unaltered viability and uncompromised proliferation activity compared with the nonlabeled MSCs. Similar to the nonlabeled MSCs, the MUE-labeled MSCs differentiated into adipogenic and osteogenic lineages. CONCLUSION Initial study results show that stem cells can be effectively labeled with SPIO by using MUE without secondary transfection agents and thus that MUE labeling is an appealing alternative cell-labeling approach that warrants investigation for intracellular magnetic labeling of stem cells. SUPPLEMENTAL MATERIAL http://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.2531081974/-/DC1.

Non-small cell lung cancer: quantitative phenotypic analysis of CT images as a potential marker of prognosis.

This was a retrospective study to investigate the predictive and prognostic ability of quantitative computed tomography phenotypic features in patients with non-small cell lung cancer (NSCLC). 661 patients with pathological confirmed as NSCLC were enrolled between 2007 and 2014. 592 phenotypic descriptors was automatically extracted on the pre-therapy CT images. Firstly, support vector machine (SVM) was used to evaluate the predictive value of each feature for pathology and TNM clinical stage. Secondly, Cox proportional hazards model was used to evaluate the prognostic value of these imaging signatures selected by SVM which subjected to a primary cohort of 138 patients, and an external independent validation of 61 patients. The results indicated that predictive accuracy for histopathology, N staging, and overall clinical stage was 75.16%, 79.40% and 80.33%, respectively. Besides, Cox models indicated the signatures selected by SVM: “correlation of co-occurrence after wavelet transform” was significantly associated with overall survival in the two datasets (hazard ratio [HR]: 1.65, 95% confidence interval [CI]: 1.41–2.75, p = 0.010; and HR: 2.74, 95%CI: 1.10–6.85, p = 0.027, respectively). Our study indicates that the phenotypic features might provide some insight in metastatic potential or aggressiveness for NSCLC, which potentially offer clinical value in directing personalized therapeutic regimen selection for NSCLC.

Central focused convolutional neural networks: Developing a data-driven model for lung nodule segmentation

HighlightsA data‐driven lung nodule segmentation method without involving shape hypothesis.Two‐branch convolutional neural networks extract both 3D and multi‐scale 2D features.A novel central pooling layer is proposed for feature selection.We propose a weighted sampling method to solve imbalanced training label problem.The method shows strong performance for segmenting juxtapleural nodules. Graphical abstract Figure. No caption available. &NA; Accurate lung nodule segmentation from computed tomography (CT) images is of great importance for image‐driven lung cancer analysis. However, the heterogeneity of lung nodules and the presence of similar visual characteristics between nodules and their surroundings make it difficult for robust nodule segmentation. In this study, we propose a data‐driven model, termed the Central Focused Convolutional Neural Networks (CF‐CNN), to segment lung nodules from heterogeneous CT images. Our approach combines two key insights: 1) the proposed model captures a diverse set of nodule‐sensitive features from both 3‐D and 2‐D CT images simultaneously; 2) when classifying an image voxel, the effects of its neighbor voxels can vary according to their spatial locations. We describe this phenomenon by proposing a novel central pooling layer retaining much information on voxel patch center, followed by a multi‐scale patch learning strategy. Moreover, we design a weighted sampling to facilitate the model training, where training samples are selected according to their degree of segmentation difficulty. The proposed method has been extensively evaluated on the public LIDC dataset including 893 nodules and an independent dataset with 74 nodules from Guangdong General Hospital (GDGH). We showed that CF‐CNN achieved superior segmentation performance with average dice scores of 82.15% and 80.02% for the two datasets respectively. Moreover, we compared our results with the inter‐radiologists consistency on LIDC dataset, showing a difference in average dice score of only 1.98%.

Computed tomography texture analysis to facilitate therapeutic decision making in hepatocellular carcinoma

This study explored the potential of computed tomography (CT) textural feature analysis for the stratification of single large hepatocellular carcinomas (HCCs) > 5 cm, and the subsequent determination of patient suitability for liver resection (LR) or transcatheter arterial chemoembolization (TACE). Wavelet decomposition was performed on portal-phase CT images with three bandwidth responses (filter 0, 1.0, and 1.5). Nine textural features of each filter were extracted from regions of interest. Wavelet-2-H (filter 1.0) in LR and wavelet-2-V (filter 0 and 1.0) in TACE were related to survival. Subsequently, LR and TACE patients were divided based on the wavelet-2-H and wavelet-2-V median at filter 1.0 into two subgroups (+ or −). LR+ patients showed the best survival, followed by LR-, TACE+, and TACE-. We estimated that LR+ patients treated using TACE would exhibit a survival similar to TACE- patients and worse than TACE+ patients, with a severe compromise in overall survival. LR was recommended for TACE- patients, whereas TACE was preferred for LR- and TACE+ patients. Independent of tumor size, CT textural features showed positive and negative correlations with survival after LR and TACE, respectively. Although further validation is needed, texture analysis demonstrated the feasibility of using HCC patient stratification for determining the suitability of LR vs. TACE.