Reproducibility of radiomics features derived from intravoxel incoherent motion diffusion-weighted MRI of cervical cancer

2020 ◽  
pp. 028418512093447 ◽  
Author(s):  
Hao Chen ◽  
Yaoyao He ◽  
Cecheng Zhao ◽  
Lili Zheng ◽  
Ning Pan ◽  
...  
Keyword(s):  
Cervical Cancer ◽  
Diffusion Coefficient ◽  
Intraclass Correlation ◽  
B Value ◽  
Intravoxel Incoherent Motion ◽  
Observer Variability ◽  
Multicenter Studies ◽  
Diffusion Weighted Images ◽  
Diffusion Weighted ◽  
Adc Map

Background The reproducibility of intravoxel incoherent motion (IVIM)-based radiomics studies in humans has not been reported. Purpose To determine the inter- and intra-observer variability on the reproducibility of IVIM-based radiomics features in cervical cancer (CC). Material and Methods The IVIM images of 25 patients with CC were retrospectively collected. Based on the high-resolution T2-weighted images, the regions of interest (ROIs) were independently delineated twice in diffusion-weighted images at a b value of 1000 s/mm2 (interval time was one month) by two radiologists. This was done at the largest transversal cross-sections of the tumors. The ROI was subsequently used in apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) maps derived from IVIM images. In total, 105 radiomics features were then finally extracted from the IVIM-derived maps. The inter- and intra-observer reproducibility of IVIM-derived features was then evaluated using the intraclass correlation coefficient. Results Inter- and intra-observer variability affected the reproducibility of radiomics features. D* map had 100% and 95% reproducible features, ADC map had 89% and 93%, D map had 97% and 86%, while f map had 54% and 62% reproducible features with good to excellent reliability in the intra-observer analysis. Similarly, D* map had 90% and 94%, ADC map had 85% and 70%, D map had 81% and 78%, while f map had 41% and 93% reproducible features with good to excellent reliability in the inter-observer analysis. Conclusion Inter- and intra-observer variability can affect radiomics analysis. Cognizant to this, multicenter studies should pay more attention to intra- and inter-observer variability.

scholarly journals The role of quantitative diffusion-weighted imaging in characterization of hypovascular liver lesions: A prospective comparison of intravoxel incoherent motion derived parameters and apparent diffusion coefficient

PLoS ONE ◽  
2021 ◽  
Vol 16 (2) ◽  
pp. e0247301
Author(s):  
Jelena Djokić Kovač ◽  
Marko Daković ◽  
Aleksandra Janković ◽  
Milica Mitrović ◽  
Vladimir Dugalić ◽  
...  
Keyword(s):  
Diffusion Coefficient ◽  
Apparent Diffusion Coefficient ◽  
Liver Metastases ◽  
Diffusion Weighted Imaging ◽  
Intraclass Correlation ◽  
Intravoxel Incoherent Motion ◽  
Liver Lesions ◽  
Diffusion Weighted ◽  
Significant Difference ◽  
Apparent Diffusion

Background The utility of intravoxel incoherent motion (IVIM) related parameters in differentiation of hypovascular liver lesions is still unknown. Purpose The purpose of this study was to evaluate the value of IVIM related parameters in comparison to apparent diffusion coefficient (ADC) for differentiation among intrahepatic mass-forming cholangiocarcinoma (IMC), and hypovascular liver metastases (HLM). Methods Seventy-four prospectively enrolled patients (21 IMC, and 53 HLM) underwent 1.5T magnetic resonance examination with IVIM diffusion-weighted imaging using seven b values (0–800 s/mm2). Two independent readers performed quantitative analysis of IVIM-related parameters and ADC. Interobserver reliability was tested using a intraclass correlation coefficient. ADC, true diffusion coefficient (D), perfusion-related diffusion coefficient (D*), and perfusion fraction (ƒ) were compared among the lesions using Kruskal-Wallis H test. The diagnostic accuracy of each parameter was assessed by receiver operating characteristic (ROC) curve analysis. Results The interobserver agreement was good for ADC (0.802), and excellent for D, D*, and ƒ (0.911, 0.927, and 0.942, respectively). ADC, and D values were significantly different among IMC and HLM (both p < 0.05), while there was no significant difference among these lesions for ƒ and D* (p = 0.101, and p = 0.612, respectively). ROC analysis showed higher diagnostic performance of D in comparison to ADC (AUC = 0.879 vs 0.821). Conclusion IVIM-derived parameters in particular D, in addition to ADC, could help in differentiation between most common hypovascular malignant liver lesions, intrahepatic mass—forming cholangiocarcinoma and hypovascular liver metastases.

Impact of different b-value combinations on radiomics features of apparent diffusion coefficient in cervical cancer

2019 ◽  
Vol 61 (4) ◽  
pp. 568-576
Author(s):  
Yaoyao He ◽  
Yi Rong ◽  
Hao Chen ◽  
Zhaoxi Zhang ◽  
Jianfeng Qiu ◽  
...  
Keyword(s):  
Cervical Cancer ◽  
Diffusion Coefficient ◽  
Apparent Diffusion Coefficient ◽  
B Value ◽  
Gray Level ◽  
Difference Matrix ◽  
Diffusion Weighted Images ◽  
Apparent Diffusion ◽  
Occurrence Matrix ◽  
The Impact

Background The impact of variable b-value combinations on apparent diffusion coefficient (ADC)-based radiomics features has not been fully addressed in literature. Purpose To investigate the correlation between radiomics features extracted from ADC maps and various b-value combinations in cervical cancer. Material and Methods Diffusion-weighted images (b-values: 0, 600, 800, and 1000 s/mm2) of 20 patients with cervical cancer were included. Tumors were identified with the largest transversal cross-section and manually segmented by radiologist. For each b-value combination, 92 radiomics features were extracted and coefficient of variance (CV) was used to evaluate the robustness of radiomics features with different b-value combinations. Features with CV > 5% were normalized by the mean feature variation across the group. Results Out of a total of 92 radiomics features, 18 were classified as robust features with CV ≤5%. Among the rest (CV > 5%), 11, 23, and 40 features demonstrated 5%< CV ≤10%, 10%< CV ≤20%, and CV > 20%, respectively. A subset of features in each category (CV > 5%) showed strong correlation with the b-value combination variation, including 44% (7/16) features in gray level co-occurrence matrix, 62% (8/13) features in gray level dependence matrix, 64% (9/14) features in first order, 50% (8/16) features in gray level run length matrix, 57% (8/14) features in gray level size matrix, and 20% (1/5) features in neighborhood gray-tone difference matrix. Conclusions Variations in b-value combinations demonstrated impact on radiomics features extracted from ADC maps for cervical cancer. The radiomics features with CV <5% can be considered as robust features and are recommended to be used in multicenter radiomics studies.

scholarly journals Renal Diffusion-Weighted Imaging in Healthy Dogs: Reproducibility, Test-Retest Repeatability, and Selection of the Optimal b-value Combination

2021 ◽  
Vol 8 ◽  
Author(s):  
Sang-Kwon Lee ◽  
Juryeoung Lee ◽  
Seolyn Jang ◽  
Eunji Lee ◽  
Chang-Yeop Jeon ◽  
...  
Keyword(s):  
Diffusion Coefficient ◽  
Diffusion Weighted Imaging ◽  
Intraclass Correlation ◽  
B Value ◽  
Adc Value ◽  
B Values ◽  
Intraobserver Reproducibility ◽  
Diffusion Weighted ◽  
Pure Diffusion ◽  
Healthy Dogs

Diffusion-weighted imaging (DWI) magnetic resonance imaging can evaluate alterations in the microstructure of the kidney. The purpose of this study was to assess the apparent diffusion coefficient (ADC) and the intravoxel incoherent motion model (IVIM) parameters of a normal kidney in healthy dogs, to evaluate the effect of b-value combinations on the ADC value, and the reproducibility and test-retest repeatability in monoexponential and IVIM analysis. In this experimental study, the ADC, pure diffusion coefficient (D), pseudodiffusion coefficient (D*), and perfusion fraction (fp) were measured from both kidneys in nine healthy beagles using nine b-values (b = 0, 50, 70, 100, 150, 200, 500, 800, and 1,000 s/mm2) twice with a 1-week interval between measurements. Interobserver and intraobserver reproducibility, and test-retest repeatability of the measurements were calculated. ADC values were measured using 10 different b-value combinations consisting of three b-values each, and were compared to the ADC obtained from nine b-values. All the ADC, D, D*, and fp values measured from the renal cortex, medulla, and the entire kidney had excellent interobserver and intraobserver reproducibility, and test-retest repeatability. The ADC obtained from a b-value combination of 0, 100, and 800 s/mm2 had the highest intraclass correlation coefficient with the ADC from nine b-values. The results of this study indicated that DWI MRI using multiple b-values is feasible for the measurement of ADC and IVIM parameters with high reproducibility and repeatability in the kidneys of healthy dogs. A combination of b = 0, 100, and 800 s/mm2 can be used for ADC measurements when multiple b-values are not available in dogs.

scholarly journals Semi-Automated Segmentation of Bone Metastases from Whole-Body MRI: Reproducibility of Apparent Diffusion Coefficient Measurements

Diagnostics ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. 499
Author(s):  
Alberto Colombo ◽  
Giulia Saia ◽  
Alcide A. Azzena ◽  
Alice Rossi ◽  
Fabio Zugni ◽  
...  
Keyword(s):  
Diffusion Coefficient ◽  
Apparent Diffusion Coefficient ◽  
Bone Metastases ◽  
Intraclass Correlation ◽  
Correlation Coefficients ◽  
Whole Body ◽  
Whole Body Mri ◽  
Diffusion Weighted Images ◽  
Diffusion Weighted ◽  
Apparent Diffusion

Using semi-automated software simplifies quantitative analysis of the visible burden of disease on whole-body MRI diffusion-weighted images. To establish the intra- and inter-observer reproducibility of apparent diffusion coefficient (ADC) measures, we retrospectively analyzed data from 20 patients with bone metastases from breast (BCa; n = 10; aged 62.3 ± 14.8) or prostate cancer (PCa; n = 10; aged 67.4 ± 9.0) who had undergone examinations at two timepoints, before and after hormone-therapy. Four independent observers processed all images twice, first segmenting the entire skeleton on diffusion-weighted images, and then isolating bone metastases via ADC histogram thresholding (ADC: 650–1400 µm2/s). Dice Similarity, Bland-Altman method, and Intraclass Correlation Coefficient were used to assess reproducibility. Inter-observer Dice similarity was moderate (0.71) for women with BCa and poor (0.40) for men with PCa. Nonetheless, the limits of agreement of the mean ADC were just ±6% for women with BCa and ±10% for men with PCa (mean ADCs: 941 and 999 µm2/s, respectively). Inter-observer Intraclass Correlation Coefficients of the ADC histogram parameters were consistently greater in women with BCa than in men with PCa. While scope remains for improving consistency of the volume segmented, the observer-dependent variability measured in this study was appropriate to distinguish the clinically meaningful changes of ADC observed in patients responding to therapy, as changes of at least 25% are of interest.

scholarly journals Evaluation of Regional Variability and Measurement Reproducibility of Intravoxel Incoherent Motion Diffusion Weighted Imaging Using a Cardiac Stationary Phase Based ECG Trigger Method

2018 ◽  
Vol 2018 ◽  
pp. 1-11 ◽  
Author(s):  
Zhiming Xiang ◽  
Zhu Ai ◽  
Jianke Liang ◽  
Guijin Li ◽  
Xiaolei Zhu ◽  
...  
Keyword(s):  
Diffusion Coefficient ◽  
Stationary Phase ◽  
Diffusion Weighted Imaging ◽  
Intravoxel Incoherent Motion ◽  
Regional Variability ◽  
Altman Plot ◽  
Bland Altman Plot ◽  
Hepatic Lobe ◽  
Diffusion Weighted ◽  
Pure Diffusion

Purpose. To evaluate the performance of an optimized ECG trigger diffusion weighted imaging (DWI) sequence in liver and its application in liver disease. Materials and Methods. Eighteen healthy volunteers underwent intravoxel incoherent motion diffusion weighted imaging (IVIM-DWI) scan of the liver twice in 1.5T MR scanner with signed informed consent approved by local ethic committees. A new method, called cardiac stationary phase based ECG trigger (CaspECG), and FB method were applied. The apparent diffusion coefficient (ADC) and the IVIM parameters, including pure diffusion coefficient (D), perfusion-related diffusion coefficient (D⁎), and perfusion fraction, (PF) were calculated, and then 18 region of interests were drawn on these parameter maps independently by two readers through whole hepatic lobe. The regional variability and reproducibility between two repeated scans were evaluated using interclass correlation coefficients (ICCs) and Bland-Altman plot, respectively, and compared between the CaspECG and FB methods. The signal-to-noise ratio (SNR) of DWI data was also evaluated. Result. Compared to the FB method, the proposed CaspECG method showed significant higher SNRs in DWI data, lower regional variability between left and right hepatic lobes, and higher reproducibility of ADC, PF, D, and D⁎ between repeat scans [left lobe, limit of agreement (LOA) of Bland-Altman plot: 10.1%, 18.3%, 19.8%, and 59.2%; right lobe, LOA: 10.25%, 14.15%, 16.45%, and 39.45%]. D⁎ showed the worst reproducibility in all parameters. Conclusion. The novel CaspECG method outperformed the FB method in compensating the cardiac motion induced artifacts in DWI data and generating more reliable quantitative parameters, with less regional variability and higher repeatability, especially in the left hepatic lobe.

Intravoxel incoherent motion diffusion-weighted imaging assessment of microvascular characteristics in the murine embryonal rhabdomyosarcoma model

2019 ◽  
Vol 61 (2) ◽  
pp. 260-266
Author(s):  
Yuan Yuan ◽  
Dewei Zeng ◽  
Yu Zhang ◽  
Juan Tao ◽  
Yajie Liu ◽  
...  
Keyword(s):  
Diffusion Coefficient ◽  
Microvessel Density ◽  
Diffusion Weighted Imaging ◽  
Embryonal Rhabdomyosarcoma ◽  
Intravoxel Incoherent Motion ◽  
Microvascular Blood Flow ◽  
Vegf Expression ◽  
Perfusion Fraction ◽  
Diffusion Weighted ◽  
D Value

Background Intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) can distinguish the false diffusion generated by microvascular blood flow from the true water molecule diffusion. Purpose To investigate the correlation between IVIM-DWI parameters and angiogenic markers such as the microvessel density and vascular endothelial growth factor (VEGF) expression in the murine embryonal rhabdomyosarcoma model. Material and Methods The murine embryonal rhabdomyosarcoma model was produced by subcutaneously injecting 107 human embryonal rhabdomyosarcoma cells into the right back of nude mice. The apparent diffusion coefficient (ADC), pseudo-diffusion coefficient (D*), true diffusion coefficient (D), and perfusion fraction (f) were obtained from 22 mice models using IVIM-DWI with b-values of 0, 50, 100, 150, 200, 400, 600, 800, 1000, and 1200 s/mm2. The microvessel density and VEGF expression were obtained by histologic examination. We then compared the correlation between IVIM-DWI parameters and microvessel density and VEGF expression. Results The average ADC, D*, D, and f values were 1.05 ± 0.27 × 10−3 mm2/s, 6.19 ± 1.78 × 10−3 mm2/s, 0.69 ± 0.09 ×10−3 mm2/s, and 17.68 ± 8.41 (%), respectively. There was moderate positive correlation between D* value and microvessel density and VEGF expression (r = 0.484, P = 0.023; r = 0.511, P = 0.015). However, there was no significant correlation between ADC, D, and f values and microvessel density and VEGF expression. Conclusion The D* value from IVIM-DWI may be used to evaluate tumor angiogenesis in the murine embryonal rhabdomyosarcoma model.

Sign in / Sign up

Export Citation Format

Share Document