Staging liver fibrosis by a continuous-time random-walk diffusion model

PURPOSE

Noninvasive assessment of liver fibrosis holds significant clinical importance. We aimed to evaluate the clinical potential of using a continuous-time random-walk diffusion model (CTRW) for staging liver fibrosis.

METHODS

This prospective study included 52 patients suspected of liver disease and scheduled for liver biopsy. All patients underwent multi-b value diffusion-weighted imaging (DWI) using a 1.5 T MR scanner to derive the anomalous diffusion coefficient (D) and temporal (α) and spatial (β) diffusion heterogeneity indexes sourced from the CTRW. The mono-exponential DWI-derived apparent diffusion coefficient (ADC), transient elastography-derived liver stiffness measurement (LSM), aspartate aminotransferase-to-platelet ratio index (APRI), and fibrosis-4 (FIB-4) index were calculated. We assessed and compared the correlations of these parameters with fibrosis stages and their efficacy in staging liver fibrosis.

RESULTS

Significant correlations with fibrosis stages were found for APRI (r = 0.336), FIB-4 (r = 0.351), LSM (r = 0.523), D (r = -0.458), and ADC (r = -0.473). Significant differences were observed between APRI, LSM, D, and ADC of different fibrosis stages. The diagnostic performance of an index that combined D, α, β, ADC, and LSM was superior to that of ADC or LSM alone for fibrosis stage F ≥ 2 and better than the index that combined D, α, β for fibrosis stage F ≥ 4.

CONCLUSIONS

Accurate liver fibrosis staging was achieved with a model that combined CTRW-derived parameters (D, α, and β), ADC, and LSM. The model could serve as a reliable tool for noninvasive fibrosis evaluation.

Assessment of chronic allograft injury in renal transplantation using diffusional kurtosis imaging

Background

Chronic allograft injury (CAI) is a significant reason for which many grafts were lost. The study was conducted to assess the usefulness of diffusional kurtosis imaging (DKI) technology in the non-invasive assessment of CAI.

Methods

Between February 2019 and October 2019, 110 renal allograft recipients were included to analyze relevant DKI parameters. According to estimated glomerular filtration rate (eGFR) (mL/min/ 1.73 m²) level, they were divided to 3 groups: group 1, eGFR ≥ 60 (n = 10); group 2, eGFR 30–60 (n = 69); group 3, eGFR < 30 (n = 31). We performed DKI on a clinical 3T magnetic resonance imaging system. We measured the area of interest to determine the mean kurtosis (MK), mean diffusivity (MD), and apparent diffusion coefficient (ADC) of the renal cortex and medulla. We performed a Pearson correlation analysis to determine the relationship between eGFR and the DKI parameters. We used the receiver operating characteristic curve to estimate the predicted values of DKI parameters in the CAI evaluation. We randomly selected five patients from group 2 for biopsy to confirm CAI.

Results

With the increase of creatinine, ADC, and MD of the cortex and medulla decrease, MK of the cortex and medulla gradually increase. Among the three different eGFR groups, significant differences were found in cortical and medullary MK (P = 0.039, P < 0.001, P < 0.001, respectively). Cortical and medullary ADC and MD are negatively correlated with eGFR (r = − 0.49, − 0.44, − 0.57, − 0.57, respectively; P < 0.001), while cortical and medullary MK are positively correlated with eGFR (r = 0.42, 0.38; P < 0.001). When 0.491 was set as the cutoff value, MK's CAI assessment showed 87% sensitivity and 100% specificity. All five patients randomly selected for biopsy from the second group confirmed glomerulosclerosis and tubular atrophy/interstitial fibrosis.

Conclusion

The DKI technique is related to eGFR as allograft injury progresses and is expected to become a potential non-invasive method for evaluating CAI.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12880-021-00595-3.

The volumetric-tumour histogram-based analysis of intravoxel incoherent motion and non-Gaussian diffusion MRI: association with prognostic factors in HER2-positive breast cancer

Background

To evaluate the imaging biomarkers of human epidermal growth factor receptor 2 (HER2) positive breast cancer in comparison to other molecular subtypes and to determine the feasibility of identifying hormone receptor (HR) status and lymph node metastasis status using volumetric-tumour histogram-based analysis through intravoxel incoherent motion (IVIM) and non-Gaussian diffusion.

Methods

This study included 145 breast cancer patients with 148 lesions between January and November in 2018. Among the 148 lesions, 74 were confirmed to be HER2-positive. The volumetric-tumour histogram-based features were extracted from the combined IVIM and non-Gaussian diffusion model. IVIM and non-Gaussian diffusion parameters obtained from images of the subjects with different molecular prognostic biomarker statuses were compared by Student’s t test or the Mann–Whitney U test. The area under the curve (AUC), sensitivity, and specificity at the best cut-off point were reported. The Spearman correlation coefficient was calculated to analyse the correlations of clinical tumor nodule metastasis (TNM) stage and Ki67 with the IVIM and non-Gaussian diffusion parameters.

Results

The entropy of mean kurtosis (MK) was significantly higher in the HER2-positive group than in the HER2-negative group (p = 0.015), with an AUC of 0.629 (95% CI 0.546, 0.707), a sensitivity of 62.6%, and a specificity of 66.2%. For HR status, the MD 5th percentile was higher in the HR-positive group of HER2-positive breast cancer (p = 0.041), with an AUC of 0.643 (95% CI 0.523, 0.751), while for lymph node status, the entropy of mean diffusivity (MK) was lower in the lymph node positive group (p = 0.040), with an AUC of 0.587 (95% CI 0.504, 0.668). The clinical TNM stage and Ki67 index were correlated with several histogram parameters.

Conclusion

Volumetric-lesion histogram analysis of IVIM and the non-Gaussian diffusion model can be used to provide prognostic information about HER2-positive breast cancers and potentially contribute to individualized anti-HER2 targeted therapy plans .

Electronic supplementary material

The online version of this article (10.1186/s12967-019-1911-6) contains supplementary material, which is available to authorized users.

Histogram analysis of diffusion kurtosis imaging estimates for in vivo assessment of 2016 WHO glioma grades: A cross-sectional observational study

PURPOSE

To assess the diagnostic performance of histogram analysis of diffusion kurtosis imaging (DKI) maps for in vivo assessment of the 2016 World Health Organization Classification of Tumors of the Central Nervous System (2016 CNS WHO) integrated glioma grades.

MATERIALS AND METHODS

Seventy-seven patients with histopathologically-confirmed glioma who provided written informed consent were retrospectively assessed between 01/2014 and 03/2017 from a prospective trial approved by the local institutional review board. Ten histogram parameters of mean kurtosis (MK) and mean diffusivity (MD) metrics from DKI were independently assessed by two blinded physicians from a volume of interest around the entire solid tumor. One-way ANOVA was used to compare MK and MD histogram parameter values between 2016 CNS WHO-based tumor grades. Receiver operating characteristic analysis was performed on MK and MD histogram parameters for significant results.

RESULTS

The 25th, 50th, 75th, and 90th percentiles of MK and average MK showed significant differences between IDH1/2_wild-type gliomas, IDH1/2_mutated gliomas, and oligodendrogliomas with chromosome 1p/19q loss of heterozygosity and IDH1/2_mutation (p<0.001). The 50th, 75th, and 90th percentiles showed a slightly higher diagnostic performance (area under the curve (AUC) range; 0.868-0.991) than average MK (AUC range; 0.855-0.988) in classifying glioma according to the integrated approach of 2016 CNS WHO.

CONCLUSIONS

Histogram analysis of DKI can stratify gliomas according to the integrated approach of 2016 CNS WHO. The 50th (median), 75th_, and the 90th percentiles showed the highest diagnostic performance. However, the average MK is also robust and feasible in routine clinical practice.

Directional sensitivity of anomalous diffusion in human brain assessed by tensorial fractional motion model

Anisotropic diffusion in the nervous system is most commonly modeled by apparent diffusion tensor, which is based on regular diffusion theory. However, the departure of diffusion-induced signal attenuation from a mono-exponential form implies that there is anomalous diffusion. Recently, a novel diffusion NMR theory based on the fractional motion (FM) model, which is an anomalous diffusion model, has been proposed. While the FM model has been applied to both healthy subjects and tumor patients, its anisotropy in the nervous system remains elusive. In this study, this issue was addressed by measuring the FM-related parameters in 12 non-collinear directions. A metric to quantify the directional deviation was derived. Furthermore, the FM-related parameters were modeled as tensors and analyzed in analogy with the conventional diffusion tensor imaging (DTI). Experimental results, which were obtained for 15 healthy subjects at 3T, exhibited pronounced anisotropy of the FM-related parameters, although the effects were smaller than the apparent diffusion coefficient (ADC). The tensorial nature for α, which is the Noah exponent in the FM model, showed behavior similar to the ADC, especially the principal eigenvector for α aligned with the dominant white matter fiber directions. The Hurst exponent H in the FM model, however, showed no correlation with the major fiber directions. The anisotropy of the FM model may provide complementary information to DTI and may have potential for tractography and detecting brain abnormalities.

Evaluation of optimized b-value sampling schemas for diffusion kurtosis imaging with an application to stroke patient data

Diffusion kurtosis imaging (DKI) is a new method of magnetic resonance imaging (MRI) that provides non-Gaussian information that is not available in conventional diffusion tensor imaging (DTI). DKI requires data acquisition at multiple b-values for parameter estimation; this process is usually time-consuming. Therefore, fewer b-values are preferable to expedite acquisition. In this study, we carefully evaluated various acquisition schemas using different numbers and combinations of b-values. Acquisition schemas that sampled b-values that were distributed to two ends were optimized. Compared to conventional schemas using equally spaced b-values (ESB), optimized schemas require fewer b-values to minimize fitting errors in parameter estimation and may thus significantly reduce scanning time. Following a ranked list of optimized schemas resulted from the evaluation, we recommend the 3b schema based on its estimation accuracy and time efficiency, which needs data from only 3 b-values at 0, around 800 and around 2600 s/mm2, respectively. Analyses using voxel-based analysis (VBA) and region-of-interest (ROI) analysis with human DKI datasets support the use of the optimized 3b (0, 1000, 2500 s/mm2) DKI schema in practical clinical applications.