Magnetic resonance imaging最新文献

Alterations in white matter (WM) microstructure of the central nervous system have been shown to be pathophysiological presentations of various neurodegenerative disorders. Current methods for measuring such WM features require ex vivo tissue samples analyzed using electron microscopy. Magnetic Resonance Imaging (MRI) diffusion-weighted pulse sequences provide a non-invasive tool for estimating such microstructural features in vivo. The current project investigated the use of two methods of analysis, including the ROI-based (Region of Interest, RBA) and voxel-based analysis (VBA), as well as four mathematical models of WM microstructure, including the ActiveAx Frequency-Independent Extra-Axonal Diffusion (AAI), ActiveAx Frequency-Dependent Extra-Axonal Diffusion (AAD), AxCaliber Frequency-Independent Extra-Axonal Diffusion (ACI), and AxCaliber Frequency-Dependent Extra-Axonal Diffusion (ACD) models. Two mice samples imaged at 7 T and 15.2 T were analyzed. Both the AAI and AAD models provide a single value for each of the fit parameters, including mean effective axon diameter $\overline{\mathrm{AxD}}$, packing fraction $f_{\mathrm{in}}$, intra-cellular and $D_{\mathrm{in}}$ and extra-cellular $D_{\mathrm{ex}}$ diffusion coefficients, as well as the frequency dependence of $D_{\mathrm{ex}}$, ${\beta }_{\mathrm{ex}}$ for the AAD model. The ACI and ACD models provide this, in addition to a distribution of axon diameters for a chosen ROI. VBA extends this, providing a parameter value for each voxel within the selected ROI, at the cost of increased computational load and analysis time. Overall, RBA-ACD and VBA-AAD were found to be optimal for parameter fitting to physically relevant values in a reasonable time frame. A full comparison of each combination of RBA and VBA with AAI, AAD, ACI, and ACD is provided to give the reader sufficient information to make an informed decision of which model is best for their own experiments.

The reconstruction of dynamic magnetic resonance images from incomplete k-space data has sparked significant research interest due to its potential to reduce scan time. However, traditional iterative optimization algorithms fail to faithfully reconstruct images at higher acceleration factors and incur long reconstruction time. Furthermore, end-to-end deep learning-based reconstruction algorithms suffer from large model parameters and lack robustness in the reconstruction results. Recently, unrolled deep learning models, have shown immense potential in algorithm stability and applicability flexibility. In this paper, we propose an unrolled deep learning network based on a second-order Half-Quadratic Splitting(HQS) algorithm, where the forward propagation process of this framework strictly follows the computational flow of the HQS algorithm. In particular, we propose a degradation-sense module by associating random sampling patterns with intermediate variables to guide the iterative process. We introduce the Information Fusion Transformer(IFT) to extract both local and non-local prior information from image sequences, thereby removing aliasing artifacts resulting from random undersampling. Finally, we impose low-rank constraints within the HQS algorithm to further enhance the reconstruction results. The experiments demonstrate that each component module of our proposed model contributes to the improvement of the reconstruction task. Our proposed method achieves comparably satisfying performance to the state-of-the-art methods and it exhibits excellent generalization capabilities across different sampling masks. At the low acceleration factor, there is a 0.7% enhancement in the PSNR. Furthermore, when the acceleration factor reached 8 and 12, the PSNR achieves an improvement of 3.4% and 5.8% respectively.

Purpose

This study assessed the clinical classification performance of an R2*-based MRI technique for LIC quantification relative to FerriScan, with intra-patient FerriScan LIC uncertainty taken into account. The variabilities of R2* and FerriScan LIC were also assessed.

Materials and methods

This was an ethics approved retrospective study, performed on patients undergoing chelation treatment for iron overload. 126 patients (69 women, 57 men), with an age of 42 +/− 16 years (range 19–86 years) were included. FerriScan and R2* MRI at 1.5 T were performed as part of a routine liver iron assessment protocol. For R2* MRI, a commercially available pulse sequence and reconstruction implementation was used, together with a previously derived calibration curve to convert R2* to LIC. Clinical classifications arising from R2*-derived LIC estimates were compared to those based on FerriScan. The accuracy and precision of the R2* technique was calculated. The variabilities of FerriScan- and R2*-derived estimates of LIC were compared with a Wilcoxon Signed Rank test. Significance was set at the 95% confidence level.

Results

The precision of R2* ranged from 0.59 to 0.92, with an overall accuracy of 72%. When intra-patient FerriScan LIC uncertainty was considered, precision and accuracy increased to >94% and 97% respectively. The R2*-LIC variability (=17%) was significantly lower than the FerriScan-LIC variability (34%) at the 95% confidence level (p < 10⁻³).

Conclusion

MRI R2*-based LIC estimates provided a similar clinical classification as FerriScan. The intra-patient uncertainty of R2*-based LIC estimates was significantly lower than FerriScan.

This study investigated the use of a Vision Transformer (ViT) for reconstructing GABA-edited Magnetic Resonance Spectroscopy (MRS) data from a reduced number of transients. Transients refer to the samples collected during an MRS acquisition by repeating the experiment to generate a signal of sufficient quality. Specifically, 80 transients were used instead of the typical 320 transients, aiming to reduce scan time. The 80 transients were pre-processed and converted into a spectrogram image representation using the Short-Time Fourier Transform (STFT). A pre-trained ViT, named Spectro-ViT, was fine-tuned and then tested using in-vivo GABA-edited MEGA-PRESS data. Its performance was compared against other pipelines in the literature using quantitative quality metrics and estimated metabolite concentration values, with the typical 320-transient scans serving as the reference for comparison. The Spectro-ViT model exhibited the best overall quality metrics among all other pipelines against which it was compared. The metabolite concentrations from Spectro-ViT's reconstructions for GABA+ achieved the best average R² value of 0.67 and the best average Mean Absolute Percentage Error (MAPE) value of 9.68%, with no significant statistical differences found compared to the 320-transient reference. The code to reproduce this research is available at https://github.com/MICLab-Unicamp/Spectro-ViT

Radiomics of cardiac magnetic resonance (MR) imaging has proved to be potentially useful in the study of various myocardial diseases. Therefore, assessing the repeatability degree in radiomic features measurement is of fundamental importance.

The aim of this study was to assess test-retest repeatability of myocardial radiomic features extracted from quantitative T1 and T2 maps.

A representative group of 24 subjects (mean age 54 ± 18 years) referred for clinical cardiac MR imaging were enrolled in the study. For each subject, T1 and T2 mapping through MOLLI and T2-prepared TrueFISP acquisition sequences, respectively, were performed at 1.5 T. Then, 98 radiomic features of different classes (shape, first-order, second-order) were extracted from a region of interest encompassing the whole left ventricle myocardium in a short axis slice. The repeatability was assessed performing different and complementary analyses: intraclass correlation coefficient (ICC) and limits of agreement (LOA) (i.e., the interval within which 95% of the percentage differences between two repeated measures are expected to lie).

Radiomic features were characterized by a relatively wide range of repeatability degree in terms of both ICC and LOA. Overall, 44.9% and 38.8% of radiomic features showed ICC values > 0.75 for T1 and T2 maps, respectively, while 25.5% and 23.4% of radiomic features showed LOA between ±10%. A subset of radiomic features for T1 (Mean, Median, 10Percentile, 90Percentile, RootMeanSquared, Imc2, RunLengthNonUniformityNormalized, RunPercentage and ShortRunEmphasis) and T2 (MaximumDiameter, RunLengthNonUniformityNormalized, RunPercentage, ShortRunEmphasis) maps presented both ICC > 0.75 and LOA between ±5%.

Overall, radiomic features extracted from T1 maps showed better repeatability performance than those extracted from T2 maps, with shape features characterized by better repeatability than first-order and textural features. Moreover, only a limited subset of 9 and 4 radiomic features for T1 and T2 maps, respectively, showed high repeatability degree in terms of both ICC and LOA. These results confirm the importance of assessing test-retest repeatability degree in radiomic feature estimation and might be useful for a more effective/reliable use of myocardial T1 and T2 mapping radiomics in clinical or research studies.

Objective

The objective of this study was to investigate the correlation between serum levels of matrix metalloproteinase-2 (MMP-2), matrix metalloproteinase-9 (MMP-9), and tissue inhibitor of metalloproteinases-1 (TIMP-1) levels and their ratios with the severity of white matter hyperintensities (WMHs) in patients with cerebral small vessel disease (CSVD).

Methods

This cross-sectional study was done on a prospective cohort of patients with CSVD. Qualitative and quantitative analyses of WMHs were performed using Fazekas grading and lesion prediction algorithm (LPA) methods. Biomarkers MMP-2, MMP-9, and TIMP-1 were measured to explore their correlation with the severity of WMHs.

Results

The sample consisted of 144 patients with CSVD. There were 63 male and 81 female patients, with an average age of 67.604 ± 8.727 years. Among these, 58.33% presented with white matter hyperintensities at Fazekas grading level 1, with an average total template volume of WMHs of 4.305 mL. MMP-2 (P = 0.025), MMP-9 (P = 0.008), TIMP-1 (P = 0.026), and age (P = 0.007) were identified as independent correlates of WMHs based on Fazekas grading. Independent correlates of the total template volume of WMHs included MMP-2 (P = 0.023), TIMP-1 (P = 0.046), age (P = 0.047), systolic blood pressure (P = 0.047), and homocysteine (Hcy) (P = 0.014). In addition, age (P = 0.003; P < 0.001), interleukin-6 (IL-6) (P < 0.001; P = 0.044), Hcy (P < 0.001; P < 0.001), glycated hemoglobin (HbA1c) (P = 0.016; P = 0.043), and chronic kidney disease (P < 0.001; P < 0.001) were associated with both WMHs Fazekas grading and the total template volume of WMHs.

Conclusion

Serum levels of MMP-9, MMP-2, and TIMP-1 were independently associated with the Fazekas grading, while serum TIMP-1 and MMP-2 levels were independently related to the total template volume of WMHs. The association of TIMP-1 and MMP-2 with the severity of CSVD-related WMHs suggests their potential role as disease-related biomarkers. However, further research is required to uncover the specific mechanisms underlying these interactions.

Objective

The research aimed to determine whether and which radiomic features from breast dynamic contrast enhanced (DCE) MRI could predict the presence of BRCA1 mutation in patients with triple-negative breast cancer (TNBC).

Material and methods

This retrospective study included consecutive patients histologically diagnosed with TNBC who underwent breast DCE-MRI in 2010–2021. Baseline DCE-MRIs were retrospectively reviewed; percentage maps of wash-in and wash-out were computed and breast lesions were manually segmented, drawing a 5 mm-Region of Interest (ROI) inside the tumor and another 5 mm-ROI inside the contralateral healthy gland. Features for each map and each ROI were extracted with Pyradiomics-3D Slicer and considered first separately (tumor and contralateral gland) and then together. In each analysis the more important features for BRCA1 status classification were selected with Maximum Relevance Minimum Redundancy algorithm and used to fit four classifiers.

Results

The population included 67 patients and 86 lesions (21 in BRCA1-mutated, 65 in non BRCA-carriers). The best classifiers for BRCA mutation were Support Vector Classifier and Logistic Regression in models fitted with both gland and tumor features, reaching an Area Under ROC Curve (AUC) of 0.80 (SD 0.21) and of 0.79 (SD 0.20), respectively. Three features were higher in BRCA1-mutated compared to non BRCA-mutated: Total Energy and Correlation from gray level cooccurrence matrix, both measured in contralateral gland in wash-out maps, and Root Mean Squared, selected from the wash-out map of the tumor.

Conclusions

This study showed the feasibility of a radiomic study with breast DCE-MRI and the potential of radiomics in predicting BRCA1 mutational status.

Purpose

The aim of this study was to evaluate the diagnostic accuracy of the B1 inhomogeneity-corrected variable flip angle (VFA) method using native T1 values in the staging of liver fibrosis.

Methods

Eighty-three patients who presented for liver biopsy due to varying degrees of liver damage, underwent MR examinations and had T1-mapping images of the liver acquired using the B1 inhomogeneity-corrected VFA VIBE method. Among them, 65 patients underwent Fibroscan, and their results were used to evaluate the elasticity of liver tissue. Additionally, T1-mapping images were collected from 19 normal control patients. Independent sample t-tests were used to analyze the correlation between T1 mapping and Fibroscan. The diagnostic efficacy of T1 mapping in patients with different stages of liver fibrosis was evaluated using receiver operating characteristic (ROC) curves.

Results

The consistency between different observer groups was intraclass correlation coefficient (ICC) =0.802. T1 mapping demonstrated significant differences between mid-stage liver fibrosis (S = 2) and late-stage liver fibrosis (S = 3), as well as moderate inflammation (G = 2) and severe inflammation (G = 3), P < 0.05. The Area Under Curve(AUC) values of T1 mapping for early liver fibrosis (S ≥ 1), significant liver fibrosis (S ≥ 2), advanced liver fibrosis (S ≥ 3), and end-stage liver fibrosis (S = 4) were 0.760, 0.709, 0.790, and 0.768, respectively. T1 mapping combined with Fibroscan had an AUC value of 0.860.

Conclusions

The B1 inhomogeneity-corrected VFA T1 mapping may be useful for the staging of liver fibrosis. It has a superior diagnostic efficiency for diagnosing advanced fibrosis (≥S3), while native T1 values combined with Fibroscan have potential value for the staging of liver fibrosis.

Objectives

This study aims to generate post-contrast MR images reducing the exposure of gadolinium-based contrast agents (GBCAs) for brainstem glioma (BSG) detection, simultaneously delineating the BSG lesion, and providing high-resolution contrast information.

Methods

A retrospective cohort of 30 patients diagnosed with brainstem glioma was included. Multi-contrast images, including pre-contrast T1 weighted (pre-T1w), T2 weighted (T2w), arterial spin labeling (ASL) and post-contrast T1w images, were collected. A multi-task generative model was developed to synthesize post-contrast T1w images and simultaneously segment BSG masks from the multi-contrast inputs. Performance evaluation was conducted using peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and mean absolute error (MAE) metrics. A perceptual study was also undertaken to assess diagnostic quality.

Results

The proposed model achieved SSIM of 0.86 ± 0.04, PSNR of 26.33 ± 0.05 and MAE of 57.20 ± 20.50 for post-contrast T1w image synthesis. Automated delineation of the BSG lesions achieved Dice similarity coefficient (DSC) score of 0.88 ± 0.27.

Conclusions

The proposed model can synthesize high-quality post-contrast T1w images and accurately segment the BSG region, yielding satisfactory DSC scores.

Clinical relevance statement

The synthesized post-contrast MR image presented in this study has the potential to reduce the usage of gadolinium-based contrast agents, which may pose risks to patients. Moreover, the automated segmentation method proposed in this paper aids radiologists in accurately identifying the brainstem glioma lesion, facilitating the diagnostic process.

Objectives

To investigate the association of quantitative parameter (apparent diffusion coefficient [ADC]) from diffusion-weighted imaging (DWI) and various quantitative and semiquantitative parameters from dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) with Ki-67 proliferation index (PI) in cervical carcinoma (CC).

Methods

A total of 102 individuals with CC who received 3.0 T MRI examination (DWI and DCE MRI) between October 2016 and December 2022 were enrolled in our investigation. Two radiologists separately assessed the ADC parameter and various quantitative and semiquantitative parameters including (volume transfer constant [K^trans], rate constant [k_ep], extravascular extracellular space volume fraction [v_e], volume fraction of plasma [v_p], time to peak [TTP], maximum concentration [MaxCon], maximal slope [MaxSlope] and area under curve [AUC]) for each tumor. Their association with Ki-67 PI was analyzed by Spearman association analysis. The discrepancy between low-proliferation and high-proliferation groups was subsequently analyzed. The receiver operating characteristic (ROC) curve analysis utilized to identify optimal cut-off points for significant parameters.

Results

Both ADC (ρ = − 0.457, p < 0.001) and K^trans (ρ = − 0.467, p < 0.001) indicated a strong negative association with Ki-67 PI. Ki-67 PI showed positive correlations with TTP, MaxCon, MaxSlope and AUC (ρ = 0.202, 0.231, 0.309, 0.235, respectively; all p values<0.05). Compared with the low-proliferation group, high-Ki-67 group presented a significantly lower ADC (0.869 ± 0.125 × 10⁻³ mm²/s vs. 1.149 ± 0.318 × 10⁻³ mm²/s; p < 0.001) and K^trans (1.314 ± 1.162 min⁻¹vs. 0.391 ± 0.390 min⁻¹; p < 0.001), also significantly higher MaxCon values (0.756 ± 0.959 vs. 0.422 ± 0.341; p < 0.05) and AUC values (2.373 ± 3.012 vs. 1.273 ± 1.000; p < 0.05). The cut-offs of ADC, K^trans, MaxCon and AUC for discrimating low- and high-Ki-67 groups were 0.920 × 10⁻³ mm²/s, 0.304 min⁻¹, 0.209 and 1.918, respectively.

Conclusions

ADC, K^trans, TTP, MaxCon, MaxSlope and AUC are associated with Ki-67 PI. ADC and K^trans exhibited high performance to discriminate low and high Ki-67 status of CC.