Various MRI sequences have shown their potential to discriminate parotid gland tumors, including but not limited to T2‐weighted, postcontrast T1‐weighted, and diffusion‐weighted images. In this ...study, we present a fully automatic system for the diagnosis of parotid gland tumors by using deep learning methods trained on multimodal MRI images. We used a two‐dimensional convolution neural network, U‐Net, to segment and classify parotid gland tumors. The U‐Net model was trained with transfer learning, and a specific design of the batch distribution optimized the model accuracy. We also selected five combinations of MRI contrasts as the input data of the neural network and compared the classification accuracy of parotid gland tumors. The results indicated that the deep learning model with diffusion‐related parameters performed better than those with structural MR images. The performance results (n = 85) of the diffusion‐based model were as follows: accuracy of 0.81, 0.76, and 0.71, sensitivity of 0.83, 0.63, and 0.33, and specificity of 0.80, 0.84, and 0.87 for Warthin tumors, pleomorphic adenomas, and malignant tumors, respectively. Combining diffusion‐weighted and contrast‐enhanced T1‐weighted images did not improve the prediction accuracy. In summary, the proposed deep learning model could classify Warthin tumor and pleomorphic adenoma tumor but not malignant tumor.
We used a two‐dimensional convolution neural network to conduct the segmentation and classifications of parotid gland tumors. The results indicated that the deep learning model with diffusion‐related parameters performed better than those with structural MR images. In summary, the deep learning model presented the potential to classify Warthin tumor and pleomorphic adenoma tumor, but not malignant tumor.
Recently, intravoxel incoherent motion (IVIM) diffusion‐weighted imaging (DWI) has also been demonstrated as an imaging tool for applications in neurological and neurovascular diseases. However, the ...use of single‐shot diffusion‐weighted echo‐planar imaging for IVIM DWI acquisition leads to suboptimal data quality: for instance, geometric distortion and deteriorated image quality at high spatial resolution. Although the recently commercialized multi‐shot acquisition methods, such as multiplexed sensitivity encoding (MUSE), can attain high‐resolution and high‐quality DWI with signal‐to‐noise ratio (SNR) performance superior to that of the conventional parallel imaging method, the prolonged scan time associated with multi‐shot acquisition is impractical for routine IVIM DWI. This study proposes an acquisition and reconstruction framework based on parametric‐POCSMUSE to accelerate the four‐shot IVIM DWI with 70% reduction of total scan time (13 min 8 s versus 4 min 8 s). First, the four‐shot IVIM DWI scan with 17 b values was accelerated by acquiring only one segment per b value except for b values of 0 and 600 s/mm2. Second, an IVIM‐estimation scheme was integrated into the parametric‐POCSMUSE to enable joint reconstruction of multi‐b images from under‐sampled four‐shot IVIM DWI data. In vivo experiments on both healthy subjects and patients show that the proposed framework successfully produced multi‐b DW images with significantly higher SNRs and lower reconstruction errors than did the conventional acceleration method based on parallel imaging. In addition, the IVIM quantitative maps estimated from the data produced by the proposed framework showed quality comparable to that of fully sampled MUSE‐reconstructed images, suggesting that the proposed framework can enable highly accelerated multi‐shot IVIM DWI without sacrificing data quality. In summary, the proposed framework can make multi‐shot IVIM DWI feasible in a routine MRI examination, with reasonable scan time and improved geometric fidelity.
The proposed reconstruction framework based on parametric‐POCSMUSE can enable 3.2‐fold scan acceleration compared with four‐shot IVIM DWI with MUSE. In addition to being able to produce multi‐b DW images with less distortion, it can also derive the IVIM quantitative maps with quality comparable to that of fully sampled four‐shot IVIM DWI data.
Background
Adipose tissue is closely related to bone mass, bone quality, and bone fractures, but the connection between fat and bone is complex and gender‐related. Fat–water magnetic resonance ...imaging (MRI) and MR spectroscopy (MRS) are very useful tools for identifying tissue fat.
Purpose
To assess gender interactions between bone mineral density (BMD), bone marrow fat, and body mass index (BMI) in the elderly using fat–water MRI and MRS.
Study Type
Prospective/cohort.
Population
Sixty‐six women and 38 men (mean age, 62.3 years; range, 50–75 years), Asian.
Field Strength
A 1.5T MR equipped with a body and spine array coil. STEAM MRS and T2* Dixon were performed.
Assessment
Vertebral bone marrow fat ratio (MFR), BMI, and BMD were measured. Correlations between these variables and differences in bone density in MFR were assessed between participants, divided into three groups based on bone density.
Statistical Tests
Multiple regression; Pearson tests; analysis of covariance; analysis of variance.
Results
Multiple regression analysis identified gender, vertebral bone MFR, and BMI as significant predictors of vertebral BMD (P < 0.001). Among the women, vertebral BMD was negatively correlated with vertebral MFR (P = 0.011), but among the men, it was positively correlated with BMI (P = 0.048), although this relationship was confounded by age and MFR. Moreover, vertebral bone marrow fat and BMI were indeed statistically uncorrelated in the elderly (P = 0.357 in women; P = 0.961 in men).
Data Conclusion
We found gender interactions between fat and bone in the elderly. Higher bone marrow fat was correlated with lower trabecular BMD in older women but not in men. On the other hand, the positive correlation between BMI and BMD was more pronounced in men than in women.
Level of Evidence: 2
Technical Efficacy Stage: 2
J. Magn. Reson. Imaging 2020;51:1382–1389.
Background
Single‐shot diffusion‐weighted imaging (ssDWI) has been shown useful for detecting active bowel inflammation in Crohn's disease (CD) without MRI contrast. However, ssDWI suffers from ...geometric distortion and low spatial resolution.
Purpose
To compare conventional ssDWI with higher‐resolution ssDWI (HR‐ssDWI) and multi‐shot DWI based on multiplexed sensitivity encoding (MUSE‐DWI) for evaluating bowel inflammation in CD, using contrast‐enhanced MR imaging (CE‐MRI) as the reference standard.
Study Type
Prospective.
Subjects
Eighty nine patients with histological diagnosis of CD from previous endoscopy (55 male/34 female, age: 17–69 years).
Field Strength/Sequences
ssDWI (2.7 mm × 2.7 mm), HR‐ssDWI (1.8 mm × 1.8 mm), MUSE‐DWI (1.8 mm × 1.8 mm) based on echo‐planar imaging, T2‐weighted imaging, and CE‐MRI sequences, all at 1.5 T.
Assessment
Five raters independently evaluated the tissue texture conspicuity, geometry accuracy, minimization of artifacts, diagnostic confidence, and overall image quality using 5‐point Likert scales. The diagnostic performance (sensitivity, specificity and accuracy) of each DWI sequences was assessed on per‐bowel‐segment basis.
Statistical Tests
Inter‐rater agreement for qualitative evaluation of each parameter was measured by the intra‐class correlation coefficient (ICC). Paired Wilcoxon signed‐rank tests were performed to evaluate the statistical significance of differences in qualitative scoring between DWI sequences. A P value <0.05 was considered to be statistically significant.
Results
Tissue texture conspicuity, geometric distortions, and overall image quality were significantly better for MUSE‐DWI than for ssDWI and HR‐ssDWI with good agreement among five raters (ICC: 0.70–0.89). HR‐ssDWI showed significantly poorer performance to ssDWI and MUSE‐DWI for all qualitative scores and had the worst diagnostic performance (sensitivity of 57.0% and accuracy of 87.3%, with 36 undiagnosable cases due to severe artifacts). MUSE‐DWI showed significantly higher sensitivity (97.5% vs. 86.1%) and accuracy (98.9% vs. 95.1%) than ssDWI for detecting bowel inflammation.
Data Conclusion
MUSE‐DWI was advantageous in assessing bowel inflammation in CD, resulting in improved spatial resolution and image quality.
Level of Evidence
2
Technical Efficacy Stage
2
In this study, the performance of machine learning in classifying parotid gland tumors based on diffusion‐related features obtained from the parotid gland tumor, the peritumor parotid gland, and the ...contralateral parotid gland was evaluated. Seventy‐eight patients participated in this study and underwent magnetic resonance diffusion‐weighted imaging. Three regions of interest, including the parotid gland tumor, the peritumor parotid gland, and the contralateral parotid gland, were manually contoured for 92 tumors, including 20 malignant tumors (MTs), 42 Warthin tumors (WTs), and 30 pleomorphic adenomas (PMAs). We recorded multiple apparent diffusion coefficient (ADC) features and applied a machine‐learning method with the features to classify the three types of tumors. With only mean ADC of tumors, the area under the curve of the classification model was 0.63, 0.85, and 0.87 for MTs, WTs, and PMAs, respectively. The performance metrics were improved to 0.81, 0.89, and 0.92, respectively, with multiple features. Apart from the ADC features of parotid gland tumor, the features of the peritumor and contralateral parotid glands proved advantageous for tumor classification. Combining machine learning and multiple features provides excellent discrimination of tumor types and can be a practical tool in the clinical diagnosis of parotid gland tumors.
In this study, the performance of machine learning in classifying parotid gland tumors based on clinical features and diffusion‐related biomarkers was evaluated. The classification accuracy was 0.81 to 0.92 for three types of tumors. Combining machine learning and hybrid features provides discrimination of tumor types and can be a practical tool in the clinical diagnosis of parotid gland tumors.
Our study aimed to compare bone scintigraphy and dual-layer detector spectral CT (DLCT) with multiphase contrast enhancement for the diagnosis of osteoblastic bone lesions in patients with prostate ...cancer. The patients with prostate cancer and osteoblastic bone lesions detected on DLCT were divided into positive bone scintigraphy group (pBS) and negative bone scintigraphy group (nBS) based on bone scintigraphy. A total of 106 patients (57 nBS and 49 pBS) was included. The parameters of each lesion were measured from DLCT including Hounsfield unit (HU), 40-140 keV monochromatic HU, effective nuclear numbers (Z
), and Iodine no water (InW) value in non-contrast phase (N), the arterial phase (A), and venous phase (V). The slope of the spectral curve at 40 and 100 keV, the different values of the parameters between A and N phase (A-N), V and N phase (V-N), and hybrid prediction model with multiparameters were used to differentiate pBS from nBS. Receiver operating characteristic analysis was performed to compare the area under the curve (AUC) for differentiating the pBS group from the nBS group. The value of conventional HU values, slope, and InW in A-N and V-N, and hybrid model were significantly higher in the pBS group than in the nBS group. The hybrid model of all significant parameters had the highest AUC of 0.988, with 95.5% sensitivity and 94.6% specificity. DLCT with arterial contrast enhancement phase has the potential to serve as an opportunistic screening tool for detecting positive osteoblastic bone lesions, corresponding to those identified in bone scintigraphy.
Enhancer of zeste homolog 2 (EZH2), a histone methyltransferase, catalyzes tri-methylation of histone H3 at Lys 27 (H3K27me3) to regulate gene expression through epigenetic machinery. EZH2 functions ...as a double-facet molecule in regulation of gene expression via repression or activation mechanisms, depending on the different cellular contexts. EZH2 interacts with both histone and non-histone proteins to modulate diverse physiological functions including cancer progression and malignancy. In this review article, we focused on the updated information regarding microRNAs (miRNAs) and long non coding RNAs (lncRNAs) in regulation of EZH2, the oncogenic and tumor suppressive roles of EZH2 in cancer progression and malignancy, as well as current pre-clinical and clinical trials of EZH2 inhibitors.
Deep learning allows automatic segmentation of teeth on cone beam computed tomography (CBCT). However, the segmentation performance of deep learning varies among different training strategies. Our ...aim was to propose a 3.5D U-Net to improve the performance of the U-Net in segmenting teeth on CBCT. This study retrospectively enrolled 24 patients who received CBCT. Five U-Nets, including 2Da U-Net, 2Dc U-Net, 2Ds U-Net, 2.5Da U-Net, 3D U-Net, were trained to segment the teeth. Four additional U-Nets, including 2.5Dv U-Net, 3.5Dv5 U-Net, 3.5Dv4 U-Net, and 3.5Dv3 U-Net, were obtained using majority voting. Mathematical morphology operations including erosion and dilation (E&D) were applied to remove diminutive noise speckles. Segmentation performance was evaluated by fourfold cross validation using Dice similarity coefficient (DSC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV). Kruskal-Wallis test with post hoc analysis using Bonferroni correction was used for group comparison. P < 0.05 was considered statistically significant. Performance of U-Nets significantly varies among different training strategies for teeth segmentation on CBCT (P < 0.05). The 3.5Dv5 U-Net and 2.5Dv U-Net showed DSC and PPV significantly higher than any of five originally trained U-Nets (all P < 0.05). E&D significantly improved the DSC, accuracy, specificity, and PPV (all P < 0.005). The 3.5Dv5 U-Net achieved highest DSC and accuracy among all U-Nets. The segmentation performance of the U-Net can be improved by majority voting and E&D. Overall speaking, the 3.5Dv5 U-Net achieved the best segmentation performance among all U-Nets.