A Coarse-to-Fine Fusion Network for Small Liver Tumor Detection and Segmentation: A Real-World Study

Liver tumor semantic segmentation is a crucial task in medical image analysis that requires multiple MRI modalities. This paper proposes a novel coarse-to-fine fusion segmentation approach to detect and segment small liver tumors of various sizes. To enhance the segmentation accuracy of small liver tumors, the method incorporates a detection module and a CSR (convolution-SE-residual) module, which includes a convolution block, an SE (squeeze and excitation) module, and a residual module for fine segmentation. The proposed method demonstrates superior performance compared to conventional single-stage end-to-end networks. A private liver MRI dataset comprising 218 patients with a total of 3605 tumors, including 3273 tumors smaller than 3.0 cm, were collected for the proposed method. There are five types of liver tumors identified in this dataset: hepatocellular carcinoma (HCC); metastases of the liver; cholangiocarcinoma (ICC); hepatic cyst; and liver hemangioma. The results indicate that the proposed method outperforms the single segmentation networks 3D UNet and nnU-Net as well as the fusion networks of 3D UNet and nnU-Net with nnDetection. The proposed architecture was evaluated on a test set of 44 images, with an average Dice similarity coefficient (DSC) and recall of 86.9% and 86.7%, respectively, which is a 1% improvement compared to the comparison method. More importantly, compared to existing methods, our proposed approach demonstrates state-of-the-art performance in segmenting small objects with sizes smaller than 10 mm, achieving a Dice score of 85.3% and a malignancy detection rate of 87.5%.

3D Visualization of Whole Brain Vessels and Quantification of Vascular Pathology in a Chronic Hypoperfusion Model Causing White Matter Damage

Chronic cerebral hypoperfusion is an important pathological factor in many neurodegenerative diseases, such as cerebral small vessel disease (CSVD). One of the most used animal models for chronic cerebral hypoperfusion is the bilateral common carotid artery stenosis (BCAS) mouse. For the therapy of CSVD and other diseases, it will be beneficial to understand the pathological alterations of the BCAS mouse, particularly vascular pathological changes. A mouse model of BCAS was used, and 8 weeks later, cognitive function of the mice was examined by using novel object recognition test and eight-arm radial maze test. 11.7 T magnetic resonance imaging (MRI) and luxol fast blue staining were used to evaluate the injury of the corpus callosum (CC), anterior commissure (AC), internal capsule (IC), and optic tract (Opt) in the cerebral white matter of mice. Three-dimensional vascular images of the whole brain of mice were acquired using fluorescence micro-optical sectioning tomography (fMOST) with a high resolution of 0.32 × 0.32 × 1.00 μm³. Then, the damaged white matter regions were further extracted to analyze the vessel length density, volume fraction, tortuosity, and the number of vessels of different internal diameters. The mouse cerebral caudal rhinal vein was also extracted and analyzed for its branch number and divergent angle in this study. BCAS modeling for 8 weeks resulted in impaired spatial working memory, reduced brain white matter integrity, and myelin degradation in mice, and CC showed the most severe white matter damage. 3D revascularization of the whole mouse brain showed that the number of large vessels was reduced and the number of small vessels was increased in BCAS mice. Further analysis revealed that the vessel length density and volume fraction in the damaged white matter region of BCAS mice were significantly reduced, and the vascular lesions were most noticeable in the CC. At the same time, the number of small vessels in the above white matter regions was significantly reduced, while the number of microvessels was significantly increased in BCAS mice, and the vascular tortuosity was also significantly increased. In addition, the analysis of caudal rhinal vein extraction revealed that the number of branches and the average divergent angle in BCAS mice were significantly reduced. The BCAS modeling for 8 weeks will lead to vascular lesions in whole brain of mice, and the caudal nasal vein was also damaged, while BCAS mice mainly mitigated the damages by increasing microvessels. What is more, the vascular lesions in white matter of mouse brain can cause white matter damage and spatial working memory deficit. These results provide evidence for the vascular pathological alterations caused by chronic hypoperfusion.

Neuroplasticity of pain processing and motor control in CAI patients: A UK Biobank study with clinical validation

Background

Pain plays an important role in chronic ankle instability (CAI), and prolonged pain may be associated with ankle dysfunction and abnormal neuroplasticity.

Purpose

To investigate the differences in resting-state functional connectivity among the pain-related brain regions and the ankle motor-related brain regions between healthy controls and patients with CAI, and explore the relationship between patients' motor function and pain.

Study design

A cross-database, cross-sectional study.

Methods

This study included a UK Biobank dataset of 28 patients with ankle pain and 109 healthy controls and a validation dataset of 15 patients with CAI and 15 healthy controls. All participants underwent resting-state functional magnetic resonance imaging scanning, and the functional connectivity (FC) among the pain-related brain regions and the ankle motor-related brain regions were calculated and compared between groups. The correlations between the potentially different functional connectivity and the clinical questionnaires were also explored in patients with CAI.

Results

The functional connection between the cingulate motor area and insula significantly differed between groups in both the UK Biobank (p = 0.005) and clinical validation dataset (p = 0.049), which was also significantly correlated with Tegner scores (r = 0.532, p = 0.041) in patients with CAI.

Conclusion

A reduced functional connection between the cingulate motor area and the insula was present in patients with CAI, which was also directly correlated with reduction in the level of patient physical activity.

Cortical and subcortical morphological alteration in Angelman syndrome

BACKGROUND

Angelman syndrome (AS) is a neurodevelopmental disorder with serious seizures. We aim to explore the brain morphometry of patients with AS and figure out whether the seizure is associated with brain development.

METHODS

Seventy-three patients and 26 healthy controls (HC) underwent high-resolution structural brain MRI. Group differences between the HC group and the AS group and also between AS patients with seizure (AS-Se) and age-matched AS patients with non-seizure (AS-NSe) were compared. The voxel-based and surface-based morphometry analyses were used in our study. Gray matter volume, cortical thickness (CTH), and local gyrification index (LGI) were assessed to analyze the cortical and subcortical structure alteration in the AS brain.

RESULTS

Firstly, compared with the HC group, children with AS were found to have a significant decrease in gray matter volume in the subcortical nucleus, cortical, and cerebellum. However, the gray matter volume of AS patients in the inferior precuneus was significantly increased. Secondly, patients with AS had significantly increased LGI in the whole brain as compared with HC. Thirdly, the comparison of AS-Se and the AS-NSe groups revealed a significant decrease in caudate volume in the AS-Se group. Lastly, we further selected the caudate and the precuneus as ROIs for volumetric analysis, the AS group showed significantly increased LGI in the precuneus and reduced CTH in the right precuneus. Between the AS-Se and the AS-NSe groups, the AS-Se group exhibited significantly lower density in the caudate, while only the CTH in the left precuneus showed a significant difference.

CONCLUSIONS

These results revealed cortical and subcortical morphological alterations in patients with AS, including globally the decreased brain volume in the subcortical nucleus, the increased gray matter volume of precuneus, and the whole-brain increase of LGI and reduction of CTH. The abnormal brain pattern was more serious in patients with seizures, suggesting that the occurrence of seizures may be related to abnormal brain changes.

Measurement of Cerebral Oxygen Extraction Fraction Using Quantitative BOLD Approach: A Review

Quantification of brain oxygenation and metabolism, both of which are indicators of the level of brain activity, plays a vital role in understanding the cerebral perfusion and the pathophysiology of brain disorders. Magnetic resonance imaging (MRI), a widely used clinical imaging technique, which is very sensitive to magnetic susceptibility, has the possibility of substituting positron emission tomography (PET) in measuring oxygen metabolism. This review mainly focuses on the quantitative blood oxygenation level-dependent (qBOLD) method for the evaluation of oxygen extraction fraction (OEF) in the brain. Here, we review the theoretic basis of qBOLD, as well as existing acquisition and quantification methods. Some published clinical studies are also presented, and the pros and cons of qBOLD method are discussed as well.

Low Regional Homogeneity of Intrinsic Cerebellar Activity in Ankle Instability: An Externally Validated rs-fMRI Study

PURPOSE

Joint deafferentation after post-ankle sprain ligament healing can disrupt sensory input from the ankle and induce maladaptive neuroplasticity, especially in the cerebellum. This study aimed to determine whether the regional homogeneity of intrinsic cerebellar activity differs between patients with ankle instability and healthy controls without a history of ankle injury.

METHODS

The current study used a primary data set of 18 patients and 22 healthy controls and an external UK Biobank data set of 16 patients with ankle instability and 69 healthy controls for a cross-database, cross-sectional investigation. All participants underwent resting-state functional magnetic resonance imaging to calculate their regional homogeneity (ReHo) value. Between-group comparisons of the sensorimotor-related subregions of the cerebellum were first performed in the primary data set to identify low cerebellar ReHo in patients with multiple comparison corrections, and the surviving subregions were then externally validated in the UK Biobank data set. Correlation analyses between the ReHo values and clinical features were also performed.

RESULTS

The ReHo value of cerebellar lobule VIIIb was significantly lower in the ankle instability group than in the controls (0.170 ± 0.016 vs 0.184 ± 0.019 in the primary data set, 0.157 ± 0.026 vs 0.180 ± 0.042 in the UK Biobank data set). The ReHo values of this subregion showed a significant positive correlation with the Cumberland Ankle Instability Tool scores in the ankle instability group (r = 0.553, P-corrected = 0.0348).

CONCLUSIONS

Patients with ankle instability had lower intraregional coherence in cerebellar lobule VIIIb than that of controls, which was also positively correlated with the intensity of self-reported ankle instability.

Modified exosomal SIRPα variants alleviate white matter injury after intracerebral hemorrhage via microglia/macrophages

Background

Despite limited efficiency, modulation of microglia/macrophages has shown to attenuate neuroinflammation after intracerebral hemorrhage (ICH). In this context, we evaluated the efficacy of modified exosomal signal regulatory protein α (SIRPα) variants (SIRPα-v Exos) in microglia/macrophages and neuroinflammation-associated white matter injury after ICH.

Methods

SIRPα-v Exos were engineered to block CD47-SIRPα interactions. After obtaining SIRPα-v Exos from lentivirus-infected mesenchymal stem cells, C57BL/6 mice suffering from ICH underwent consecutive intravenous injections of SIRPα-v Exos (6 mg/kg) for 14 days. Afterwards, the volume of hematoma and neurological dysfunctions were assessed in mice continuously until 35 days after ICH. In addition, demyelination, electrophysiology and neuroinflammation were evaluated. Furthermore, the mechanisms of microglial regulation by SIRPα-v Exos were investigated in vitro under coculture conditions.

Results

The results demonstrated that the clearance of hematoma in mice suffering from ICH was accelerated after SIRPα-v Exo treatment. SIRPα-v Exos improved long-term neurological dysfunction by ameliorating white matter injury. In addition, SIRPα-v Exos recruited regulatory T cells (Tregs) to promote M2 polarization of microglia/macrophages in the peri-hematoma tissue. In vitro experiments further showed that SIRPα-v Exos regulated primary microglia in a direct and indirect manner in synergy with Tregs.

Conclusion

Our studies revealed that SIRPα-v Exos could accelerate the clearance of hematoma and ameliorate secondary white matter injury after ICH through regulation of microglia/macrophages. SIRPα-v Exos may become a promising treatment for ICH in clinical practice.

Graphical Abstract

Incorporating multiple magnetic resonance diffusion models to differentiate low- and high-grade adult gliomas: a machine learning approach

BACKGROUND

Accurate grading of gliomas is a challenge in imaging diagnosis. This study aimed to evaluate the performance of a machine learning (ML) approach based on multiparametric diffusion-weighted imaging (DWI) in differentiating low- and high-grade adult gliomas.

METHODS

A model was developed from an initial cohort containing 74 patients with pathology-confirmed gliomas, who underwent 3 tesla (3T) diffusion magnetic resonance imaging (MRI) with 21 b values. In all, 112 histogram features were extracted from 16 parameters derived from seven diffusion models [monoexponential, intravoxel incoherent motion (IVIM), diffusion kurtosis imaging (DKI), fractional order calculus (FROC), continuous-time random walk (CTRW), stretched-exponential, and statistical]. Feature selection and model training were performed using five randomly permuted five-fold cross-validations. An internal test set (15 cases of the primary dataset) and an external cohort (n=55) imaged on a different scanner were used to validate the model. The diagnostic performance of the model was compared with that of a single DWI model and DWI radiomics using accuracy, sensitivity, specificity, and the area under the curve (AUC).

RESULTS

Seven significant multiparametric DWI features (two from the stretched-exponential and FROC models, and three from the CTRW model) were selected to construct the model. The multiparametric DWI model achieved the highest AUC (0.84, versus 0.71 for the single DWI model, P<0.05), an accuracy of 0.80 in the internal test, and both AUC and accuracy of 0.76 in the external test.

CONCLUSIONS

Our multiparametric DWI model differentiated low- (LGG) from high-grade glioma (HGG) with better generalization performance than the established single DWI model. This result suggests that the application of an ML approach with multiple DWI models is feasible for the preoperative grading of gliomas.

Noninvasive Delineation of Glioma Infiltration with Combined 7T Chemical Exchange Saturation Transfer Imaging and MR Spectroscopy: A Diagnostic Accuracy Study

For precise delineation of glioma extent, amino acid PET is superior to conventional MR imaging. Since metabolic MR sequences such as chemical exchange saturation transfer (CEST) imaging and MR spectroscopy (MRS) were developed, we aimed to evaluate the diagnostic accuracy of combined CEST and MRS to predict glioma infiltration. Eighteen glioma patients of different tumor grades were enrolled in this study; 18F-fluoroethyltyrosine (FET)-PET, amide proton transfer CEST at 7 Tesla(T), MRS and conventional MR at 3T were conducted preoperatively. Multi modalities and their association were evaluated using Pearson correlation analysis patient-wise and voxel-wise. Both CEST (R = 0.736, p < 0.001) and MRS (R = 0.495, p = 0.037) correlated with FET-PET, while the correlation between CEST and MRS was weaker. In subgroup analysis, APT values were significantly higher in high grade glioma (3.923 ± 1.239) and IDH wildtype group (3.932 ± 1.264) than low grade glioma (3.317 ± 0.868, p < 0.001) or IDH mutant group (3.358 ± 0.847, p < 0.001). Using high FET uptake as the standard, the CEST/MRS combination (AUC, 95% CI: 0.910, 0.907−0.913) predicted tumor infiltration better than CEST (0.812, 0.808−0.815) or MRS (0.888, 0.885−0.891) alone, consistent with contrast-enhancing and T2-hyperintense areas. Probability maps of tumor presence constructed from the CEST/MRS combination were preliminarily verified by multi-region biopsies. The combination of 7T CEST/MRS might serve as a promising non-radioactive alternative to delineate glioma infiltration, thus reshaping the guidance for tumor resection and irradiation.

Alteration of brain functional networks induced by electroacupuncture stimulation in rats with ischemia–reperfusion: An independent component analysis

Motor dysfunction is the major sequela of ischemic stroke. Motor recovery after stroke has been shown to be associated with remodeling of large-scale brain networks, both functionally and structurally. Electroacupuncture (EA) is a traditional Chinese medicine application that has frequently been recommended as an alternative therapy for ischemic stroke and is reportedly effective for alleviating motor symptoms in patients. In the present study, the effect of EA on the alterations of functional resting state networks (RSNs) was explored after middle cerebral artery occlusion/reperfusion (MCAO/R) injury using resting-state functional MRI. Rats were randomly assigned to three groups, including the sham group, MCAO/R group and MCAO/R+EA group. The ladder rung walking test was conducted prior to and after modeling to assess behavioral changes. RSNs were identified based on the independent component analysis (ICA) performed on the fMRI data from groups. EA treatment effectively reduced the occurrence of contralateral forelimb foot faults. Furthermore, our results suggested the disrupted function of the whole-brain network following ischemic stroke and the modulatory effect of acupuncture. The sensorimotor network (SMN), interoceptive network (IN), default mode network (DMN) and salience network (SN) were related to the therapeutic effect of EA on stroke recovery. Collectively, our findings confirmed the effect of EA on motor function recovery after cerebral ischemia reperfusion and shed light on the assessment of EA intervention-induced effects on brain networks. This study provides neuroimaging evidence to explain the therapeutic effects of EA in ischemic stroke and will lay the groundwork for further studies.

Decoding transcriptional signatures of the association between free water and macroscale organizations in healthy adolescents

We leveraged a novel index of diffusion MRI to investigate the relationships among cortical free water, macro-organizations and gene expression in healthy adults. Few research has been conducted to investigate the role of free water in the healthy adults due to it can easily be affected also by aging diseases. High quality data of 350 subjects from Human Connectome Project were used in our study. Cortical free water was estimated by using a bi-tensor model. The free water was high in the limbic, insular and somatosensory cortex, while being lower in motor and association cortex. The negative correlation between the free water and cortical thickness has been consistently identified in almost all the cortical regions. Negative correlation between the cortical free water and structural covariance (rho=-0.38, p_spin=0.005) revealed the free water was sensitive to cortical heterogeneity. Using human gene expression dataset, we found the gene expression pattern of the relationship between the free water and cortical thickness spatially coupled with primary gradient of structural covariance network (rho=0.40, p_spin=0.004). Our findings indicated the free water was sensitive to the cortical cellular status. The relationship between free water and macroscale organization also reflected hierarchal structures of cerebral cortex.

Corrigendum: Impaired Glymphatic Transport Kinetics Following Induced Acute Ischemic Brain Edema in a Mouse pMCAO Model

[This corrects the article DOI: 10.3389/fneur.2022.860255.].

Impaired Glymphatic Transport Kinetics Following Induced Acute Ischemic Brain Edema in a Mouse pMCAO Model

Background

Cerebral edema forms immediately after blood flow interruption in ischemic stroke, which largely increased the death and disability. The glymphatic (glial-lymphatic) pathway is a major regulator of the brain liquid dynamics and homeostasis. This study aimed to investigate the transport kinetics of the glymphatic system after the appearance of ischemic edema.

Methods

In this study, a coated filament was attached to the left middle cerebral artery (MCA) of mice to establish a mouse model of permanent middle cerebral artery occlusion with an intact blood-brain barrier (BBB). The glymphatic function was then quantified using contrast-enhanced MRI (11.7T) by employing an injection of gadobenate dimeglumine (BOPTA-Gd) into the cisterna magna of mice. We then evaluated the expression and polarization of aquaporin-4 (AQP4) as a proxy for the physiological state of the glymphatic system.

Results

Our results revealed a positive correlation between the signal intensity in T1-weighted images and the corresponding apparent diffusion coefficient (ADC) values in the cortex, striatum, and periventricular zone, suggesting that impaired glymphatic transport kinetics in these regions is correlated to the cytotoxic edema induced by the occlusion of MCA. Furthermore, the increased depolarization of AQP4 in the parenchyma perivascular space (PVS) was consistent with glymphatic failure following the induced early cerebral ischemic edema.

Conclusions

Glymphatic transport kinetics were suppressed between the onset of cytotoxic edema and the disruption of the BBB, which correlated with the diminishing ADC values that vary based on edema progression, and is associated with depolarization of AQP4 in the parenchyma PVSs.

MRI-Based Radiomic Machine-Learning Model May Accurately Distinguish between Subjects with Internet Gaming Disorder and Healthy Controls

Purpose To identify cerebral radiomic features related to the diagnosis of Internet gaming disorder (IGD) and construct a radiomics-based machine-learning model for IGD diagnosis. Methods A total of 59 treatment-naïve subjects with IGD and 69 age- and sex-matched healthy controls (HCs) were recruited and underwent anatomic and diffusion-tensor magnetic resonance imaging (MRI). The features of the morphometric properties of gray matter and diffusion properties of white matter were extracted for each participant. After excluding the noise feature with single-factor analysis of variance, the remaining 179 features were included in an all-relevant feature selection procedure within cross-validation loops to identify features with significant discriminative power. Random forest classifiers were constructed and evaluated based on the identified features. Results No overall differences in the total brain volume (1,555,295.64 ± 152,316.31 mm³ vs. 154,491.19 ± 151,241.11 mm³), total gray (709,119.83 ± 59,534.46 mm³ vs. 751,018.21 ± 58,611.32 mm³) and white (465,054.49 ± 51,862.65 mm³ vs. 470,600.22 ± 47,006.67 mm³) matter volumes, and subcortical region volume (63,882.71 ± 5110.42 mm³ vs. 64,764.36 ± 4332.33 mm³) between the IGD and HC groups were observed. The mean classification accuracy was 73%. An altered cortical shape in the bilateral fusiform, left rostral middle frontal (rMFG), left cuneus, left parsopercularis (IFG), and regions around the right uncinate fasciculus (UF) and left internal capsule (IC) contributed significantly to group discrimination. Conclusions: Our study found the brain morphology alterations between IGD subjects and HCs through a radiomics-based machine-learning method, which may help revealing underlying IGD-related neurobiology mechanisms.

Deep learning-based identification of acute ischemic core and deficit from non-contrast CT and CTA

The accurate identification of irreversible infarction and salvageable tissue is important in planning the treatments for acute ischemic stroke (AIS) patients. Computed tomographic perfusion (CTP) can be used to evaluate the ischemic core and deficit, covering most of the territories of anterior circulation, but many community hospitals and primary stroke centers do not have the capability to perform CTP scan in emergency situation. This study aimed to identify AIS lesions from widely available non-contrast computed tomography (NCCT) and CT angiography (CTA) using deep learning. A total of 345AIS patients from our emergency department were included. A multi-scale 3D convolutional neural network (CNN) was used as the predictive model with inputs of NCCT, CTA, and CTA+ (8 s delay after CTA) images. An external cohort with 108 patients was included to further validate the generalization performance of the proposed model. Strong correlations with CTP-RAPID segmentations (r = 0.84 for core, r = 0.83 for deficit) were observed when NCCT, CTA, and CTA+ images were all used in the model. The diagnostic decisions according to DEFUSE3 showed high accuracy when using NCCT, CTA, and CTA+ (0.90±0.04), followed by the combination of NCCT and CTA (0.87±0.04), CTA-alone (0.76±0.06), and NCCT-alone (0.53±0.09).

MAGnitude-Image-to-Complex K-space (MAGIC-K) Net: A Data Augmentation Network for Image Reconstruction

Deep learning has demonstrated superior performance in image reconstruction compared to most conventional iterative algorithms. However, their effectiveness and generalization capability are highly dependent on the sample size and diversity of the training data. Deep learning-based reconstruction requires multi-coil raw k-space data, which are not collected by routine scans. On the other hand, large amounts of magnitude images are readily available in hospitals. Hence, we proposed the MAGnitude Images to Complex K-space (MAGIC-K) Net to generate multi-coil k-space data from existing magnitude images and a limited number of required raw k-space data to facilitate the reconstruction. Compared to some basic data augmentation methods applying global intensity and displacement transformations to the source images, the MAGIC-K Net can generate more realistic intensity variations and displacements from pairs of anatomical Digital Imaging and Communications in Medicine (DICOM) images. The reconstruction performance was validated in 30 healthy volunteers and 6 patients with different types of tumors. The experimental results demonstrated that the high-resolution Diffusion Weighted Image (DWI) reconstruction benefited from the proposed augmentation method. The MAGIC-K Net enabled the deep learning network to reconstruct images with superior performance in both healthy and tumor patients, qualitatively and quantitatively.

Recommendation for Cardiac Magnetic Resonance Imaging-Based Phenotypic Study: Imaging Part

Cardiac magnetic resonance (CMR) imaging provides important biomarkers for the early diagnosis of many cardiovascular diseases and has been reported to reveal phenome-wide associations of cardiac/aortic structure and functionality in population studies. Nevertheless, due to the complexity of operation and variations among manufactural vendors, magnetic field strengths, coils, sequences, scan parameters, and image analysis approaches, CMR is rarely used in large cohort studies. Existing guidelines mainly focused on the diagnosis of cardiovascular diseases, which did not aim to basic research. The purpose of this study was to propose a recommendation for CMR based phenotype measurements for cohort study. We classify the imaging sequences of CMR into three categories according to the importance and universality of corresponding measurable phenotypes. The acquisition time and repeatability of the phenotypic measurement were also taken into consideration during the categorization. Unlike other guidelines, this recommendation focused on quantitative measurement of large amount of phenotypes from CMR.

Assessment of Vascular Network Connectivity of Hepatocellular Carcinoma Using Graph-Based Approach

BACKGROUND

The angiogenesis of liver cancer is a key condition for its growth, invasion, and metastasis. This study aims to investigate vascular network connectivity of hepatocellular carcinoma (HCC) using graph-based approach.

METHODS

Orthotopic HCC xenograft models (n=10) and the healthy controls (n=10) were established. After 21 days of modeling, hepatic vascular casting and Micro-CT scanning were performed for angiography, followed by blood vessels automatic segmentation and vascular network modeling. The topologic parameters of vascular network, including clustering coefficient (CC), network structure entropy (NSE), and average path length (APL) were quantified. Topologic parameters of the tumor region, as well as the background liver were compared between HCC group and normal control group.

RESULTS

Compared with normal control group, the tumor region of HCC group showed significantly decreased CC [(0.046 ± 0.005) vs. (0.052 ± 0.006), P=0.026], and NSE [(0.9894 ± 0.0015) vs. (0.9927 ± 0.0010), P<0.001], and increased APL [(0.433 ± 0.138) vs. (0.188 ± 0.049), P<0.001]. Compared with normal control group, the background liver of HCC group showed significantly decreased CC [(0.047 ± 0.004) vs. (0.052 ± 0.006), P=0.041] and increased NSE [0.9938 (0.9936~0.9940) vs. (0.9927 ± 0.0010), P=0.035]. No significant difference was identified for APL between the two groups.

CONCLUSION

Graph-based approach allows quantification of vascular connectivity of HCC. Disrupted vascular topological connectivity exists in the tumor region, as well as the background liver of HCC.

Baseline Cerebral Ischemic Core Quantified by Different Automatic Software and Its Predictive Value for Clinical Outcome

Purpose

This study aims to investigate the agreement of three software packages in measuring baseline ischemic core volume (ICV) and penumbra volume (PV), and determine their predictive values for unfavorable clinical outcome in patients with endovascular thrombectomy (EVT).

Methods

Patients with acute ischemic stroke who underwent computed tomographic perfusion (CTP) were recruited. Baseline CTP measurements including ICV and PV were calculated by three software packages of IntelliSpace Portal (ISP), Rapid Processing of Perfusion and Diffusion (RAPID), and fast-processing of ischemic stroke (F-STROKE). All patients received EVT, and the modified Rankin scale (mRS) at 90 days after EVT was assessed to determine the clinical outcomes (favorable: mRS = 0-2; unfavorable: mRS = 3-6). The agreement of CTP measurements among three software packages was determined using intraclass correlation coefficient (ICC). The associations between CTP measurements and unfavorable clinical outcome were analyzed using logistic regression. Receiver operating characteristic curves were conducted to calculate the area under the curve (AUC) of CTP measurements in predicting unfavorable clinical outcome.

Results

Of 223 recruited patients (68.2 ± 11.3 years old; 145 males), 17.0% had unfavorable clinical outcome after EVT. Excellent agreement between F-STROKE and RAPID was found in measuring ICV (ICC 0.965; 95% CI 0.956-0.973) and PV (ICC 0.966; 95% CI 0.956-0.973). ICVs measured by three software packages were significantly associated with unfavorable clinical outcome before (odds ratios 1.012-1.018, all P < 0.01) and after (odds ratios 1.003-1.014, all P < 0.05) adjusted for confounding factors (age, gender, TOAST classification, and NIHSS on admission). In predicting unfavorable clinical outcome, ICV measured by F-STROKE showed similar performance to that measured by RAPID (AUC 0.701 vs. 0.717) but higher performance than that measured by ISP (AUC 0.629).

Conclusions

The software of F-STROKE has excellent agreement with the widely used analysis tool of RAPID in measuring ICV and PV. The ischemic core volume measured by both F-STROKE and RAPID is a stronger predictor for unfavorable clinical outcome after EVT compared to ISP.

Brain Development From Newborn to Adolescence: Evaluation by Neurite Orientation Dispersion and Density Imaging

Neurite orientation dispersion and density imaging (NODDI) is a diffusion model specifically designed for brain magnetic resonance imaging. Despite recent studies suggesting that NODDI modeling might be more sensitive to brain development than diffusion tensor imaging (DTI), these studies were limited to a relatively small age range and mainly based on the manually operated region of interest analysis. Therefore, this study applied NODDI to investigate brain development in a large sample size of 214 subjects ranging in ages from 0 to 14. The whole brain was automatically segmented into 122 regions. The maturation trajectory of each region was characterized by the time course of diffusion metrics and further quantified using nonlinear regression. The NODDI-derived metrics, neurite density index (NDI) and orientation dispersion index (ODI), increased with age. And these two metrics were superior to the DTI-derived metrics in SVM regression models of age. The NDI in white matter exhibited a more rapid growth than that in gray matter (including the cortex and deep nucleus). These diffusion indicators experienced conspicuous increases during early childhood and the growth speed slowed down in adolescence. Region-specific maturation patterns were described throughout the brain, including white matter, cortical and deep gray matter. These development patterns were evaluated and discussed on the basis of NODDI’s model assumptions. To summarize, this study verified the high sensitivity of NODDI to age over a crucial developmental period from newborn to adolescence. Moreover, the existing knowledge of brain development has been complemented, suggesting that NODDI has a potential capability in the investigation of brain development.

Imaging-Based Staging of Hepatic Fibrosis in Patients with Hepatitis B: A Dynamic Radiomics Model Based on Gd-EOB-DTPA-Enhanced MRI

Accurate grading of liver fibrosis can effectively assess the severity of liver disease and help doctors make an appropriate diagnosis. This study aimed to perform the automatic staging of hepatic fibrosis on patients with hepatitis B, who underwent gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced magnetic resonance imaging with dynamic radiomics analysis. The proposed dynamic radiomics model combined imaging features from multi-phase dynamic contrast-enhanced (DCE) images and time-domain information. Imaging features were extracted from the deep learning-based segmented liver volume, and time-domain features were further explored to analyze the variation in features during contrast enhancement. Model construction and evaluation were based on a 132-case data set. The proposed model achieved remarkable performance in significant fibrosis (fibrosis stage S1 vs. S2–S4; accuracy (ACC) = 0.875, area under the curve (AUC) = 0.867), advanced fibrosis (S1–S2 vs. S3–S4; ACC = 0.825, AUC = 0.874), and cirrhosis (S1–S3 vs. S4; ACC = 0.850, AUC = 0.900) classifications in the test set. It was more dominant compared with the conventional single-phase or multi-phase DCE-based radiomics models, normalized liver enhancement, and some serological indicators. Time-domain features were found to play an important role in the classification models. The dynamic radiomics model can be applied for highly accurate automatic hepatic fibrosis staging.

An Approach Based on Mammographic Imaging and Radiomics for Distinguishing Male Benign and Malignant Lesions: A Preliminary Study

Purpose

To develop and validate an imaging-radiomics model for the diagnosis of male benign and malignant breast lesions.

Methods

Ninety male patients who underwent preoperative mammography from January 2011 to December 2018 were enrolled in this study (63 in the training cohort and 27 in the validation cohort). The region of interest was segmented into a mediolateral oblique view, and 104 radiomics features were extracted. The minimum redundancy and maximum relevance (mRMR) and the least absolute shrinkage and selection operator (LASSO) methods were used to exclude radiomics features to establish the radiomics score (rad-score). Mammographic features were evaluated by two radiologists. Univariate logistic regression was used to select for imaging features, and multivariate logistic regression was used to construct an imaging model. An imaging-radiomics model was eventually established, and a nomogram was developed based on the imaging-radiomics model. Area under the curve (AUC) and decision curve analysis (DCA) were applied to assess the clinical value.

Results

The AUC based on the imaging model in the validation cohort was 0.760, the sensitivity was 0.750, and the specificity was 0.727. The AUC, sensitivity and specificity based on the radiomics in the validation cohort were 0.820, 0.750, and 0.867, respectively. The imaging-radiomics model was better than the imaging and radiomics models; the AUC, sensitivity, and specificity of the imaging-radiomics model in the validation cohort were 0.870, 0.824, and 0.900, respectively.

Conclusion

The imaging-radiomics model created by the imaging characteristics and radiomics features exhibited a favorable discriminatory ability for male breast cancer.

R1ρ at high spin-lock frequency could be a complementary imaging biomarker for liver iron overload quantification

PURPOSE

To compare the correlations among the R1ρ, R2, and R2* relaxation rates with liver iron concentration (LIC) in the assessment of rat liver iron content and explore the application potential of R1ρ in assessing liver iron content.

METHODS

Iron dextran (dosage of 0, 25, 50, 100, and 200 mg/kg body weight) was injected into 35 male rats to increase the amount of iron storage in the liver. After one week, all rats were euthanized with isoflurane. A portion of the largest hepatic lobe was extracted to quantify the LIC by inductively coupled plasma, and the remaining liver tissue was stored in 4% buffered paraformaldehyde for 24 h before MRI. Spin-lock preparation with a RARE (rapid acquisition with relaxation enhancement) readout (9 different spin-lock times and 7 different spin-lock frequencies (FSLs)) and multi-echo UTE (ultrashort TE) pulses were developed to quantify R1ρ and R2 * on a Bruker 11.7 T MR system. For comparisons with R1ρ and R2*, R2 was acquired using the CPMG sequence.

RESULTS

Mean R1ρ values displayed dispersion, with decrease in R1ρ at higher FSLs. Spearman's correlation analysis (two-tailed) indicated that the R1ρ values were significantly associated with LIC at FSL = 2000, 2500, and 3000 Hz (r = 0.365 and P = 0.031, r = 0.608 and P < 0.001, and r = 0.764 and P < 0.001, respectively), and were not significantly associated with LIC at FSL = 500, 1000, 1250, and 1500 Hz (all P > 0.05). R2 and R2* showed significant linear correlations with LIC (r = 0.787 and P < 0.001, and r = 0.859 and P < 0.001, respectively). Correlation analysis across R1ρ, R2, and R* also suggested that the correlation strength between R1ρ and R2 and between R1ρ and R* showed an increasing trend with increase in FSL.

CONCLUSION

In this study, a strong association was observed between R1ρ and LIC at high FSLs further confirming previous findings. The results demonstrated that R1ρ at high FSL might serve as a complementary imaging biomarker for liver iron overload quantification.

Brain controllability and morphometry similarity of internet gaming addiction

Internet gaming addiction (IGD) is a common disease in teenagers which usually reflects the abnormalities in brain function or structure. Several computational models have been applied to investigate the characteristic of IGD brain networks, for instance, the conception of brain controllability. The primary objective of this study was to explore the relationship between brain controllability and IGD related clinical behaviour. A sample of 101 subjects, including 49 IGD patients and 52 normal controls, were recruited to undergo MR T1 and DTI scanning. Specifically, the MR images were used to generate the white matter connectivity matrix and the morphometry similarity network. The morphometry similarity network was then divided into several communities using modular decomposition. After, average controllability, modal controllability and synchronizability were calculated through measuring the adjacency matrix. The results indicated that the IGD group had greater synchronizability and modal controllability compared to that of the control group, and different morphological-based brain communities had different controllability properties. Furthermore, the addiction demonstrated the mediating effects between nodal or modular brain controllability as well as anxiety. In conclusion, brain controllability could be a potential biomarker of IGD.

Fibronectin-Targeting and Cathepsin B-Activatable Theranostic Nanoprobe for MR/Fluorescence Imaging and Enhanced Photodynamic Therapy for Triple Negative Breast Cancer

Because of the lack of specific targets, the highly aggressive triple negative breast cancer (TNBC) is unable to benefit from endocrine therapy or conventional targeting therapy. Even worse, current diagnostic and therapeutic approaches have limited value for TNBC. Therefore, developing TNBC-specific theranostic probes for accurate diagnosis and further selective therapy will build a powerful toolbox for TNBC management. In this contribution, we developed a sequential strategy to enhance the specificity of TNBC theranostics. In this theranostic system, a versatile nanoprobe (Pep-SQ@USPIO) was integrated legitimately for the fibronectin-targeting MR imaging and CTSB-activatable fluorescence imaging, followed with enhanced photodynamic therapy (PDT) of TNBC. First, the fibronectin overexpressed in the extracellular matrix (ECM) of TNBC was used as a biomarker for targeting theranostics using the Cys-Arg-Glu-Lys-Ala (CREKA) peptide. For another, the fluorescence and PDT capacity of self-developed squaraine photosensitizer (SQ) were prequenched by ultrasmall superparamagnetic iron oxide (USPIO), an MR imaging contrast agent. Once the linker, Gly-Phe-Leu-Gly (GFLG) peptide, was selectively cleaved by TNBC-derived CTSB, the liberated SQ photosensitizer allowed light-up fluorescence imaging and enhanced PDT of TNBC. Remarkably, this research demonstrates that tumor-ECM-targeting and endogenous enzyme-activated nanoprobes open a new avenue for TNBC theranostics.

Machine Learning-Based Analysis of MR Multiparametric Radiomics for the Subtype Classification of Breast Cancer

Objective: To investigate whether machine learning analysis of multiparametric MR radiomics can help classify immunohistochemical (IHC) subtypes of breast cancer.

Study design: One hundred and thirty-four consecutive patients with pathologically-proven invasive ductal carcinoma were retrospectively analyzed. A total of 2,498 features were extracted from the DCE and DWI images, together with the new calculated images, including DCE images changing over six time points (DCE_sequential) and DWI images changing over three b-values (DWI_sequential). We proposed a novel two-stage feature selection method combining traditional statistics and machine learning-based methods. The accuracies of the 4-IHC classification and triple negative (TN) vs. non-TN cancers were assessed.

Results: For the 4-IHC classification task, the best accuracy of 72.4% was achieved based on linear discriminant analysis (LDA) or subspace discrimination of assembled learning in conjunction with 20 selected features, and only small dependent emphasis of Kendall-tau-b for sequential features, based on the DWI_sequential with the LDA model, yielding an accuracy of 53.7%. The linear support vector machine (SVM) and medium k-nearest neighbor using eight features yielded the highest accuracy of 91.0% for comparing TN to non-TN cancers, and the maximum variance for DWI_sequential alone, together with a linear SVM model, achieved an accuracy of 83.6%.

Conclusions: Whole-tumor radiomics on MR multiparametric images, DCE images changing over time points, and DWI images changing over different b-values provide a non-invasive analytical approach for breast cancer subtype classification and TN cancer identification.

High-field mr diffusion-weighted image denoising using a joint denoising convolutional neural network

BACKGROUND

Low signal-to-noise ratio (SNR) has been a major limiting factor for the application of higher-resolution diffusion-weighted imaging (DWI). Most of the conventional denoising models suffer from the drawbacks of shallow feature extraction and hand-crafted parameter tuning. Although multiple studies have shown the promising applications of image denoising using convolutional neural networks (CNNs), none of them have considered denoising multiple b-value DWIs using a multichannel CNN model.

PURPOSE

To present a joint denoising CNN (JD-CNN) model to improve the SNR of multiple b-value DWI.

STUDY TYPE

Retrospective technical development.

POPULATION

Twenty healthy rats and two rats with clinically confirmed focal cortical dysplasia were included to evaluate the performance of the proposed method.

FIELD STRENGTH/SEQUENCE

11.7T MRI, a multiple b-values DWI sequence.

ASSESSMENT

The total variation (TV) and BM3D denoising methods were also performed on the same dataset for comparison. Peak SNR (PSNR) and normalized mean square error (NMSE) were calculated for the assessment of image qualities.

STATISTICAL TESTS

A paired Student's t-test was conducted to compare the diffusion parameter measurements between different approaches. P < 0.01 was considered statistically significant.

RESULTS

Simulation results showed substantial improvement of image quality after JD-CNN denoising (PSNR of original image: 23.15 ± 1.77; PSNR of denoised image: 42.94 ± 2.12). The proposed method outperforms the state-of-the-art methods on high b-value DWIs in terms of PSNR (TV: 33.51 ± 0.83, BM3D: 35.12 ± 0.94, JD-CNN: 46.52 ± 0.98). In addition, the NMSE of the estimated apparent diffusion coefficient (ADC) reduces from 0.72 ± 0.13 to 0.45 ± 0.06 (P < 0.01) with the application of the JD-CNN model.

DATA CONCLUSION

The proposed method is able to remove noise with a wide range of noise levels in multiple b-value DWI and improve the diffusion parameter estimation. This shows potential clinical promise.

LEVEL OF EVIDENCE

2 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2019;50:1937-1947.

IMAge/enGINE: a freely available software for rapid computation of high-dimensional quantification

Background

High-dimensional image data including diffusion weighted imaging, diffusion tensor imaging and dynamic imaging are important in exploring the connectivity, cellularity, pharmacokinetic and blood supply. IMAge/enGINE is software especially designed for high-dimensional medical image computing.

Methods

IMAge/enGINE is implemented based on open-source and cross-platform tools such as Qt, ITK and VTK. It processes the high-dimensional image data in a slice-by-slice computation mechanism. For computational efficiency, C++ is used for implementing IMAge/enGINE and multi-thread computing is handled in the scale of voxels. The architecture of IMAge/enGINE is modularized for easier extension.

Results

IMAge/enGINE has following features: (I) IMAge/enGINE is free for research use; (II) it has an easy-to-use graphic user interface designed for clinical users without programming or engineering background; (III) its frame work is open-source and extensible. Developers can implement algorithms as modules and integrate them into IMAge/enGINE or generate their own application.

Conclusions

The source of IMAge/enGINE is hosted at https://github.com/VusionMed/IMAge-enGINE. Multiple diffusion and perfusion models are implemented and integrated into IMAge/enGINE and its binaries can be downloaded freely at http://www.vusion.com.cn/?page_id=14971.