Matrix metalloproteinase 9 expression and glioblastoma survival prediction using machine learning on digital pathological images.

This study aimed to apply pathomics to predict Matrix metalloproteinase 9 (MMP9) expression in glioblastoma (GBM) and investigate the underlying molecular mechanisms associated with pathomics. Here, we included 127 GBM patients, 78 of whom were randomly allocated to the training and test cohorts for pathomics modeling. The prognostic significance of MMP9 was assessed using Kaplan-Meier and Cox regression analyses. PyRadiomics was used to extract the features of H&E-stained whole slide images. Feature selection was performed using the maximum relevance and minimum redundancy (mRMR) and recursive feature elimination (RFE) algorithms. Prediction models were created using support vector machines (SVM) and logistic regression (LR). The performance was assessed using ROC analysis, calibration curve assessment, and decision curve analysis. MMP9 expression was elevated in patients with GBM. This was an independent prognostic factor for GBM. Six features were selected for the pathomics model. The area under the curves (AUCs) of the training and test subsets were 0.828 and 0.808, respectively, for the SVM model and 0.778 and 0.754, respectively, for the LR model. The C-index and calibration plots exhibited effective estimation abilities. The pathomics score calculated using the SVM model was highly correlated with overall survival time. These findings indicate that MMP9 plays a crucial role in GBM development and prognosis. Our pathomics model demonstrated high efficacy for predicting MMP9 expression levels and prognosis of patients with GBM.

Evaluation of deep learning-based reconstruction late gadolinium enhancement images for identifying patients with clinically unrecognized myocardial infarction.

BACKGROUND: The presence of infarction in patients with unrecognized myocardial infarction (UMI) is a critical feature in predicting adverse cardiac events. This study aimed to compare the detection rate of UMI using conventional and deep learning reconstruction (DLR)-based late gadolinium enhancement (LGEO and LGEDL, respectively) and evaluate optimal quantification parameters to enhance diagnosis and management of suspected patients with UMI. METHODS: This prospective study included 98 patients (68 men; mean age: 55.8 ± 8.1 years) with suspected UMI treated at our hospital from April 2022 to August 2023. LGEO and LGEDL images were obtained using conventional and commercially available inline DLR algorithms. The myocardial signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and percentage of enhanced area (Parea) employing the signal threshold versus reference mean (STRM) approach, which correlates the signal intensity (SI) within areas of interest with the average SI of normal regions, were analyzed. Analysis was performed using the standard deviation (SD) threshold approach (2SD-5SD) and full width at half maximum (FWHM) method. The diagnostic efficacies based on LGEDL and LGEO images were calculated. RESULTS: The SNRDL and CNRDL were two times better than the SNRO and CNRO, respectively (P < 0.05). Parea-DL was elevated compared to Parea-O using the threshold methods (P < 0.05); however, no intergroup difference was found based on the FWHM method (P > 0.05). The Parea-DL and Parea-O also differed except between the 2SD and 3SD and the 4SD/5SD and FWHM methods (P < 0.05). The receiver operating characteristic curve analysis revealed that each SD method exhibited good diagnostic efficacy for detecting UMI, with the Parea-DL having the best diagnostic efficacy based on the 5SD method (P < 0.05). Overall, the LGEDL images had better image quality. Strong diagnostic efficacy for UMI identification was achieved when the STRM was ≥ 4SD and ≥ 3SD for the LGEDL and LGEO, respectively. CONCLUSIONS: STRM selection for LGEDL magnetic resonance images helps improve clinical decision-making in patients with UMI. This study underscored the importance of STRM selection for analyzing LGEDL images to enhance diagnostic accuracy and clinical decision-making for patients with UMI, further providing better cardiovascular care.

Deep learning-based radiomics of computed tomography angiography to predict adverse events after initial endovascular repair for acute uncomplicated Stanford type B aortic dissection.

PURPOSE: This study aimed to construct a predictive model integrating deep learning-derived radiomic features from computed tomography angiography (CTA) and clinical biomarkers to forecast postoperative adverse events (AEs) in patients with acute uncomplicated Stanford type B aortic dissection (uTBAD) undergoing initial thoracic endovascular aortic repair (TEVAR). METHODS: We retrospectively evaluated 369 patients treated with TEVAR for acute uTBAD from January 2015 to December 2022. A three-dimensional (3D) deep convolutional neural network (CNN) automated radiomic feature extraction from CTA images. Feature selection, using Analysis of Variance (ANOVA) and the Least Absolute Shrinkage and Selection Operator (LASSO) algorithms, refined a radiomic score (Rad-Score). This score, alongside clinical parameters, was modelled via Extreme Gradient Boosting (XGBoost) analysis. Model calibration was assessed by calibration curves. RESULTS: The integration of the Rad-Score with clinical factors including albumin and C-reactive protein levels moderately enhanced predictive efficiency, exhibiting an area under the curve (AUC) of 1.000 (95%CI, 1.000-1.000) in the training cohort and 0.990 (95%CI, 0.966-1.000) in the internal validation cohort. In an independent validation cohort from another hospital, the combined model yielded an AUC of 0.985 (95%CI, 0.965-1.000), with an accuracy, precision, sensitivity, and specificity of 0.92, 0.92, 0.94, and 0.91, respectively. CONCLUSIONS: The synergistic application of deep learning-based radiomics from CTA and clinical indicators holds promise for anticipating AEs post-initial thoracic endovascular aortic repair in patients with acute uTBAD. The clinical utility of the constructed combined model, offering prognostic foresight during follow-up, has been substantiated.

Development and Validation of a Deep Learning Model to Reduce the Interference of Rectal Artifacts in MRI-based Prostate Cancer Diagnosis.

Purpose To develop an MRI-based model for clinically significant prostate cancer (csPCa) diagnosis that can resist rectal artifact interference. Materials and Methods This retrospective study included 2203 male patients with prostate lesions who underwent biparametric MRI and biopsy between January 2019 and June 2023. Targeted adversarial training with proprietary adversarial samples (TPAS) strategy was proposed to enhance model resistance against rectal artifacts. The automated csPCa diagnostic models trained with and without TPAS were compared using multicenter validation datasets. The impact of rectal artifacts on the diagnostic performance of each model at the patient and lesion levels was compared using the area under the receiver operating characteristic curve (AUC) and the area under the precision-recall curve (AUPRC). The AUC between models was compared using the DeLong test, and the AUPRC was compared using the bootstrap method. Results The TPAS model exhibited diagnostic performance improvements of 6% at the patient level (AUC: 0.87 vs 0.81, P < .001) and 7% at the lesion level (AUPRC: 0.84 vs 0.77, P = .007) compared with the control model. The TPAS model demonstrated less performance decline in the presence of rectal artifact-pattern adversarial noise than the control model (ΔAUC: -17% vs -19%, ΔAUPRC: -18% vs -21%). The TPAS model performed better than the control model in patients with moderate (AUC: 0.79 vs 0.73, AUPRC: 0.68 vs 0.61) and severe (AUC: 0.75 vs 0.57, AUPRC: 0.69 vs 0.59) artifacts. Conclusion This study demonstrates that the TPAS model can reduce rectal artifact interference in MRI-based csPCa diagnosis, thereby improving its performance in clinical applications. Keywords: MR-Diffusion-weighted Imaging, Urinary, Prostate, Comparative Studies, Diagnosis, Transfer Learning Clinical trial registration no. ChiCTR23000069832 Supplemental material is available for this article. Published under a CC BY 4.0 license.

Neurodegeneration within the rostral spinal cord is associated with brain gray matter volume atrophy in the early stage of cervical spondylotic myelopathy.

STUDY DESIGN: Case-control study. OBJECTIVES: Investigating the association between neurodegeneration within rostral spinal cord and brain gray matter volume (GMV) and assessing the relationship between remote neurodegenerative changes and clinical outcomes at the early phase of Cervical Spondylotic Myelopathy (CSM). SETTING: University/hospital. METHODS: Using Spinal Cord Toolbox, spinal cord morphometrics (cross-sectional area [CSA], gray matter area [GMA], white matter area [WMA]) of 40 patients with CSM and 28 healthy controls (HCs) were computed and compared using two-sample t test. Brain GMV of the two groups was analyzed using voxel-based morphometry approach. Pearson's correlation between spinal cord morphometrics and altered brain GMV and Spearman's relationship between remote neurodegenerations and clinical outcomes were conducted in CSM group. RESULTS: Compared to HCs, CSA and WMA at C2/3 and GMV in right postcentral gyrus (PoCG.R) and left supplementary motor area (SMA.L) were significantly decreased in patients with CSM. CSA and WMA at C2/3 were associated with GMV in SMA.L and MCG.R in patients with CSM. CSA at C2/3 and GMV in PoCG.R were related to modified Japanese Orthopedic Association score in patients with CSM. CONCLUSIONS: The associations between CSA and WMA at C2/3 and GMV in SMA.L and MCG.R suggest a concordant change pattern and adaptive mechanisms for neuronal plasticity underlying remote neurodegeneration in early CSM. The atrophy of CSA at C2/3 and GMV loss in PoCG.R can serve as potential neuroimaging biomarkers of early structural changes within spinal cord and brain preceding marked clinical disabilities in patients with CSM.

Atrial and Ventricular Strain Imaging Using CMR in the Prediction of Ventricular Arrhythmia in Patients with Myocarditis.

(1) Objective: Myocarditis can be associated with ventricular arrhythmia (VA), individual non-invasive risk stratification through cardiovascular magnetic resonance (CMR) is of great clinical significance. Our study aimed to explore whether left atrial (LA) and left ventricle (LV) myocardial strain serve as independent predictors of VA in patients with myocarditis. (2) Methods: This retrospective study evaluated CMR scans in 141 consecutive patients diagnosed with myocarditis based on the updated Lake Louise criteria (29 females, mean age 41 ± 20). The primary endpoint was VA; this encompassed ventricular fibrillation, sustained ventricular tachycardia, nonsustained ventricular tachycardia, and frequent premature ventricular complexes. LA and LV strain function were performed on conventional cine SSFP sequences. (3) Results: After a median follow-up time of 23 months (interquartile range (18-30)), 17 patients with acute myocarditis reached the primary endpoint. In the multivariable Cox regression analysis, LA reservoir (hazard ratio [HR] and 95% confidence interval [CI]: 0.93 [0.87-0.99], p = 0.02), LA booster (0.87 95% CI [0.76-0.99], p = 0.04), LV global longitudinal (1.26 95% CI [1.02-1.55], p = 0.03), circumferential (1.37 95% CI [1.08-1.73], p = 0.008), and radial strain (0.89 95% CI [0.80-0.98], p = 0.01) were all independent determinants of VA. Patients with LV global circumferential strain > -13.3% exhibited worse event-free survival compared to those with values ≤ -13.3% (p < 0.0001). (4) Conclusions: LA and LV strain mechanism on CMR are independently associated with VA events in patients with myocarditis, independent to LV ejection fraction, and late gadolinium enhancement location. Incorporating myocardial strain parameters into the management of myocarditis may improve risk stratification.

Microvascular permeability and texture analysis of bone marrow in diabetic rabbits with critical limb ischemia based on dynamic contrast-enhanced magnetic resonance imaging.

BACKGROUND: Early on in the development of diabetes, skeletal muscles can exhibit microarchitectural changes that can be detected using texture analysis (TA) based on volume transfer constant (Ktrans) maps. Nevertheless, there have been few studies and thus we evaluated microvascular permeability and the TA of the bone marrow in diabetics with critical limb ischemia (CLI). METHODS: Eighteen male rabbits were randomly assigned equally into an operation group with hindlimb ischemia and diabetes, a sham-operated group with diabetes only, and a control group. Dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) was performed on all rabbits at predetermined intervals (1, 5, 10, 15, 20, and 25 days post-surgery). The pharmacokinetic model was used to generate the permeability parameters, while the textural parameters were derived from the Ktrans map. Data analysis methods included the independent sample t-test, Mann-Whitney U test, repeated-measures analysis of variance, and Pearson correlation tests. RESULTS: The Ktrans values reached a minimum on day 1 after ischemia induction, then gradually recovered, but remained lower than those of the sham-operated group. The volume fraction only showed a significant difference between the operation group and the sham-operated group on day 5 post-surgery, but not in the extravascular extracellular space volume fraction at all time points. A significantly reduced Ktrans on day 1, a decreased number of bone trabeculae (Tb.N), and the area of bone trabeculae (Tb.Ar), and an increased microvessel density on day 25 in the operation group compared with the sham-operated group were observed. At each time point, there was a discernible difference between the two groups in the mean value, mean of positive pixels, and sumAverage. CONCLUSIONS: The early stages of diabetic bone marrow with CLI can be evaluated by DCE-MRI for microvascular permeability. Texture analysis based on DCE-MRI could act as an imaging discriminator and new radiological analysis tool for critical limb ischemia in diabetes mellitus.

Usefulness of T2-Weighted Images with Deep-Learning-Based Reconstruction in Nasal Cartilage.

OBJECTIVE: This study aims to evaluate the feasibility of visualizing nasal cartilage using deep-learning-based reconstruction (DLR) fast spin-echo (FSE) imaging in comparison to three-dimensional fast spoiled gradient-echo (3D FSPGR) images. MATERIALS AND METHODS: This retrospective study included 190 set images of 38 participants, including axial T1- and T2-weighted FSE images using DLR (T1WIDL and T2WIDL, belong to FSEDL) and without using DLR (T1WIO and T2WIO, belong to FSEO) and 3D FSPGR images. Subjective evaluation (overall image quality, noise, contrast, artifacts, and identification of anatomical structures) was independently conducted by two radiologists. Objective evaluation including signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) was conducted using manual region-of-interest (ROI)-based analysis. Coefficient of variation (CV) and Bland-Altman plots were used to demonstrate the intra-rater repeatability of measurements for cartilage thickness on five different images. RESULTS: Both qualitative and quantitative results confirmed superior FSEDL to 3D FSPGR images (both p < 0.05), improving the diagnosis confidence of the observers. Lower lateral cartilage (LLC), upper lateral cartilage (ULC), and septal cartilage (SP) were relatively well delineated on the T2WIDL, while 3D FSPGR showed poorly on the septal cartilage. For the repeatability of cartilage thickness measurements, T2WIDL showed the highest intra-observer (%CV = 8.7% for SP, 9.5% for ULC, and 9.7% for LLC) agreements. In addition, the acquisition time for T1WIDL and T2WIDL was respectively reduced by 14.2% to 29% compared to 3D FSPGR (both p < 0.05). CONCLUSIONS: Two-dimensional equivalent-thin-slice T1- and T2-weighted images using DLR showed better image quality and shorter scan time than 3D FSPGR and conventional construction images in nasal cartilages. The anatomical details were preserved without losing clinical performance on diagnosis and prognosis, especially for pre-rhinoplasty planning.

Reduced myocardial strain of interventricular septum among male amateur marathon runners: a cardiac magnetic resonance study.

OBJECTIVES: To analyze the long-term effect of multiple marathons on cardiac structure and function in amateur marathon runners compared with healthy controls. DESIGN: Cross-sectional study using male amateur marathon runners (n = 32) and age-matched cohort of male healthy controls (n = 12). METHODS: A total of 32 male amateur marathon runners (age 44 ±â€¯7 years) and 12 male healthy controls (age 42 ±â€¯8 years) underwent cardiac magnetic resonance (CMR). The relevant parameters of cardiac structure and function were studied employing feature-tracking strain analysis. RESULTS: Amateur marathon runners showed lower heart rates, body mass index and body surface area. The left ventricular (LV) mass index, LV end-diastolic volume index and right ventricular end-systolic volume index were significantly higher in amateur marathon runners compared with healthy controls. Furthermore, walls of interventricular septum (IVS) in amateur marathon runners were thicker than healthy controls. There was no significant difference between two groups in the global myocardial strain (MS) in LV. However, the segmental radial and circumferential strains of the LV were lower in amateur marathon runners compared to healthy controls, specifically in the 8th and 9th segments. Finally, we also found as the total running intensity increased, so did global longitudinal strain. CONCLUSIONS: We reported higher wall thickness and lower regional radial and circumferential strain in the IVS region in amateur marathon runners, suggesting that prolonged and high-intensity exercise may cause cardiac remodeling. Further studies are needed to investigate whether this is an adaptive or maladaptive change in amateur marathon runners.

Radiomics Features on Magnetic Resonance Images Can Predict C5aR1 Expression Levels and Prognosis in High-Grade Glioma.

BACKGROUND: The complement component C5a receptor 1 (C5aR1) regulates cancer immunity. This retrospective study aimed to assess its prognostic value in high-grade glioma (HGG) and predict C5aR1 expression using a radiomics approach. METHODS: Among 298 patients with HGG, 182 with MRI data were randomly divided into training and test groups for radiomics analysis. We examined the association between C5aR1 expression and prognosis through Kaplan-Meier and Cox regression analyses. We used maximum relevance-minimum redundancy and recursive feature elimination algorithms for radiomics feature selection. We then built a support vector machine (SVM) and a logistic regression model, investigating their performances using receiver operating characteristic, calibration curves, and decision curves. RESULTS: C5aR1 expression was elevated in HGG and was an independent prognostic factor (hazard ratio = 3.984, 95% CI: 2.834-5.607). Both models presented with >0.8 area under the curve values in the training and test datasets, indicating efficient discriminatory ability, with SVM performing marginally better. The radiomics score calculated using the SVM model correlated significantly with overall survival (p < 0.01). CONCLUSIONS: Our results highlight C5aR1's role in HGG development and prognosis, supporting its potential as a prognostic biomarker. Our radiomics model can noninvasively and effectively predict C5aR1 expression and patient prognosis in HGG.

Frontoparietal network homogeneity as a biomarker for mania and remitted bipolar disorder and a predictor of early treatment response in bipolar mania patient.

OBJECTIVE: Previous studies have revealed the frontoparietal network (FPN) plays a key role in the imaging pathophysiology of bipolar disorder (BD). However, network homogeneity (NH) in the FPN among bipolar mania (BipM), remitted bipolar disorder (rBD), and healthy controls (HCs) remains unknown. The present study aimed to explore whether NH within the FPN can be used as an imaging biomarker to differentiate BipM from rBD and to predict treatment efficacy for patients with BipM. METHODS: Sixty-six patients with BD (38 BipM and 28 rBD) and 60 HCs participated in resting-state functional magnetic resonance imaging and neuropsychological tests. Independent component analysis and NH analysis were applied to analyze the imaging data. RESULTS: Relative to HCs, BipM patients displayed increased NH in the left middle frontal gyrus (MFG), and rBD patients displayed increased NH in the right inferior parietal lobule (IPL). Compared to rBD patients, BipM patients displayed reduced NH in the right IPL. Furthermore, support vector machine results exhibited that NH values in the right IPL could distinguish BipM patients from rBD patients with 69.70 %, 57.89 %, and 91.67 % for accuracy, sensitivity, and specificity, respectively, and support vector regression results exhibited a significant association between predicted and actual symptomatic improvement based on the reduction ratio of the Young` Mania Rating Scale total scores (r = 0.466, p < 0.01). CONCLUSION: The study demonstrated distinct NH values in the FPN could serve as a valuable neuroimaging biomarker capable of differentiating patients with BipM and rBD, and NH values of the left MFG as a potential predictor of early treatment response in patients with BipM.

An integrated convolutional neural network for classifying small pulmonary solid nodules.

Achieving accurate classification of benign and malignant pulmonary nodules is essential for treating some diseases. However, traditional typing methods have difficulty obtaining satisfactory results on small pulmonary solid nodules, mainly caused by two aspects: (1) noise interference from other tissue information; (2) missing features of small nodules caused by downsampling in traditional convolutional neural networks. To solve these problems, this paper proposes a new typing method to improve the diagnosis rate of small pulmonary solid nodules in CT images. Specifically, first, we introduce the Otsu thresholding algorithm to preprocess the data and filter the interference information. Then, to acquire more small nodule features, we add parallel radiomics to the 3D convolutional neural network. Radiomics can extract a large number of quantitative features from medical images. Finally, the classifier generated more accurate results by the visual and radiomic features. In the experiments, we tested the proposed method on multiple data sets, and the proposed method outperformed other methods in the small pulmonary solid nodule classification task. In addition, various groups of ablation experiments demonstrated that the Otsu thresholding algorithm and radiomics are helpful for the judgment of small nodules and proved that the Otsu thresholding algorithm is more flexible than the manual thresholding algorithm.

High-Resolution Single-Shot Fast Spin-Echo MR Imaging with Deep Learning Reconstruction Algorithm Can Improve Repeatability and Reproducibility of Follicle Counting.

OBJECTIVE: To investigate the diagnostic performance of high-resolution single-shot fast spin-echo (SSFSE) imaging with deep learning (DL) reconstruction algorithm on follicle counting and compare it with original SSFSE images and conventional fast spin-echo (FSE) images. METHODS: This study included 20 participants (40 ovaries) with clinically confirmed polycystic ovary syndrome (PCOS) who underwent high-resolution ovary MRI, including three-plane T2-weighted FSE sequences and slice-matched T2-weighted SSFSE sequences. A DL reconstruction algorithm was applied to the SSFSE sequences to generate SSFSE-DL images, and the original SSFSE images were also saved. Subjective evaluations such as the blurring artifacts, subjective noise, and clarity of the follicles on the SSFSE-DL, SSFSE, and conventional FSE images were independently conducted by two observers. Intra-class correlation coefficients and Bland-Altman plots were used to present the repeatability and reproducibility of the follicle number per ovary (FNPO) based on the three types of images. RESULTS: SSFSE-DL images showed less blurring artifact, subjective noise, and better clarity of the follicles than SSFSE and FSE (p < 0.05). For the repeatability of the FNPO, SSFSE-DL showed the highest intra-observer (ICC = 0.930; 95% CI: 0.878-0.962) and inter-observer (ICC = 0.914; 95% CI: 0.843-0.953) agreements. The inter-observer 95% limits of agreement (LOA) for SSFSE-DL, SSFSE, and FSE ranged from -3.7 to 4.5, -4.4 to 7.0, and -7.1 to 7.6, respectively. The intra-observer 95% LOA for SSFSE-DL, SSFSE, and FSE ranged from -3.5 to 4.0, -5.1 to 6.1, and -5.7 to 4.2, respectively. The absolute values of intra-observer and inter-observer differences for SSFSE-DL were significantly lower than those for SSFSE and FSE (p < 0.05). CONCLUSIONS: Compared with the original SSFSE images and the conventional FSE images, high-resolution SSFSE images with DL reconstruction algorithm can better display follicles, thus improving FNPO assessment.

Hippocampal Subfield Volumes in Amateur Marathon Runners.

PURPOSE: Numerous studies have implicated the involvement of structure and function of the hippocampus in physical exercise, and the larger hippocampal volume is one of the relevant benefits reported in exercise. It remains to be determined how the different subfields of hippocampus respond to physical exercise. METHODS: A 3D T1-weighted magnetic resonance imaging was acquired in 73 amateur marathon runners (AMR) and 52 healthy controls (HC) matched with age, sex, and education. The Montreal Cognitive Assessment, the Pittsburgh Sleep Quality Index (PSQI), and the Fatigue Severity Scale were assessed in all participants. We obtained hippocampal subfield volumes using FreeSurfer 6.0. We compared the volumes of the hippocampal subfield between the two groups and ascertained correlation between the significant subfield metrics and the significant behavioral measure in AMR group. RESULTS: The AMR had significantly better sleep than HC, manifested as with lower score of PSQI. Sleep duration in AMR and HC was not significantly different from each other. In the AMR group, the left and right hippocampus, cornu ammonis 1 (CA1), CA4, granule cell and molecular layers of the dentate gyrus, molecular layer, left CA2-3, and left hippocampal-amygdaloid transition area volumes were significantly larger compared with those in the HC group. In AMR group, the correlations between the PSQI and the hippocampal subfield volumes were not significant. No correlations were found between hippocampal subfield volumes and sleep duration in AMR group. CONCLUSIONS: We reported larger volumes of specific hippocampal subfields in AMR, which may provide a hippocampal volumetric reserve that protects against age-related hippocampal deterioration. These findings should be further investigated in longitudinal studies.

Predicting prognosis of nasopharyngeal carcinoma based on deep learning: peritumoral region should be valued.

BACKGROUND: The purpose of this study was to explore whether incorporating the peritumoral region to train deep neural networks could improve the performance of the models for predicting the prognosis of NPC. METHODS: A total of 381 NPC patients who were divided into high- and low-risk groups according to progression-free survival were retrospectively included. Deeplab v3 and U-Net were trained to build segmentation models for the automatic segmentation of the tumor and suspicious lymph nodes. Five datasets were constructed by expanding 5, 10, 20, 40, and 60 pixels outward from the edge of the automatically segmented region. Inception-Resnet-V2, ECA-ResNet50t, EfficientNet-B3, and EfficientNet-B0 were trained with the original, segmented, and the five new constructed datasets to establish the classification models. The receiver operating characteristic curve was used to evaluate the performance of each model. RESULTS: The Dice coefficients of Deeplab v3 and U-Net were 0.741(95%CI:0.722-0.760) and 0.737(95%CI:0.720-0.754), respectively. The average areas under the curve (aAUCs) of deep learning models for classification trained with the original and segmented images and with images expanded by 5, 10, 20, 40, and 60 pixels were 0.717 ± 0.043, 0.739 ± 0.016, 0.760 ± 0.010, 0.768 ± 0.018, 0.802 ± 0.013, 0.782 ± 0.039, and 0.753 ± 0.014, respectively. The models trained with the images expanded by 20 pixels obtained the best performance. CONCLUSIONS: The peritumoral region NPC contains information related to prognosis, and the incorporation of this region could improve the performance of deep learning models for prognosis prediction.

Detection of Acute Optic Neuritis using Contrast-Enhanced 3-Dimensional Cube T1-Weighted Imaging: A Preliminary Study.

Objective; We aimed to assess whole-brain imaging with contrast-enhanced (CE) 3- dimensional (3D) Cube T1WI in improving the diagnostic accuracy of acute optic neuritis (ON) compared to conventional CE 2-dimensional (2D) T1WI. METHODS: At a field strength of 3 T, CE 3D Cube T1-weighted and conventional CE 2D T1- weighted MR images were retrospectively analyzed for 32 patients (64 optic nerves) with clinically confirmed acute ON. The study cohort included 36 pathological nerves. Image assessments including the overall image quality, clarity of the optic nerve, and visual contrast enhancement were performed by two blinded neuroradiologists using a 4-point scale. The sensitivity, specificity, and accuracy of the conventional 2D T1WI and 3D Cube T1WI were calculated according to the clinical diagnosis. RESULTS: The application of 3D Cube T1WI improved the overall image quality compared to 2D Ax T1WI and 2D Cor T1WI (P < 0.05). The clarity of the optic nerve and the visual contrast enhancement were higher for the 3D Cube T1WI compared to the 2D Ax T1WI and 2D Cor T1WI for at least one reader. The sensitivity, specificity, and accuracy were 89%, 86%, 88% for the 3D Cube T1WI respectively, and 75%, 79%, 77% for the conventional 2D T1WI respectively. The lesions detected by the conventional 2D T1WI were all detected by the 3D Cube T1WI. CONCLUSION: Our data show that whole-brain imaging with CE 3D Cube T1WI is a viable alternative for the detection of acute ON without sacrificing scanning efficiency.

CENP-A is a potential prognostic biomarker and correlated with immune infiltration levels in glioma patients.

Background: Centromeric protein A (CENP-A), an essential protein involved in chromosomal segregation during cell division, is associated with several cancer types. However, its role in gliomas remains unclear. This study examined the clinical and prognostic significance of CENP-A in gliomas. Methods: Data of patients with glioma were collected from the Cancer Genome Atlas. Logistic regression, the Kruskal-Wallis test, and the Wilcoxon signed-rank test were performed to assess the relationship between CENP-A expression and clinicopathological parameters. The Cox regression model and Kaplan-Meier curve were used to analyze the association between CENP-A and survival outcomes. A prognostic nomogram was constructed based on Cox multivariate analysis. Gene set enrichment analysis (GSEA) was conducted to identify key CENP-A-related pathways and biological processes. Results: CENP-A was upregulated in glioma samples. Increased CENP-A levels were significantly associated with the world health organization (WHO) grade [Odds ratio (OR) = 49.88 (23.52-129.06) for grade 4 vs. grades 2 and 3], primary therapy outcome [OR = 2.44 (1.64-3.68) for progressive disease (PD) and stable disease (SD) vs. partial response (PR) and complete response (CR)], isocitrate dehydrogenase (IDH) status [OR = 13.76 (9.25-20.96) for wild-type vs. mutant], 1p/19q co-deletion [OR = 5.91 (3.95-9.06) for no codeletion vs. co-deletion], and age [OR = 4.02 (2.68-6.18) for > 60 vs. ≤ 60]. Elevated CENP-A expression was correlated with shorter overall survival in both univariate [hazard ratio (HR): 5.422; 95% confidence interval (CI): 4.044-7.271; p < 0.001] and multivariate analyses (HR: 1.967; 95% CI: 1.280-3.025; p < 0.002). GSEA showed enrichment of numerous cell cycle-and tumor-related pathways in the CENP-A high expression phenotype. The calibration plot and C-index indicated the favorable performance of our nomogram for prognostic prediction in patients with glioma. Conclusion: We propose a role for CENP-A in glioma progression and its potential as a biomarker for glioma diagnosis and prognosis.

Two-stage hybrid network for segmentation of COVID-19 pneumonia lesions in CT images: a multicenter study.

COVID-19 has been spreading continuously since its outbreak, and the detection of its manifestations in the lung via chest computed tomography (CT) imaging is essential to investigate the diagnosis and prognosis of COVID-19 as an indispensable step. Automatic and accurate segmentation of infected lesions is highly required for fast and accurate diagnosis and further assessment of COVID-19 pneumonia. However, the two-dimensional methods generally neglect the intraslice context, while the three-dimensional methods usually have high GPU memory consumption and calculation cost. To address these limitations, we propose a two-stage hybrid UNet to automatically segment infected regions, which is evaluated on the multicenter data obtained from seven hospitals. Moreover, we train a 3D-ResNet for COVID-19 pneumonia screening. In segmentation tasks, the Dice coefficient reaches 97.23% for lung segmentation and 84.58% for lesion segmentation. In classification tasks, our model can identify COVID-19 pneumonia with an area under the receiver-operating characteristic curve value of 0.92, an accuracy of 92.44%, a sensitivity of 93.94%, and a specificity of 92.45%. In comparison with other state-of-the-art methods, the proposed approach could be implemented as an efficient assisting tool for radiologists in COVID-19 diagnosis from CT images.

Freezing of Gait in Multiple System Atrophy.

Background and Purpose: Freezing of gait (FOG) is a common gait disturbance phenomenon in multiple system atrophy (MSA) patients. The current investigation assessed the incidence FOG in a cross-sectional clinical study, and clinical correlations associated with it. Methods: Ninety-nine MSA patients from three hospitals in China were consecutively enrolled in the study. Eight patients were subsequently excluded from the analysis due to incomplete information. The prevalence of FOG symptoms in the MSA cohort was determined, and clinical manifestations in MSA patients with and without FOG were assessed. Results: Of 91 MSA patients, 60 (65.93%) exhibited FOG. The incidence of FOG increased with disease duration and motor severity and was correlated with modified Hoehn and Yahr (H-Y) stages [odds ratio (OR), 0.54; 95% confidence interval (CI), 0.33-3.92], longer disease duration (OR, 0.54, 95% CI, 0.37-0.78), higher Unified Multiple System Atrophy Rating Scale (UMSARS) score (OR, 0.96, 95% CI, 0.93-0.99), MSA-cerebellum subtype (OR, 2.99, 95% CI, 1.22-7.33), levodopa-equivalent dose (LDED) (OR, 0.998, 95% CI, 0.997-1.00), and higher Scale for the Assessment and Rating of Ataxia (SARA) score (OR, 0.80, 95% CI, 0.72-0.89) (logistic regression). Motor dysfunction was significantly positively associated with lower quality of life scores (p < 0.01). Conclusion: FOG is a common symptom in MSA patients and it is correlated with poor quality of life, disease progression and severity, levodopa-equivalent dose, and cerebellum impairment.

Microvascular Permeability and Texture Analysis of the Skeletal Muscle of Diabetic Rabbits With Critical Limb Ischaemia Based on DCE-MRI.

BACKGROUND: We evaluated skeletal muscle vascular permeability in diabetic rabbits with critical limb ischaemia using quantitative dynamic contrast agent-enhanced (DCE) magnetic resonance imaging (MRI) and explored the feasibility of using DCE-MRI Ktrans-based texture analysis for assessing early slight ischaemia-related skeletal muscle structural changes. METHOD: Twenty-four male New Zealand white rabbits (2.7 ± 0.3 kg; n = 12 each in sham-operated and experimental groups) underwent axial MRI of the vastus lateralis muscle at 1, 2, and 3 weeks after alloxan injection. Between-group and intra-group postoperative permeability and texture parameters were compared. Texture features of experimental groups in the third week were modelled by receiver operating characteristic (ROC) curve analysis. Correlations of permeability and of statistical texture parameters with peripheral blood endothelial progenitor cells (EPCs) and microvascular density (MVD) were analysed. RESULTS: In the experimental group, the transfer constant (Ktrans) was statistically significant at all time-points (F = 5.800, P = 0.009). Their vastus lateralis muscle Ktrans was significantly lower in the third than in the first week (P = 0.018) and correlated positively with peripheral blood EPCs in the experimental group [r = 0.598, (95% CI: 0.256, 0.807)]. The rate constant was negatively associated with vastus lateralis muscle MVD [r = -0.410, (95% CI: -0.698, -0.008)]. The area under the ROC curve of texture parameters based on Ktrans in ischaemic limbs was 0.882. CONCLUSIONS: Quantitative DCE-MRI parameters could evaluate microvascular permeability of ischaemic limb skeletal muscle, and texture analysis based on DCE-MRI Ktrans allowed evaluation of early slight skeletal muscle structural changes.