Search

The purpose of this study was to develop Bi-parametric Magnetic Resonance Imaging (BP-MRI) based radiomics models for differentiation between benign and malignant prostate lesions, and to cross-vendor validate the generalization ability of the models. The prebiopsy BP-MRI data (T2-Weighted Image [T2WI] and the Apparent Diffusion Coefficient [ADC]) of 459 patients with clinical suspicion of prostate cancer were acquired using two scanners from different vendors. The prostate biopsies are the reference standard for diagnosing benign and malignant prostate lesions. The training set was 168 patients' data from Siemens (Vendor 1), and the inner test set was 70 patients' data from the same vendor. The external test set was 221 patients' data from GE (Vendor 2). The lesion Region of Interest (ROI) was manually delineated by experienced radiologists. A total of 851 radiomics features including shape, first-order statistical, texture, and wavelet features were extracted from ROI in T2WI and ADC, respectively. Two feature-ranking methods (Minimum Redundancy Maximum Relevance [MRMR] and Wilcoxon Rank-Sum Test [WRST]) and three classifiers (Random Forest [RF], Support Vector Machine [SVM], and the Least Absolute Shrinkage and Selection Operator [LASSO] regression) were investigated for their efficacy in building single-parametric radiomics signatures. A biparametric radiomics model was built by combining the optimal single-parametric radiomics signatures. A comprehensive diagnosis model was built by combining the biparametric radiomics model with age and Prostate Specific Antigen (PSA) value using multivariable logistic regression. All models were built in the training set and independently validated in the inner and external test sets, and the performances of models in the diagnosis of benign and malignant prostate lesions were quantified by the Area Under the Receiver Operating Characteristic Curve (AUC). The mean AUCs of the inner and external test sets were calculated for each model. The non-inferiority test was used to test if the AUC of model in external test was not inferior to the AUC of model in inner test. Combining MRMR and LASSO produced the best-performing single-parametric radiomics signatures with the highest mean AUC of 0.673 for T2WI (inner test AUC = 0.729 vs. external test AUC = 0.616, p = 0.569) and the highest mean AUC of 0.810 for ADC (inner test AUC = 0.822 vs. external test AUC = 0.797, p = 0.102). The biparametric radiomics model produced a mean AUC of 0.833 (inner test AUC = 0.867 vs. external test AUC = 0.798, p = 0.051). The comprehensive diagnosis model had an improved mean AUC of 0.911 (inner test AUC = 0.935 vs. external test AUC = 0.886, p = 0.010). The comprehensive diagnosis model for differentiating benign from malignant prostate lesions was accurate and generalizable.

Objectives This study aimed to explore the spatial distribution of brain metastases (BMs) from small cell lung cancer (SCLC) a homogenous sample, and to identify the metastatic risk levels in brain regions. Methods T1-enhanced magnetic resonance imaging (MRI) from SCLC patients were retrospectively reviewed from three medical institutions in China. All images were registered to the standard brain template provided by the Montreal Neurological Institute (MNI) 152 database, followed by transformation of the location of all BMs to the space of standard brain. The MNI structural atlas and Anatomical Automatic Labeling (AAL) atlas were then used to identify the anatomical brain regions, and the observed and expected rates of BMs were compared using 2-tailed proportional hypothesis testing. The locations and sizes of brain lesions were analyzed after image standardization. Results A total of 215 eligible patients with 1033 lesions were screened by MRI, including 157 (73%) males and 58 (27%) females. The incidence of crucial structures were as follows: hippocampus 0.68%, parahippocampal 0.97%, brainstem 2.05%, cauate 0.68%, putamen 0.68%, pallidum 0.2%, thalamus 1.36%. No BMs were found in the amygdala, pituitary gland, or pineal gland. The cumulative frequency of the important structures was 6.62%. Based on the results of MNI structural atlas, the cerebellum, deep white matter and brainstem was identified as a higher risk region than expected for BMs (P = 9.80 ×10−15, 9.04 ×10−6), whereas temporal lobe were low-risk regions (P = 1.65 ×10−4). More detailed AAL atlas revealed that the low-risk regions for BMs was inferior frontal gyrus (P = 6.971 ×10−4), while the high-risk regions for BMs was cerebellar hemispheres (P = 1.177 ×10−9). Conclusion Many crucial structures including the hippocampus, parahippocampus, pituitary gland and thalamus etc. have low frequency of brain metastases in a population of SCLC patients. This study provides the help to investigate the clinical feasibility of HA-WBRT and non-uniform dose of PCI in a population of SCLC patients.

BACKGROUND Preoperative prediction of extracapsular extension (ECE) of prostate cancer (PCa) is important to guide clinical decision-making and improve patient prognosis. PURPOSE To investigate the value of multiparametric magnetic resonance imaging (mpMRI)-based peritumoral radiomics for preoperative prediction of the presence of ECE. STUDY TYPE Retrospective. POPULATION Two hundred eighty-four patients with PCa from two centers (center 1: 226 patients; center 2: 58 patients). Cases from center 1 were randomly divided into training (158 patients) and internal validation (68 patients) sets. Cases from center 2 were assigned to the external validation set. FIELD STRENGTH/SEQUENCE A 3.0 T MRI scanners (three vendors). Sequence: Pelvic T2-weighted turbo/fast spin echo sequence and diffusion weighted echo planar imaging sequence. ASSESSMENT The peritumoral region (PTR) was obtained by 3-12 mm (half of the tumor length) 3D dilatation of the intratumoral region (ITR). Single-MRI radiomics signatures, mpMRI radiomics signatures, and integrated models, which combined clinical characteristics with the radiomics signatures were built. The discrimination ability was assessed by area under the receiver operating characteristic curve (AUC) in the internal and external validation sets. STATISTICAL TESTS Fisher's exact test, Mann-Whitney U-test, DeLong test. RESULTS The PTR radiomics signatures demonstrated significantly better performance than the corresponding ITR radiomics signatures (AUC: 0.674 vs. 0.554, P

To develop a preoperative MRI-based radiomic-clinical nomogram for prediction of residual disease (RD) in patients with advanced high-grade serous ovarian carcinoma (HGSOC). In total, 217 patients with advanced HGSOC were enrolled from January 2014 to June 2019 and randomly divided into a training set (n = 160) and a validation set (n = 57). Finally, 841 radiomic features were extracted from each tumor on T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted imaging (CE-T1WI) sequence, respectively. We used two fusion methods, the maximal volume of interest (MV) and the maximal feature value (MF), to fuse the radiomic features of bilateral tumors, so that patients with bilateral tumors have the same kind of radiomic features as patients with unilateral tumors. The radiomic signatures were constructed by using mRMR method and LASSO classifier. Multivariable logistic regression analysis was used to develop a radiomic-clinical nomogram incorporating radiomic signature and conventional clinico-radiological features. The performance of the nomogram was evaluated on the validation set. In total, 342 tumors from 217 patients were analyzed in this study. The MF-based radiomic signature showed significantly better prediction performance than the MV-based radiomic signature (AUC = 0.744 vs. 0.650, p = 0.047). By incorporating clinico-radiological features and MF-based radiomic signature, radiomic-clinical nomogram showed favorable prediction ability with an AUC of 0.803 in the validation set, which was significantly higher than that of clinico-radiological signature and MF-based radiomic signature (AUC = 0.623, 0.744, respectively). The proposed MRI-based radiomic-clinical nomogram provides a promising way to noninvasively predict the RD status. • MRI-based radiomic-clinical nomogram is feasible to noninvasively predict residual disease in patients with advanced HGSOC. • The radiomic signature based on MF showed significantly better prediction performance than that based on MV. • The radiomic-clinical nomogram showed a favorable prediction ability with an AUC of 0.803.

Three-dimensional fluid-attenuated inversion-recovery (3 D-FLAIR) and real inversion-recovery (3 D-real IR) sequences are used to detect endolymphatic hydrops (EH), but medium inversion-time inversion-recovery imaging with magnitude reconstruction (MIIRMR) may be more sensitive.We investigated the inner-ear visualisation success rate and EH detection rates of 3 D-FLAIR and 3 D-real IR, and whether salvage MIIRMR could improve EH detection.Fifty-one patients (102 ears) with episodic or chronic vestibular syndrome were injected intra-tympanically with 8-fold diluted gadolinium, and 3 D-FLAIR and 3 D-real IR images obtained 24-h post-injection. If 3 D-FLAIR inner-ear visualisation failed, additional MIIRMR was performed. The success and EH detection rate increase by MIIRMR was calculated. The diagnostic performance of combined MIIRMR + 3D-FLAIR + 3D-real IR for Meniere's disease (MD) was evaluated.The success rates of 3 D-FLAIR and 3 D-real IR were 88.90% and 72.55%, respectively. MIIRMR increased the success and EH detection rates by 11.10% and 6.86%, respectively. In MD, MIIRMR increased these rates by 10.53% and 10.53%, respectively. 3 D-FLAIR + 3D-real IR + MIIRMR had 92.11% sensitivity, 79.68% specificity, 72.92% positive-predictive value, and 94.44% negative-predictive value for MD diagnosis.MIIRMR can improve success and EH detection rates when 3 D-FLAIR fails. Combined MIIRMR + 3D-FLAIR + 3D-real IR is more valuable for diagnosing MD than conventional sequences.

Background Preoperative differentiation of primary central nervous system lymphoma (PCNSL) from glioblastoma (GBM) is important to guide neurosurgical decision-making. Purpose To validate the generalization ability of radiomics models based on multiparametric-MRI (MP-MRI) for differentiating PCNSL from GBM. Study type Retrospective. Population In all, 240 patients with GBM (n = 129) or PCNSL (n = 111). Field strength/sequence 3.0T scanners (two vendors). Sequences: fluid-attenuation inversion recovery, diffusion-weighted imaging (DWI), and contrast-enhanced T1 -weighted imaging (CE-T1 WI). Apparent diffusion coefficients (ADCs) were derived from DWI. Assessment Cross-vendor and mixed-vendor validation were conducted. In cross-vendor validation, the training set was 149 patients' data from vendor 1, and test set was 91 patients' data from vendor 2. In mixed-vendor validation, a training set was 80% of data from both vendors, and the test set remained at 20% of data. Single and multisequence radiomics models were built. The diagnoses by radiologists with 5, 10, and 20 years' experience were obtained. The integrated models were built combining the diagnoses by the best-performing radiomics model and each radiologist. Model performance was validated in the test set using area under the ROC curve (AUC). Histological results were used as the reference standard. Statistical tests DeLong test: differences between AUCs. U-test: differences of numerical variables. Fisher's exact test: differences of categorical variables. Results In cross-vendor and mixed-vendor validation, the combination of CE-T1 WI and ADC produced the best-performing radiomics model, with AUC of 0.943 vs. 0.935, P = 0.854. The integrated models had higher AUCs than radiologists, with 5 (0.975 vs. 0.891, P = 0.002 and 0.995 vs. 0.885, P = 0.007), 10 (0.975 vs. 0.913, P = 0.029 and 0.995 vs. 0.900, P = 0.030), and 20 (0.975 vs. 0.945, P = 0.179 and 0.995 vs. 0.923, P = 0.046) years' experiences. Data conclusion Radiomics for differentiating PCNSL from GBM was generalizable. The model combining MP-MRI and radiologists' diagnoses had superior performance compared to the radiologists alone. Level of evidence 4 TECHNICAL EFFICACY STAGE: 2.

BACKGROUND The main endemic areas of alveolar echinococcosis (AE) are in Central Europe and Western China. Both the infiltration of intrahepatic vascular and bile duct structures as well as extrahepatic disease can lead to further complications and may increase morbidity in patients with AE. AIM To evaluate vascular/biliary involvement in hepatic AE and its distant extrahepatic disease manifestations in an international collective was the aim. METHODS Consecutively, five experienced examiners evaluated contrast-enhanced abdominal computed tomography (CT) scans for 200 patients with hepatic AE of each of four locations (n = 50) in Germany, France and China. Therefore, we retrospectively included the 50 most recent abdominal contrast-enhanced CT examinations at each center, performed because of hepatic AE from September 21, 2007 to March 21, 2018. AE liver lesions were classified according to the echinococcosis multilocularis Ulm classification for CT (EMUC-CT). Distant extrahepatic manifestations were documented either by whole body positron emission tomography–CT or with the addition of thoracic CT and cranial magnetic resonance imaging. Vascular/biliary involvement of the hepatic disease as well as the presence of distant extrahepatic manifestations were correlated with the EMUC-CT types of liver lesion. Statistical analysis was performed using SAS Version 9.4 (SAS Institute Inc., Cary, NC, United States). RESULTS Distant extrahepatic AE manifestations were significantly more frequent in China than in Europe (P = 0.0091). A significant relationship was found between the presence of distant extrahepatic disease and AE liver lesion size (P = 0.0075). Vascular/biliary structures were involved by the liver lesions significantly more frequently in China than in Europe (P < 0.0001), and vascular/biliary involvement depended on lesion size. Different morphological types of AE liver lesions led to varying frequencies of vascular/biliary involvement and were associated with different frequencies of distant extrahepatic manifestations: Vascular/biliary involvement as a function of lesions primary morphology ranged from 5.88% of type IV liver lesions to 100% among type III lesions. Type IV differed significantly in these associations from types I, II, and III (P < 0.0001). With respect to extrahepatic disease, the primary morphology types IV and V of liver lesions were not associated with any case of distant extrahepatic disease. In contrast, distant extrahepatic manifestations in types I–III were found to varying degrees, with a maximum of 22% for type III. CONCLUSION Different CT morphological patterns of hepatic AE lesions influence vascular/biliary involvement and the occurrence of distant extrahepatic manifestations. There are intercontinental differences regarding the characteristics of AE manifestation.

Epithelial ovarian cancers (EOC) can be divided into type I and type II according to etiology and prognosis. Accurate subtype differentiation can substantially impact patient management. In this study, we aimed to construct an MR image–based radiomics model to differentiate between type I and type II EOC. In this multicenter retrospective study, a total of 294 EOC patients from January 2010 to February 2019 were enrolled. Quantitative MR imaging features were extracted from the following axial sequences: T2WI FS, DWI, ADC, and CE-T1WI. A combined model was constructed based on the combination of these four MR sequences. The diagnostic performance was evaluated by ROC-AUC. In addition, an occlusion test was carried out to identify the most critical region for EOC differentiation. The combined radiomics model exhibited superior diagnostic capability over all four single-parametric radiomics models, both in internal and external validation cohorts (AUC of 0.806 and 0.847, respectively). The occlusion test revealed that the most critical region for differential diagnosis was the border zone between the solid and cystic components, or the less compact areas of solid component on direct visual inspection. MR image–based radiomics modeling can differentiate between type I and type II EOC and identify the most critical region for differential diagnosis. • Combined radiomics models exhibited superior diagnostic capability over all four single-parametric radiomics models, both in internal and external validation cohorts (AUC of 0.834 and 0.847, respectively). • The occlusion test revealed that the most crucial region for differentiating type Ι and type ΙΙ EOC was the border zone between the solid and cystic components, or the less compact areas of solid component on direct visual inspection on T2WI FS. • The light-combined model (constructed by T2WI FS, DWI, and ADC sequences) can be used for patients who are not suitable for contrast agent use.

Antibodies to myelin oligodendrocyte glycoprotein (MOG-ab) and antibodies to aquaporin-4 (AQP4-ab) have been suggested to play roles in commonly separated subsets of patients with neuromyelitis optica spectrum disorder (NMOSD) phenotypes. The aim of this study is to quantitatively delineate and compare the brain lesion distributions of AQP4-ab-positive and MOG-ab-positive patients. Fifty-seven and twenty-eight clinical MRI scans were collected from fifty-two AQP4-ab-positive and twenty-four MOG-ab-positive patients, respectively. T2 lesions were segmented manually on each axial FLAIR image. Probabilistic lesion distribution maps were created for each group after spatial normalization. Lobe-wise and voxel-wise quantitative comparisons of the two distributions were performed. A classification model based on the lesion distribution features was constructed to differentiate the two patient groups. Infratentorial and supratentorial brain lesions were found in both AQP4-ab-positive and MOG-ab-positive patients, with large inter-group overlap mainly in deep white matter (WM). In comparison with those in the AQP4 group, the brain lesions of the MOG-ab-positive patients had a larger size, dispersed distribution, and higher probabilities in the cerebellum, pons, midbrain, and GM and juxtacortical WM in temporal, sublobar, frontal, and parietal lobes. The area under the receiver operating characteristic curve of the lesion-distribution-based classification model was 0.951. MOG-ab-positive and AQP4-ab-positive groups showed similar but quantitatively different brain lesion distributions. These results may help clinicians in considering MOG versus AQP4 in initial diagnosis, and add rationale for sending corresponding serologic testing. • Brain lesion distributions of AQP-ab-positive and MOG-ab-positive NMOSD patients • Larger size, dispersed distribution, higher lesion probabilities in the cerebellum, pons, midbrain, and GM and juxtacortical WM in the MOG group • The lesion-distribution-based classification model differentiates the two groups with AUC = 0.951

Microvascular invasion (MVI) of small hepatocellular carcinoma (sHCC) (≤ 3.0 cm) is an independent prognostic factor for poor progression-free and overall survival. Radiomics can help extract imaging information associated with tumor pathophysiology.To develop and validate radiomics scores and a nomogram of gadolinium ethoxybenzyl-diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced magnetic resonance imaging (MRI) for preoperative prediction of MVI in sHCC.In total, 415 patients were diagnosed with sHCC by postoperative pathology. A total of 221 patients were retrospectively included from our hospital. In addition, we recruited 94 and 100 participants as independent external validation sets from two other hospitals. Radiomics models of Gd-EOB-DTPA-enhanced MRI and diffusion-weighted imaging (DWI) were constructed and validated using machine learning. As presented in the radiomics nomogram, a prediction model was developed using multivariable logistic regression analysis, which included radiomics scores, radiologic features, and clinical features, such as the alpha-fetoprotein (AFP) level. The calibration, decision-making curve, and clinical usefulness of the radiomics nomogram were analyzed. The radiomic nomogram was validated using independent external cohort data. The areas under the receiver operating curve (AUC) were used to assess the predictive capability.Pathological examination confirmed MVI in 64 (28.9%), 22 (23.4%), and 16 (16.0%) of the 221, 94, and 100 patients, respectively. AFP, tumor size, non-smooth tumor margin, incomplete capsule, and peritumoral hypointensity in hepatobiliary phase (HBP) images had poor diagnostic value for MVI of sHCC. Quantitative radiomic features (1409) of MRI scans) were extracted. The classifier of logistic regression (LR) was the best machine learning method, and the radiomics scores of HBP and DWI had great diagnostic efficiency for the prediction of MVI in both the testing set (hospital A) and validation set (hospital B, C). The AUC of HBP was 0.979, 0.970, and 0.803, respectively, and the AUC of DWI was 0.971, 0.816, and 0.801 (AFP and conventional Gd-EOB-DTPA-enhanced MRI features have poor diagnostic accuracies for MVI in patients with sHCC. Machine learning with an LR classifier yielded the best radiomics score for HBP and DWI. The radiomics nomogram developed as a noninvasive preoperative prediction method showed favorable predictive accuracy for evaluating MVI in sHCC.

Background Preoperative differentiation of borderline from malignant epithelial ovarian tumors (BEOT vs. MEOT) is challenging and can significantly impact surgical management. Purpose To develop a multiple instance convolutional neural network (MICNN) that can differentiate BEOT from MEOT, and to compare its diagnostic performance with that of radiologists. Study type Retrospective study of eight clinical centers. Subjects Between January 2010 and June 2018, a total of 501 women (mean age, 48.93 ± 14.05 years) with histopathologically confirmed BEOT (N = 165) or MEOT (N = 336) were divided into the training (N = 342) and validation cohorts (N = 159). Field strength/sequence Three axial sequences from 1.5 or 3 T scanner were used: fast spin echo T2-weighted imaging with fat saturation (T2WI FS), echo planar diffusion-weighted imaging, and 2D volumetric interpolated breath-hold examination of contrast-enhanced T1-weighted imaging (CE-T1WI) with FS. Assessment Three monoparametric MICNN models were built based on T2WI FS, apparent diffusion coefficient map, and CE-T1WI. Based on these monoparametric models, we constructed an early multiparametric (EMP) model and a late multiparametric (LMP) model using early and late information fusion methods, respectively. The diagnostic performance of the models was evaluated using the receiver operating characteristic (ROC) curve and compared to the performance of six radiologists with varying levels of experience. Statistical tests We used DeLong test, chi-square test, Mann-Whitney U-test, and t-test, with significance level of 0.05. Results Both EMP and LMP models differentiated BEOT from MEOT, with an area under the ROC curve (AUC) of 0.855 (95% CI, 0.795-0.915) and 0.884 (95% CI, 0.831-0.938), respectively. The AUC of the LMP model was significantly higher than the radiologists' pooled AUC (0.884 vs. 0.797). Data conclusion The developed MICNN models can effectively differentiate BEOT from MEOT and the diagnostic performances (AUCs) were more superior than that of the radiologists' assessments. Level of evidence 3 TECHNICAL EFFICACY STAGE: 2.

To develop an MRI-based multi-lesion radiomics model for discrimination of relapsing-remitting multiple sclerosis (RRMS) and its mimicker neuropsychiatric systemic lupus erythematosus (NPSLE).A total of 112 patients with RRMS (n = 63) or NPSLE (n = 49) were assigned to training and test sets with a ratio of 3:1. All lesions across the whole brain were manually segmented on T2-weighted fluid-attenuated inversion recovery images. For each single lesion, 371 radiomics features were extracted and trained using machine learning algorithms, producing Radiomics Index for Lesion (RIL) for each lesion and a single-lesion radiomics model. Then, for each subject, single lesions were assigned to one of two disease courts based on their distance to decision threshold, and a Radiomics Index for Subject (RIS) was calculated as the mean RIL value of lesions on the higher-weighted court. Accordingly, a subject-level discrimination model was constructed and compared with performances of two radiologists.The subject-based discrimination model satisfactorily differentiated RRMS and NPSLE in both training (AUC = 0.967, accuracy = 0.892, sensitivity = 0.917, and specificity = 0.872) and test sets (AUC = 0.962, accuracy = 0.931, sensitivity = 1.000, and specificity = 0.875), significantly better than the single-lesion radiomics method (training: p0.001; test: p = 0.001) Besides, the discrimination model significantly outperformed the senior radiologist in the training set (training: p = 0.018; test: p = 0.077) and the junior radiologist in both the training and test sets (training: p = 0.008; test: p = 0.023).The multi-lesion radiomics model could effectively discriminate between RRMS and NPSLE, providing a supplementary tool for accurate differential diagnosis of the two diseases.• Radiomic features of brain lesions in RRMS and NPSLE were different. • The multi-lesion radiomics model constructed using a merging strategy was comprehensively superior to the single-lesion-based model for discrimination of RRMS and NPSLE. • The RRMS-NPSLE discrimination model showed a significantly better performance or a trend toward significance than the radiologists.

Objective To provide the reference biometric measurements of the normal foetal brain by prenatal MRI and describe the distribution of measurements in the foetuses with ventricular septal defect (VSD). Methods This retrospective study analysed the biometric measurements of 218 foetuses between 18 − 37 gestational weeks with normal MRI findings from July 2014 to August 2019, as well as 18 foetuses with VSD. The measurements included fronto-occipital diameter (FOD), biparietal diameter (BPD), and transverse cerebellar diameter (TCD). All the prenatal MRI examinations have been taken on the same 1.5 T MR unit with a standard protocol of the foetal brain. All the linear measurements of the foetal brain were obtained on the T2-weighted imaging. The distribution of measurements in 18 foetuses with VSD was plotted on centile curves. Results The reference data were presented in mean, standard deviation, 95% predicted confidence intervals, and the 3rd, 10th, 25th, 50th, 75th, 90th, 97th centiles at each gestational age. The value of TCD in 56% (10/18 cases) foetuses with VSD was lower than the 3rd centile, and the rate for FOD and BPD was 33% (6/18 cases) and 22% (4/18 cases) separately. On the curves, most VSD cases with measurements lower than the 3rd centile were in relatively early gestational stage (≤28 weeks). Conclusions We have presented reference linear biometry of the foetal brain by prenatal MRI from 18 to 37 gestational weeks, which could help us to interpret and monitor the brain development for foetuses with VSD and other congenital heart diseases.Key messages:We have presented reference linear biometry of the foetal brain by prenatal MRI from 18 to 37 gestational weeks in multiple statistical methods: mean and standard deviation; 95% predicted confidence intervals and the 3rd, 10th, 25th, 50th, 75th, 90th, 97th centiles.Our data showed that the involvement of the brain in VSD may be not globally, but regionally, and the cerebellum may be more possible to be involved.We speculated that the earlier the VSD diagnosed the worse the brain involved, which might suggest a poor outcome and necessary follow-up.

Background Obsessive-compulsive disorder (OCD) is increasingly conceptualized as a brain connectivity disorder. Recently, abnormalities in remote resting-state functional connectivity (FC) have been well demonstrated in the frontoparietal areas that linked impairments in large-scale intrinsic brain networks with aberrant fronto-striatal interactions. Beyond the remote FC abnormalities in OCD, many studies using regional homogeneity (ReHo) analysis have reported local FC alterations. However, their results were not entirely consistent. Methods We conducted a voxel-wise meta-analysis of ReHo studies to identify consistent local FC abnormalities in patients with OCD. A seed-based d mapping approach was used. Results Eight studies that compared 200 patients with OCD and 187 healthy controls were included. Increased ReHo in the lateral orbitofrontal cortices and dorsomedial prefrontal cortices bilaterally, and decreased ReHo in the fusiform gyri bilaterally were the most consistent and reliable findings in patients with OCD relative to healthy controls. Limitations The number of available studies included in the meta-analysis was relatively small. Many potential confounds on changes in ReHo warrant further attention. Conclusions These regions are critically implicated in the pathophysiology of OCD. Our findings in local FC alterations are complementary to the abnormalities in remote FC in OCD, contributing to the modeling of brain functional connectomes in OCD.

To develop and validate radiomic models in evaluating biological characteristics of rectal cancer based on multiparametric magnetic resonance imaging (MP-MRI). This study consisted of 345 patients with rectal cancer who underwent MP-MRI. We focused on evaluating five postoperative confirmed characteristics: lymph node (LN) metastasis, tumor differentiation, fraction of Ki-67-positive tumor cells, human epidermal growth factor receptor 2 (HER-2), and KRAS-2 gene mutation status. Data from 197 patients were used to develop the biological characteristics evaluation models. Radiomic features were extracted from MP-MRI and then refined for reproducibility and redundancy. The refined features were investigated for usefulness in building radiomic signatures by using two feature-ranking methods (MRMR and WLCX) and three classifiers (RF, SVM, and LASSO). Multivariable logistic regression was used to build an integrated evaluation model combining radiomic signatures and clinical characteristics. The performance was evaluated using an independent validation dataset comprising 148 patients. The MRMR and LASSO regression produced the best-performing radiomic signatures for evaluating HER-2, LN metastasis, tumor differentiation, and KRAS-2 gene status, with AUC values of 0.696 (95% CI, 0.610–0.782), 0.677 (95% CI, 0.591–0.763), 0.720 (95% CI, 0.621–0.819), and 0.651 (95% CI, 0.539–0.763), respectively. The best-performing signatures for evaluating Ki-67 produced an AUC value of 0.699 (95% CI, 0.611–0.786), and it was developed by WLCX and RF algorithm. The integrated evaluation model incorporating radiomic signature and MRI-reported LN status had improved AUC of 0.697 (95% CI, 0.612–0.781). Radiomic signatures based on MP-MRI have potential to noninvasively evaluate the biological characteristics of rectal cancer. • Radiomic features were extracted from MP-MRI images of the rectal tumor. • The proposed radiomic signatures demonstrated discrimination ability in identifying the histopathological, immunohistochemical, and genetic characteristics of rectal cancer. • All MRI sequences were important and could provide complementary information in radiomic analysis.

Background Differential diagnosis of primary central nervous system lymphoma (PCNSL) and glioblastoma (GBM) is useful to guide treatment strategies. Purpose To investigate the use of a convolutional neural network (CNN) model for differentiation of PCNSL and GBM without tumor delineation. Study type Retrospective. Population A total of 289 patients with PCNSL (136) or GBM (153) were included, the average age of the cohort was 54 years, and there were 173 men and 116 women. Field strength/sequence 3.0 T Axial contrast-enhanced T1 -weighted spin-echo inversion recovery sequence (CE-T1 WI), T2 -weighted fluid-attenuation inversion recovery sequence (FLAIR), and diffusion weighted imaging (DWI, b = 0 second/mm2 , 1000 seconds/mm2 ). Assessment A single-parametric CNN model was built using CE-T1 WI, FLAIR, and the apparent diffusion coefficient (ADC) map derived from DWI, respectively. A decision-level fusion based multi-parametric CNN model (DF-CNN) was built by combining the predictions of single-parametric CNN models through logistic regression. An image-level fusion based multi-parametric CNN model (IF-CNN) was built using the integrated multi-parametric MR images. The radiomics models were developed. The diagnoses by three radiologists with 6 years (junior radiologist Y.Y.), 11 years (intermediate-level radiologist Y.T.), and 21 years (senior radiologist Y.L.) of experience were obtained. Statistical analysis The 5-fold cross validation was used for model evaluation. The Pearson's chi-squared test was used to compare the accuracies. U-test and Fisher's exact test were used to compare clinical characteristics. Results The CE-T1 WI, FLAIR, and ADC based single-parametric CNN model had accuracy of 0.884, 0.782, and 0.700, respectively. The DF-CNN model had an accuracy of 0.899 which was higher than the IF-CNN model (0.830, P = 0.021), but had no significant difference in accuracy compared to the radiomics model (0.865, P = 0.255), and the senior radiologist (0.906, P = 0.886). Data conclusion A CNN model can differentiate PCNSL from GBM without tumor delineation, and comparable to the radiomics models and radiologists. Level of evidence 4 TECHNICAL EFFICACY: Stage 2.

Background: This study will evaluate the diagnostic accuracy of magnetic resonance imaging (MRI) to investigate tumor resectability at primary debulking surgery among women experiencing advanced-stage ovarian cancer. Methods: We will systematically search the randomized controlled trials (RCTs) for potentially eligible studies from electronic databases, including 4 English databases (PubMed, EMBASE, Web of Science, and Cochrane Library) and 3 Chinese databases (China National Knowledge Infrastructure, WanFang, and China Biomedical Database). The study language will be restricted to English and Chinese. Also, 2 independent authors will collect and carry out data extraction as well as quality assessment. Data will be synthesized using appropriate statistical methods. Results: We will summarize present study's evidence to assess the diagnostic accuracy of MRI for evaluating tumor resectability at primary debulking surgery in women experiencing advanced-stage ovarian cancer. Conclusion: The present study will put forward the latest high-quality evidence for MRI's clinical application for evaluating tumor resectability in women experiencing advanced ovarian cancer. Ethics and dissemination: Since the present study is a systematic review and meta-analysis based on the published literature, ethical approval will not be necessary. Protocol registration number: DOI 10.17605/OSF.IO/UWDRF (https://osf.io/uwdrf/)

Background: In developed nations, ovarian cancer has resulted in the most fatalities from gynecological cancer. Laparoscopy is primarily utilized as the test to diagnose ovarian cancer. Besides being costly, there are surgical risks associated with laparoscopies. At present, clinical practitioners have access to non-invasive tests for diagnosing ovarian cancer. This study aims to evaluate the diagnostic accuracy of magnetic resonance imaging (MRI) for diagnosing ovarian cancer. Methods: In order to obtain eligible studies, cross-sectional studies or randomized controlled trials are searched in electronic databases. The databases include 5 English databases (PubMed, the Cochrane Library, PsycINFO, EMBASE, and Web of Science) and 3 Chinese databases (China Biomedical Literature Database, China National Knowledge Infrastructure, and WanFang database). The databases are searched from their origin to October 2020. Quality Assessment of Diagnostic Accuracy Studies-2 is used to assess the methodological quality of the selected studies. RevMan 5.3 and SAS NLMIXED software are used to assess the data synthesis, sensitivity analysis, and risk of bias assessment. Results: This study evaluates the pooled diagnostic value of MRI for diagnosing ovarian cancer. Conclusions: This study will summarize previously published evidence of MRI in relation to diagnosing ovarian cancer. Ethics and dissemination: Since this study does not utilize data from patients, this protocol does not require ethical approval. Protocol registration number: DOI 10.17605/OSF.IO/A6SPQ (https://osf.io/a6spq)

To develop a preoperative MRI-based radiomic-clinical nomogram for prediction of residual disease (RD) in patients with advanced high-grade serous ovarian carcinoma (HGSOC).In total, 217 patients with advanced HGSOC were enrolled from January 2014 to June 2019 and randomly divided into a training set (n = 160) and a validation set (n = 57). Finally, 841 radiomic features were extracted from each tumor on T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted imaging (CE-T1WI) sequence, respectively. We used two fusion methods, the maximal volume of interest (MV) and the maximal feature value (MF), to fuse the radiomic features of bilateral tumors, so that patients with bilateral tumors have the same kind of radiomic features as patients with unilateral tumors. The radiomic signatures were constructed by using mRMR method and LASSO classifier. Multivariable logistic regression analysis was used to develop a radiomic-clinical nomogram incorporating radiomic signature and conventional clinico-radiological features. The performance of the nomogram was evaluated on the validation set.In total, 342 tumors from 217 patients were analyzed in this study. The MF-based radiomic signature showed significantly better prediction performance than the MV-based radiomic signature (AUC = 0.744 vs. 0.650, p = 0.047). By incorporating clinico-radiological features and MF-based radiomic signature, radiomic-clinical nomogram showed favorable prediction ability with an AUC of 0.803 in the validation set, which was significantly higher than that of clinico-radiological signature and MF-based radiomic signature (AUC = 0.623, 0.744, respectively).The proposed MRI-based radiomic-clinical nomogram provides a promising way to noninvasively predict the RD status.• MRI-based radiomic-clinical nomogram is feasible to noninvasively predict residual disease in patients with advanced HGSOC. • The radiomic signature based on MF showed significantly better prediction performance than that based on MV. • The radiomic-clinical nomogram showed a favorable prediction ability with an AUC of 0.803.

Background Gamma Knife radiosurgery (GKS) was recommended for treating patients with breast cancer brain metastasis (BCBM), but predictions of the existing prognostic models for therapeutic responsiveness vary substantially. Purpose To investigate the prognostic value of pretreatment clinical, MRI radiologic, and texture features in patients with BCBM undergoing GKS. Material and Methods The data of 81 BCBMs in 44 patients were retrospectively reviewed. Progressive disease was defined as an increase of at least 20% in the longest diameter of the target lesion or the presence of new intracranial lesions on contrast-enhanced T1-weighted (CE-T1W) imaging. Radiomic features were extracted from pretreatment CE-T1W images, T2-weighted (T2W) images, and ADC maps. Cox proportional hazard analyses were performed to identify independent predictors associated with BCBM-specific progression-free survival (PFS). A nomogram was constructed and its calibration ability was assessed. Results The cumulative BCBM-specific PFS was 52.27% at six months and 11.36% at one year, respectively. Age (hazard ratio [HR] 1.04; 95% confidence interval [CI] 1.01–1.06; P = 0.004) and CE-T1W-based kurtosis (HR 0.72; 95% CI 0.57–0.92; P = 0.008) were the independent predictors. The combination of CE-T1W-based kurtosis and age displayed a higher C-index (C-index 0.70; 95% CI 0.63–0.77) than did CE-T1W-based kurtosis (C-index 0.65; 95% CI 0.57–0.73) or age (C-index 0.63; 95% CI 0.56–0.70) alone. The nomogram based on the combinative model provided a better performance over age ( P Conclusion Pretreatment CE-T1W-based kurtosis combined with age could improve prognostic ability in patients with BCBM undergoing GKS.

Epithelial ovarian cancers (EOC) can be divided into type I and type II according to etiology and prognosis. Accurate subtype differentiation can substantially impact patient management. In this study, we aimed to construct an MR image-based radiomics model to differentiate between type I and type II EOC.In this multicenter retrospective study, a total of 294 EOC patients from January 2010 to February 2019 were enrolled. Quantitative MR imaging features were extracted from the following axial sequences: T2WI FS, DWI, ADC, and CE-T1WI. A combined model was constructed based on the combination of these four MR sequences. The diagnostic performance was evaluated by ROC-AUC. In addition, an occlusion test was carried out to identify the most critical region for EOC differentiation.The combined radiomics model exhibited superior diagnostic capability over all four single-parametric radiomics models, both in internal and external validation cohorts (AUC of 0.806 and 0.847, respectively). The occlusion test revealed that the most critical region for differential diagnosis was the border zone between the solid and cystic components, or the less compact areas of solid component on direct visual inspection.MR image-based radiomics modeling can differentiate between type I and type II EOC and identify the most critical region for differential diagnosis.• Combined radiomics models exhibited superior diagnostic capability over all four single-parametric radiomics models, both in internal and external validation cohorts (AUC of 0.834 and 0.847, respectively). • The occlusion test revealed that the most crucial region for differentiating type Ι and type ΙΙ EOC was the border zone between the solid and cystic components, or the less compact areas of solid component on direct visual inspection on T2WI FS. • The light-combined model (constructed by T2WI FS, DWI, and ADC sequences) can be used for patients who are not suitable for contrast agent use.

Background In unresectable hepatocellular carcinoma (HCC), methods to predict patients at increased risk of progression are required. Purpose To investigate the feasibility of radiomics model in predicting early progression of unresectable HCC after transcatheter arterial chemoembolization (TACE) therapy using preoperative multiparametric magnetic resonance imaging (MP-MRI). Study type Retrospective. Population A total of 84 patients with BCLC B stage HCC from one medical center. According to the modified response evaluation criteria in solid tumors, patients who progressed at 6 months after TACE therapy were assigned as the progressive disease (PD) group (n = 32). Patients whose MRI was performed on four devices were divided into a training cohort (n = 67). Patients whose MRI was performed on other than the previous four devices were used as the testing set (n = 17). Field strength/sequence 3.0T, 1.5T axial T2 -weighted imaging (T2 WI), diffusion-weighted imaging (DWI, b = 0, 500 s/mm2 ), and apparent diffusion coefficient (ADC) ASSESSMENT: PD was confirmed via imaging studies with MRI. Risk factors, including age, alpha fetoprotein (AFP), size, and radiomic-related features of PD were assessed. In addition, the discrimination ability of each radiomics signature was tested on an independent testing set. Statistical tests The area under the receiver-operator characteristic (ROC) curve (AUC) was used to evaluate the predictive accuracy of the radiomic signature in both the training and testing sets. The results indicated that the MP-MRI model achieved the greatest benefit. Results In the testing set, the model based on DWI features presented an AUC of (b = 0, 0.786; b = 500, 0.729), followed by T2 WI features (0.729) and ADC (0.714). The AUC of the MP-MRI signature was increased to 0.800 compared to any single MRI signature. The multivariate logistic analysis identified the radiomics signature as independent parameters of PD, while clinical information such as age, AFP, size, etc., had no significance in the PD group. Data conclusion Preoperative MP-MRI has the potential to predict the outcome of TACE therapy for unresectable HCC. In addition, these image features may be complementary to the current staging systems of HCC patients. Level of evidence 2. Technical efficacy stage 3. J. Magn. Reson. Imaging 2020;52:1083-1090.

BACKGROUND Preoperative differentiation of borderline from malignant epithelial ovarian tumors (BEOT from MEOT) can impact surgical management. MRI has improved this assessment but subjective interpretation by radiologists may lead to inconsistent results. PURPOSE To develop and validate an objective MRI-based machine-learning (ML) assessment model for differentiating BEOT from MEOT, and compare the performance against radiologists' interpretation. STUDY TYPE Retrospective study of eight clinical centers. POPULATION In all, 501 women with histopathologically-confirmed BEOT (n = 165) or MEOT (n = 336) from 2010 to 2018 were enrolled. Three cohorts were constructed: a training cohort (n = 250), an internal validation cohort (n = 92), and an external validation cohort (n = 159). FIELD STRENGTH/SEQUENCE Preoperative MRI within 2 weeks of surgery. Single- and multiparameter (MP) machine-learning assessment models were built utilizing the following four MRI sequences: T2 -weighted imaging (T2 WI), fat saturation (FS), diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC), and contrast-enhanced (CE)-T1 WI. ASSESSMENT Diagnostic performance of the models was assessed for both whole tumor (WT) and solid tumor (ST) components. Assessment of the performance of the model in discriminating BEOT vs. early-stage MEOT was made. Six radiologists of varying experience also interpreted the MR images. STATISTICAL TESTS Mann-Whitney U-test: significance of the clinical characteristics; chi-square test: difference of label; DeLong test: difference of receiver operating characteristic (ROC). RESULTS The MP-ST model performed better than the MP-WT model for both the internal validation cohort (area under the curve [AUC] = 0.932 vs. 0.917) and external validation cohort (AUC = 0.902 vs. 0.767). The model showed capability in discriminating BEOT vs. early-stage MEOT, with AUCs of 0.909 and 0.920, respectively. Radiologist performance was considerably poorer than both the internal (mean AUC = 0.792; range, 0.679-0.924) and external (mean AUC = 0.797; range, 0.744-0.867) validation cohorts. DATA CONCLUSION Performance of the MRI-based ML model was robust and superior to subjective assessment of radiologists. If our approach can be implemented in clinical practice, improved preoperative prediction could potentially lead to preserved ovarian function and fertility for some women. LEVEL OF EVIDENCE Level 4. TECHNICAL EFFICACY Stage 2. J. Magn. Reson. Imaging 2020;52:897-904.

Photodynamic therapy (PDT) has increasingly become an efficient and attractive cancer treatment modality based on reactive oxygen species (ROS) that can induce tumor death after irradiation with ultraviolet or visible light. Herein, to overcome the limited tissue penetration in traditional PDT, a novel near-infrared (NIR) light-activated NaScF4: 40% Yb, 2% Er@CaF2 upconversion nanoparticle (rUCNP) is successfully designed and synthesized. Chlorin e6, a photosensitizer and a chelating agent for Mn2+, is loaded into human serum albumin (HSA) that further conjugates onto rUCNPs. To increase the ability to target glioma tumor, an acyclic Arg–Gly–Asp peptide (cRGDyK) is linked to rUCNPs@HSA(Ce6-Mn). This nanoplatform enables efficient adsorption and conversion of NIR light (980 nm) into bright red emission (660 nm), which can trigger the photosensitizer Ce6-Mn complex for PDT and T1-weighted magnetic resonance imaging (T1-weighted MRI) for glioma diagnosis. Our in vitro and in vivo experiments demonstrate that NIR light-activated and glioma tumor-targeted PDT can generate large amounts of intracellular ROS that induce U87 cell apoptosis and suppress glioma tumor growth owing to the deep tissue penetration of irradiated light and excellent tumor-targeting ability. Thus, this nanoplatform holds potential for applications in T1-weighted MRI diagnosis and PDT of glioma for antitumor therapy. Statement of Significance A near-infrared (NIR) light-activated nanoplatform for photodynamic therapy (PDT) was designed and synthesized. The Red-to-Green (R/G) ratio of NaScF4: 40% Yb, 2% Er almost reached 9, a value that was much higher than that of a traditional Yb/Er-codoped upconversion nanoparticle (rUCNP). By depositing a CaF2 shell, the red-emission intensities of the rUCNPs were seven times strong as that of NaScF4: 40% Yb, 2% Er. The enhanced red-emitting rUCNPs could be applied in many fields such as bioimaging, controlled release, and real-time diagnosis. The nanoplatform had a strong active glioma-targeting ability, and all results achieved on subcutaneous glioma demonstrated that our NIR light-activated red-emitting upconverting nanoplatform was efficient for PDT. By loading Ce6-Mn complex into rUCNPs@HSA-RGD, the nanoplatform could be used as a T1-weighted magnetic resonance imaging agent for tumor diagnosis.

Segmentation of colorectal tumors is the basis of preoperative prediction, staging, and therapeutic response evaluation. Due to the blurred boundary between lesions and normal colorectal tissue, it is hard to realize accurate segmentation. Routinely manual or semi-manual segmentation methods are extremely tedious, time-consuming, and highly operator-dependent. In the framework of FCNs, a segmentation method for colorectal tumor was presented. Normalization was applied to reduce the differences among images. Borrowing from transfer learning, VGG-16 was employed to extract features from normalized images. We conducted five side-output blocks from the last convolutional layer of each block of VGG-16 along the network, these side-output blocks can deep dive multiscale features, and produced corresponding predictions. Finally, all of the predictions from side-output blocks were fused to determine the final boundaries of the tumors. A quantitative comparison of 2772 colorectal tumor manual segmentation results from T2-weighted magnetic resonance images shows that the average Dice similarity coefficient, positive predictive value, specificity, sensitivity, Hammoude distance, and Hausdorff distance were 83.56, 82.67, 96.75, 87.85%, 0.2694, and 8.20, respectively. The proposed method is superior to U-net in colorectal tumor segmentation (P 0.05). There is no difference between cross-entropy loss and Dice-based loss in colorectal tumor segmentation (P 0.05). The results indicate that the introduction of FCNs contributed to accurate segmentation of colorectal tumors. This method has the potential to replace the present time-consuming and nonreproducible manual segmentation method.

Malignant epithelial ovarian tumors (MEOTs) are the most lethal gynecologic malignancies, accounting for 90% of ovarian cancer cases. By contrast, borderline epithelial ovarian tumors (BEOTs) have low malignant potential and are generally associated with a good prognosis. Accurate preoperative differentiation between BEOTs and MEOTs is crucial for determining the appropriate surgical strategies and improving the postoperative quality of life. Multimodal magnetic resonance imaging (MRI) is an essential diagnostic tool. Although state-of-the-art artificial intelligence technologies such as convolutional neural networks can be used for automated diagnoses, their application have been limited owing to their high demand for graphics processing unit memory and hardware resources when dealing with large 3D volumetric data. In this study, we used multimodal MRI with a multiple instance learning (MIL) method to differentiate between BEOT and MEOT. We proposed the use of MAC-Net, a multiple instance convolutional neural network (MICNN) with modality-based attention (MA) and contextual MIL pooling layer (C-MPL). The MA module can learn from the decision-making patterns of clinicians to automatically perceive the importance of different MRI modalities and achieve multimodal MRI feature fusion based on their importance. The C-MPL module uses strong prior knowledge of tumor distribution as an important reference and assesses contextual information between adjacent images, thus achieving a more accurate prediction. The performance of MAC-Net is superior, with an area under the receiver operating characteristic curve of 0.878, surpassing that of several known MICNN approaches. Therefore, it can be used to assist clinical differentiation between BEOTs and MEOTs.

The purpose of this study was to evaluate the value of prenatal magnetic resonance imaging (MRI) in characterizing fetal intracranial space occupying lesions in comparison to prenatal ultrasound. This retrospective study included 50 fetuses (mean age 26 years, mean gestational weeks 31 + 1 GW) with intracranial space occupying lesions, suspected by prenatal screening ultrasound. T2-weighted, T1-weighted, SSFP, and diffusion-weighted sequences of the fetal brain were obtained on a 1.5 T unit. Pathology (n = 5), postmortem MRI (n = 3), or postnatal US (n = 42) was available as standard of reference. The fetal MRI provided correct diagnosis in 49 cases (98%), while 35 (70%) by ultrasound, and MRI failed in 1 case (2%), while ultrasound failed in 15 cases (30%). Fetal MR and ultrasound were concordant in 35 of 50 cases (70%), completely discordant in 4 (8%), and partially discordant in 11 (22%) cases. MRI could provide detailed information about the minor lesions, such as focal hemorrhage and periventricular nodules. Meanwhile, it could provide whole view of the lesion in order to delineate the surrounding anatomical structure. But there are still some limitations of its soft-tissue resolution in a case with teratoma; more effort is needed to improve the sequences.

Purpose Mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS), as a brain network disorder, has given increased importance to resective surgery. We aimed to examine the short-term postoperative functional network reorganization in patients with unilateral MTLE-HS using resting-state functional magnetic resonance imaging (Rs-fMRI) data with a longitudinal design. Method Seventeen (10 right-side and 7 left-side) patients with unilateral MTLE-HS were recruited. Rs-fMRI data were recorded from each subject before surgery and at an average of 4.5 months after surgery. Preoperative and postoperative amplitude of low-frequency fluctuations (ALFF) values as well as regions of interesting (ROI)-wise functional connectivity (FC) were compared to characterize the dynamic alterations of cerebral activity. Results After surgery, patients with unilateral MTLE-HS presented bilateral diffuse ALFF value changes, which were more widespread in the left-side resection group. Regarding the postoperative Rs-FC alterations, the left MTLE-HS patients presented an interhemispheric FC alteration in a relatively symmetric manner between the bilateral frontal lobe and putamen; the right MTLE-HS patients exhibited an intrahemispheric FC reduction in the basal ganglia-thalamocortical circuit. Conclusion Patients with unilateral MTLE-HS presented bilateral diffuse regional and interregional neural activity alteration after surgery, even over a short period, which exhibited a divergent pattern between the different sides of resection. This finding may contribute to the understanding of the effect and the mechanisms of surgery on the whole brain network in patients with unilateral MTLE-HS.

PURPOSE Accurate segmentation of rectal tumors is a basic and crucial task for diagnosis and treatment of rectal cancer. To avoid tedious manual delineation, an automatic rectal tumor segmentation model is proposed. METHODS A pretrained Resnet50 model was introduced for feature extraction. To reduce the complexity of the model, all layers after the 13th residual block of ResNet50 were removed, and three side-output modules were added to the hidden layer of ResNet50 to guide multiscale feature learning. The final boundaries of tumors were determined by fusion of the predictions from side-output modules. The proposed model was compared with two other models, and the effects of different region of interest (ROI) sizes, loss functions, and side-output fusion strategy were also evaluated. RESULTS The models were trained and evaluated on data from four clinical centers; T2-weighted magnetic resonance images (T2W-MRIs) from 461 patients in the first clinical center were used for training, while T2W-MRIs from 51 patients in the same clinical center and 56 patients in three other clinical centers were used for performance evaluation. The proposed model was superior to the two other models and achieved an average Dice similarity coefficient of 82.39%, sensitivity of 86.32%, specificity of 92.25%, and Hausdorff distance of 12.10 px. In addition, when the ROI contained rectal tumors, the smaller the ROI size, the higher the segmentation accuracy. For a certain ROI size, there were no considerable differences in segmentation results among the loss functions. Compared to the models with single side-output module, the proposed model performed better. CONCLUSIONS The results show that the proposed model has potential clinical applications in rectal cancer treatment, particularly with regard to therapeutic response evaluation and preoperative planning.

Increased expression of the 3.1 isoform of the KCNH2 potassium channel has been associated with cognitive dysfunction and with schizophrenia, yet little is known about the underlying pathophysiological mechanisms. Here, by using in vivo wireless local field potential recordings during working memory processing, in vitro brain slice whole-cell patching recordings and in vivo stereotaxic hippocampal injection of AAV-encoded expression, we identified specific and delayed disruption of hippocampal-mPFC synaptic transmission and functional connectivity associated with reductions of SERPING1, CFH, and CD74 in the KCNH2-3.1 overexpression transgenic mice. The differentially expressed genes in mice are enriched in neurons and microglia, and reduced expression of these genes dysregulates the complement cascade, which has been previously linked to synaptic plasticity. We find that knockdown of these genes in primary neuronal–microglial cocultures from KCNH2-3.1 mice impairs synapse formation, and replenishing reduced CFH gene expression rescues KCNH2-3.1-induced impaired synaptogenesis. Translating to humans, we find analogous dysfunctional interactions between hippocampus and prefrontal cortex in coupling of the fMRI blood oxygen level-dependent (BOLD) signal during working memory in healthy subjects carrying alleles associated with increased KCNH2-3.1 expression in brain. Our data uncover a previously unrecognized role of the truncated KCNH2-3.1 potassium channel in mediating complement activation, which may explain its association with altered hippocampal–prefrontal connectivity and synaptic function. These results provide a potential molecular link between increased KCNH2-3.1 expression, synapse alterations, and hippocampal–prefrontal circuit abnormalities implicated in schizophrenia.

Objective To build radiomic model in differentiating dissecting aneurysm (DA) from complicated saccular aneurysm (SA) based on high-resolution magnetic resonance imaging (HR-MRI) through machine-learning algorithm. Methods Overall, 851 radiomic features from 77 cases were retrospectively analyzed, and the ElasticNet algorithm was used to build the radiomic model. A clinico-radiological model using clinical features and conventional MRI findings was also built. An integrated model was then built by incorporating the radiomic model and clinico-radiological model. The diagnostic abilities of these models were evaluated using leave one out cross validation and quantified using the receiver operating characteristic (ROC) analysis. The diagnostic performance of radiologists was also evaluated for comparison. Results Five features were used to form the radiomic model, which yielded an area under the ROC curve (AUC) of 0.912 (95 % CI 0.846-0.976), sensitivity of 0.852, and specificity of 0.861. The radiomic model achieved a better diagnostic performance than the clinico-radiological model (AUC=0.743, 95 % CI 0.623-0.862), integrated model (AUC=0.888, 95 % CI 0.811-0.965), and even many radiologists. Conclusion Radiomic features derived from HR-MRI can reliably be used to build a radiomic model for effectively differentiating between DA and complicated SA, and it can provide an objective basis for the selection of clinical treatment plan.

To evaluate the efficiency of radiomics model on CT images of intratumoral and peritumoral lung parenchyma for preoperative prediction of lymph node (LN) metastasis in clinical stage T1 peripheral lung adenocarcinoma patients. Three hundred sixty-six peripheral lung adenocarcinoma patients with clinical stage T1 were evaluated using five CT scanners. For each patient, two volumes of interest (VOIs) on CT were defined as the gross tumor volume (GTV) and the peritumoral volume (PTV, 1.5 cm around the tumor). One thousand nine hundred forty-six radiomic features were obtained from each VOI, and then refined for reproducibility and redundancy. The refined features were investigated for usefulness in building radiomic signatures by mRMR feature ranking method and LASSO classifier. Multivariable logistic regression analysis was used to develop a radiomic nomogram incorporating the radiomic signature and clinical parameters. The prediction performance was evaluated on the validation cohort. The radiomic signatures using the features of GTV and PTV showed a good ability in predicting LN metastasis with an AUC of 0.829 (95% CI, 0.745–0.913) and 0.825 (95% CI, 0.733–0.918), respectively. By incorporating the features of GTV and PTV, the AUC of radiomic signature increased to 0.843 (95% CI, 0.770–0.916). The AUC of radiomic nomogram was 0.869 (95% CI, 0.800–0.938). Radiomic signatures of GTV and PTV both had a good prediction ability in the prediction of LN metastasis, and there is no significant difference of AUC between the two groups. The proposed nomogram can be conveniently used to facilitate the preoperative prediction of LN metastasis in T1 peripheral lung adenocarcinomas. • Radiomics from peritumoral lung parenchyma increase the efficiency of the prediction for lymph node metastasis in clinical stage T1 lung adenocarcinoma on CT. • A radiomic nomogram was developed and validated to predict LN metastasis. • Different scan parameters on CT showed that radiomics signature had good predictive performance.

OBJECTIVE: To evaluate how well different magnitudes of compression‐induced degenerative changes using a bent rat tail model simulated human lumbar lordosis. It has been shown that compression plays an important role in intervertebral disc degeneration (IDD). METHODS: Sprague–Dawley rats (n = 25) were instrumented with a special compressive apparatus that was used to bend the intervertebral disc between the 8th and the 10th caudal vertebral bodies using two Kirschner wires inserted percutaneously into the middle of two tail vertebrae. Then, rats were divided into five different static compression loads (control, sham, 1.8 N, 4.5 N, and 7.2 N). The degeneration of the discs was evaluated by magnetic resonance imaging (MRI), histology, gene expression of anabolism and catabolism after 2 weeks. We used the signal characteristics of the disc in T2‐weighted MRI to reflect the changes caused by degeneration as this is the most relevant and clinically recognized way to assess IDD. Pfirrmann classification was used to classify disc images. The tail discs from C(8–9) and C(9–10) with their two adjacent half vertebrae were carefully cut out and decalcified. Then the sections were paraffin‐embedded and cut into 5‐μm sections by histotome. Finally, they were stained with Safranin O‐Fast Green and hematoxylin, and hematoxylin and eosin, respectively. Images were taken using a microscope and staining and compression‐induced changes were assessed by a Masuda's grading scale. The relative expression levels of mRNA encoding rat anabolic genes and catabolic genes were evaluated by real‐time reverse transcription (RT)‐polymerase chain reaction (PCR). The mRNA expression fold change of the target gene was calculated using the 2(−ΔΔCt) method in the loaded and unloaded disc. RESULTS: As the loading magnitude increased, static compression produced a significantly progressive decrease in nucleus intensity on T2‐weighted MRI, a decrease of aggrecan and Type II collagen, an increase in Matrix metallopeptidase‐3 (MMP‐3) and MMP‐13 expressions, and a histomorphological degeneration. The sham group had a score of 1.4 ± 0.3, the 1.8 N group had a score of 2.4 ± 0.3, the 4.5 N group had a score of 3.2 ± 0.3, and the 7.2 N group had a score of 4.4 ± 0.3, which was based on the Pfirrmann classification score, in which the control group had a score of 1. These results demonstrated that the sham group was not significantly different from the control group. Histological analysis showed that in the loaded disc, the size of the nucleus was reduced and that the annular layer was disorganized. Based on the Masuda grading scale, scores were as follows: for the control group, 3.8 ± 0.35; sham, 4.2 ± 0.35; 1.8 N, 5.4 ± 0.35; 4.5 N, 7.6 ± 0.35; and 7.2 N, 10 ± 0.35. The gene expression was divided into the following: anabolic genes (aggrecan, collagen type1‐α1, and collagen type2‐α1) and catabolic genes (MMP‐3 and MMP‐13). Aggrecan and collagen type 2 were, respectively, downregulated from 0.42 ± 0.04 to 0.21 ± 0.04 and from 0.93 ± 0.06 to 0.17 ± 0.06 as the magnitude of compression increased, whereas collagen type 1 was significantly upregulated, from 2.49 ± 0.19 to 4.40 ± 0.19, when compared with the control group (from 1.8 to 7.2 N, P < 0.05). Catabolic genes MMP‐3 and MMP‐13 were significantly upregulated in all experimental groups (P < 0.05, MMP‐3: from 1.46 ± 0.18 to 3.44 ± 0.18; MMP‐13: from 1.19 ± 0.12 to 2.82 ± 0.13); however, MMP‐13 exhibited no significant changes but tended to be upregulated when compared with the 1.8 N group with the 4.5 N group. CONCLUSIONS: Different stresses led to different processes of degenerative changes, the concave disc degenerating more severely as stress gradually increased.

To solve the problem that the method based on tumor morphology or overall average parameters of tumor cannot conduct the early evaluation of tumor treatment response,we proposed a voxel-wise method.The voxel-wise method uses the method combining rigid and elastic registration algorithm to align the tumor area before and after treatment on the images which are acquired by the dynamic contrast enhanced magnetic resonance imaging(DCEMRI).We calculated voxel-wise volume transport constant(Ktrans)using pharmacokinetic model,and designed a threshold d to get the volume fraction of voxels which Ktrans increased significantly(F+),Ktrans decreased significantly(F-)or had no significant change(F0).Linear regression analysis was performed to get the correlation between volume fractions and pathological tumor cell necrosis rate(TCNR).We then determined the ability of volume fractions to evaluate treatment response at early stage by receiver operating characteristic(ROC)curve analysis.We performed experiments on 10 patients with soft tissue sarcomas.The results indicated that F- had significant negative correlation with TCNR(R2=0.832 8,P=0.000 2),F0 has significant positively correlation with TCNR(R2=0.788 4,P=0.000 6).In addition,F-(AUC=0.905,P=0.053),F0(AUC=0.857,P=0.087)had a good ability in early tumor treatment response evaluation.Therefore,F- and F0 can be used as effective imaging biomarkers for early evaluation of tumor treatment.

Summary: Background: There is a close relationship between the vertebral trabecular morphology and the condition of the associated disc. Objective: The relationship between disc degeneration and vertebral trabecular inhomogeneity is unclear. This study aimed to analyse the regional changes of vertebral trabecular morphology after disc degeneration. Methods: Thirty male Sprague–Dawley rats were randomly assigned to five groups. Group 1 served as an experimental group for the assessment of disc degeneration induced by needle puncture. Group 2 served as a sham group for trabecular morphology analysis. In Group 3, rats had their tail bent between the eighth and tenth coccygeal vertebrae. In Group 4, the tail of rats was bent with a compression load of 4.5 N. In Group 5, rats first underwent disc degeneration induced by a needle puncture before their tail was bent with a compressive load of 4.5 N. Magnetic resonance imaging was performed on all groups, and histological examination was performed on rodents from Group 1. The ninth coccygeal vertebrae of rats from Groups 2–5 were scanned by Micro-computed tomography. Trabecular morphologic changes were assessed in the concave and convex regions by bone volume fraction, trabecular number, trabecular thickness and trabecular separation. Results: Vertebral trabecular morphology in the concave region improved significantly, whereas the convex region was of significantly lower trabecular morphologic parameters with disc degeneration. The difference in trabecular morphologic parameters between the convex and concave regions increased significantly after disc degeneration. Conclusion: Disc degeneration promotes regional inhomogeneity in the vertebral trabecular morphology, with the convex region of the vertebrae having the worse trabecular bone morphology than the concave region. The translational potential of this article: Our study indicates that disc degeneration promotes regional inhomogeneity in the vertebral trabecular morphology. Regional variations in trabecular microarchitecture are helpful to predict vertebral fragility. This may help to elucidate the mechanisms by which disc degeneration contributes to vertebral fracture. Keywords: Animal model, Disc degeneration, Regional inhomogeneity, Trabecular morphology

Objective Exercise has a positive effect on physical fitness. Tai Chi Chuan is a traditional Chinese aerobic exercise. We assessed the effect of Tai Chi on the degeneration of lumbar vertebrae and lumbar discs with magnetic resonance images. Methods This retrospective cohort study involved 2 groups of participants: 27 Tai Chi practitioners with more than 4 years of experience with regular Tai Chi exercise and 24 sex- and age-matched participants without Tai Chi experience. The lumbar magnetic resonance images of all participants were collected. The numbers of degenerated lumbar vertebrae and lumbar discs were evaluated by the same radiologist, who was blind to the grouping. Results The Tai Chi practitioners had significantly fewer degenerated lumbar vertebrae (1.9) and lumbar discs (2.3) than the control group (2.6 and 2.9, respectively). The most severely affected lumbar vertebrae and discs were L5 and L4/L5, respectively. Conclusion Regular performance of the simplified Tai Chi 24 form could possibly retard the degeneration of lumbar vertebrae and lumbar discs in middle-aged and aged people.

Photodynamic therapy (PDT) has increasingly become an efficient and attractive cancer treatment modality based on reactive oxygen species (ROS) that can induce tumor death after irradiation with ultraviolet or visible light. Herein, to overcome the limited tissue penetration in traditional PDT, a novel near-infrared (NIR) light-activated NaScFA near-infrared (NIR) light-activated nanoplatform for photodynamic therapy (PDT) was designed and synthesized. The Red-to-Green (R/G) ratio of NaScF

To investigate the value of diffusion-weighted imaging (DWI) for assessing histopathologic changes observed in chronic kidney disease (CKD). Fifty-two patients with CKD underwent DWI before renal biopsy. The renal apparent diffusion coefficient (ADC) values and histopathologic changes were analyzed. The pathologic changes were scored using a semi-quantitative method (no lesion as 0, mild lesion as 1, moderate lesion as 2, and severe lesion as 3). The relationships between renal histopathologic scores, types, classification, and right renal ADCs were analyzed using ANOVA and Pearson’s or Spearman’s correlation tests. Negative correlations were found between the right renal ADCs and scores of tubulointerstitial lesions (r = − 0.354, P = 0.012), the severity of tubular atrophy (r = − 0.439, P = 0.002), and the severity of interstitial fibrosis (r = − 0.272, P = 0.049). There were no correlations between the ADCs and scores of glomeruli and peritubular vessel lesions (P > 0.05). There were significant differences among groups based on pathology types (P = 0.009). There was no significant relationship between renal ADCs and the pathologic classification (P > 0.05). DWI may be helpful to detect tubulointerstitial injury, including tubular atrophy and interstitial fibrosis. DWI may have the potential to serve as an effective auxiliary method to help nephrologists to evaluate patients with CKD.

Objective Seizure-related respiratory and cardiac dysfunctions were once thought to be the direct cause of sudden unexpected death in epilepsy (SUDEP), but both may be secondary to postictal cerebral inhibition. An important issue that has not been explored to date is the neural network basis of cerebral inhibition. Our aim was to investigate the features of neural networks in patients at high risk for SUDEP using a blood oxygen level-dependent (BOLD) resting-state functional connectivity (FC) approach. Subjects and methods Resting-state functional magnetic resonance imaging (Rs-fMRI) data were recorded from 13 patients at high risk for SUDEP and 12 patients at low risk for SUDEP. Thirteen cerebral regions that are closely related to cardiorespiratory activity were selected as regions of interest (ROIs). The ROI-wise resting-state FC analysis was compared between the two groups. Results Compared with patients at low risk for SUDEP, patients at high risk exhibited significant reductions in the resting-state FC between the pons and the right thalamus, the midbrain and the right thalamus, the bilateral anterior cingulate cortex (ACC) and the right thalamus, and the left thalamus and the right thalamus. Conclusions This investigation is the first to use neuroimaging methods in research on the mechanism of SUDEP and demonstrates the abnormally decreased resting-state FC in the ACC–thalamus–brainstem circuit in patients at high risk for SUDEP. These findings highlight the need to understand the fundamental neural network dysfunction in SUDEP, which may fill the missing link between seizure-related cardiorespiratory dysfunction and SUDEP, and provide a promising neuroimaging biomarker for risk prediction of SUDEP.

Autism spectrum disorder (ASD) has been widely recognized as a complex neurodevelopmental disorder. A large number of neuroimaging studies suggest abnormalities in brain structure and function of patients with ASD, but there is still no consistent conclusion. We sought to investigate both of the structural and functional brain changes in 3-7-year-old children with ASD compared with typically developing controls (TDs), and to assess whether these alterations are associated with autistic behavioral symptoms. Firstly, we applied an optimized method of voxel-based morphometry (VBM) analysis on structural magnetic resonance imaging (sMRI) data to assess the differences of gray matter volume between thirty-one autistic boys aged 3 to 7 and thirty-one age- and handness-matched male TDs. Secondly, we used clusters with between-group differences as seed regions to generate intrinsic functional connectivity maps based on resting-state functional connectivity magnetic resonance imaging (rs-fcMRI) in order to evaluate the functional impairments induced by structural alterations. Brain-behavior correlations were assessed among gray matter volume (GMV), functional connectivity and symptom severity in children with ASD. VBM analyses revealed increased GMV in left superior temporal gyrus (STG) and left postcentral gyrus (PCG) in ASD children, comparing with TDs. Using left PCG as a seed region, ASD children displayed significantly more positive connectivity with right angular gyrus (AG) and greater negative connectivity with right superior parietal gyrus (SPG) and right superior occipital gyrus (SOG), which were associated with the severity of symptoms in social interaction, communication and self-care ability. We suggest that stronger functional connectivity between left PCG and right AG, SPG and SOG detected in young boys with ASD may serve as important indicators of disease severity. Our study provided preliminary functional evidence that may underlie impaired higher-order multisensory integration in ASD children.

Background This study aims to describe the imaging characteristic of dural sinus malformation in 4 fetuses with different locations. Materials and Methods We report a series of 4 fetuses with dural sinus malformation in Hubei Maternal and Children's Hospital from July 2013 to February 2016. All the mothers undertook the prenatal magnetic resonance (MRI) imaging because of the intracranial space-occupying lesions discovered by prenatal ultrasound. Results Two of the 4 cases demonstrated typical MRI of dural sinus malformation with thrombosis in the vicinity of torcular herophili (also known as sinus confluence), whereas the other 2 cases showed lesions in the superior sagittal sinus and transverse sinus separately. Three pregnancies were terminated, whereas the other one was delivered in the local hospital. Conclusions Sonographer should realize the sign of dural sinus malformation. Atypical location of the dural sinus malformation, such as superior sagittal sinus and transverse sinus, should be paid special attention to. Further, prenatal MRI is necessary to identify the lesion.

Objective To find early predictors of histologic response in soft tissue sarcoma through volume transfer constant (Ktrans) analysis based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Materials and methods 11 Patients with soft tissue sarcoma of the lower extremity that underwent preoperative chemoradiotherapy followed by limb salvage surgery were included in this retrospective study. For each patient, DCE-MRI data sets were collected before and two weeks after therapy initiation, and histologic tumor cell necrosis rate (TCNR) was reported at surgery. The DCE-MRI volumes were aligned by registration. Then, the aligned volumes were used to obtain the Ktrans variation map. Accordingly, three sub-volumes (with increased, decreased or unchanged Ktrans) were defined and identified, and fractions of the sub-volumes, denoted as F+, F- and F0, respectively, were calculated. The predictive ability of volume fractions was determined by using area under a receiver operating characteristic curve (AUC). Linear regression analysis was performed to investigate the relationship between TCNR and volume fractions. In addition, the Ktrans values of the sub-volumes were compared. Results The AUC for F- (0.896) and F0 (0.833) were larger than that for change of tumor longest diameter ΔD (0.625) and the change of mean Ktrans Δ K t r a n s ¯ (0.792). Moreover, the regression results indicated that TCNR was directly proportional to F0 (R2 = 0.75, P = 0.0003), while it was inversely proportional to F- (R2 = 0.77, P = 0.0002). However, TCNR had relatively weak linear relationship with Δ K t r a n s ¯ (R2 = 0.64, P = 0.0018). Additionally, TCNR did not have linear relationship with DD (R2 = 0.16, P = 0.1246). Conclusion The volume fraction F- and F0 have potential as early predictors of soft tissue sarcoma histologic response.

Objective To determine whether adding diffusion-weighted imaging (DWI) to gadoxetic acid-enhanced 3.0T magnetic resonance imaging (MRI) can improve the detection of hepatocellular carcinoma (HCC), particularly for small lesions (≤2 cm) in patients with liver cirrhosis. Methods Data of patients diagnosed with focal liver lesions who had undergone gadoxetic acid-enhanced 3.0T MRI and DWI were retrospectively reviewed. Two radiologists (the observers) reviewed independently MRI images in two reading sessions, that is, gadoxetic acid-enhanced images alone and the combination of DWI (b values: 0 and 600 s/mm2) and gadoxetic acid-enhanced images. They assigned to each lesion a confidence level based on a five-point scale. The area under the receiver operating characteristic curve (AUROC), sensitivity and positive predictive value (PPV) for the detection of HCC were calculated. Results Both observers found the AUROC of the gadoxetic acid-enhanced images was slightly higher than that of the combined DWI and gadoxetic acid-enhanced MRI images in the detection of HCC (observer 1: 0.947 ± 0.030 vs 0.896 ± 0.042, Z = 1.478, P = 0.139; observer 2: 0.917 ± 0.038 vs 0.868 ± 0.048, Z = 1.296, P = 0.195). The sensitivity for the gadoxetic acid set alone was slightly higher than that for the combined set for observer 1 (97% vs 84%) and slightly lower for observer 2 (74% vs 82%). The PPVs were slightly higher for the gadoxetic acid set alone than for the combined set for both observers (observer 1, 89% vs 80%; observer 2, 93% vs 78%); however, none of the differences were statistically significant (P > 0.05). Conclusion There is no benefit in adding DWI to gadoxetic acid-enhanced MRI for the detection of HCC at 3.0T.