Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study.

One outstanding challenge for machine learning in diagnostic biomedical imaging is algorithm interpretability. A key application is the identification of subtle epileptogenic focal cortical dysplasias (FCDs) from structural MRI. FCDs are difficult to visualize on structural MRI but are often amenable to surgical resection. We aimed to develop an open-source, interpretable, surface-based machine-learning algorithm to automatically identify FCDs on heterogeneous structural MRI data from epilepsy surgery centres worldwide. The Multi-centre Epilepsy Lesion Detection (MELD) Project collated and harmonized a retrospective MRI cohort of 1015 participants, 618 patients with focal FCD-related epilepsy and 397 controls, from 22 epilepsy centres worldwide. We created a neural network for FCD detection based on 33 surface-based features. The network was trained and cross-validated on 50% of the total cohort and tested on the remaining 50% as well as on 2 independent test sites. Multidimensional feature analysis and integrated gradient saliencies were used to interrogate network performance. Our pipeline outputs individual patient reports, which identify the location of predicted lesions, alongside their imaging features and relative saliency to the classifier. On a restricted 'gold-standard' subcohort of seizure-free patients with FCD type IIB who had T1 and fluid-attenuated inversion recovery MRI data, the MELD FCD surface-based algorithm had a sensitivity of 85%. Across the entire withheld test cohort the sensitivity was 59% and specificity was 54%. After including a border zone around lesions, to account for uncertainty around the borders of manually delineated lesion masks, the sensitivity was 67%. This multicentre, multinational study with open access protocols and code has developed a robust and interpretable machine-learning algorithm for automated detection of focal cortical dysplasias, giving physicians greater confidence in the identification of subtle MRI lesions in individuals with epilepsy.

Aberrant intrinsic functional brain networks in patients with functional constipation.

PURPOSE: Patients with functional constipation (FCon) often suffer from mental and psychological problems. To explore the possible neurological interaction, we used resting-state functional magnetic imaging (RS-fMRI) to compare the alterations in intrinsic brain functional networks at multiple levels between patients with FCon and healthy controls (HC). METHODS: Twenty-eight patients with FCon and twenty-nine HC were recruited for a series of examinations and RS-fMRI. Both graph theory analysis and functional connectivity (FC) analysis were used to investigate brain functional alterations between the two groups. Correlation analyses were performed among neuropsychological scores, clinical indexes, and neuroimaging data. RESULTS: Compared with the HC, the assortativity showed significantly increased in global level in patients with FCon. In regional level, we found obviously increased nodal degree and nodal efficiency in somatosensory network (SMN), decreased nodal degree, and increased nodal efficiency in default mode network (DMN) in the FCon group. Furthermore, FC analysis demonstrated several functional alterations within and between the networks, particularly including the SMN and visual network (VN) in sub-network and large-scale network analysis. Moreover, correlation analysis indicated that nodal metrics and aberrant FC among functional brain networks were associated with emotion and scores of constipation in patients with FCon. CONCLUSION: All these findings reflect the differences in intrinsic brain functional networks between FCon and HC. Our study highlighted SMN, DMN, and VN as critical network and may be involved in the neurophysiology of FCon, which may contribute to improve personalized treatment in patients with FCon.

Atlas of lesion locations and postsurgical seizure freedom in focal cortical dysplasia: A MELD study.

OBJECTIVE: Drug-resistant focal epilepsy is often caused by focal cortical dysplasias (FCDs). The distribution of these lesions across the cerebral cortex and the impact of lesion location on clinical presentation and surgical outcome are largely unknown. We created a neuroimaging cohort of patients with individually mapped FCDs to determine factors associated with lesion location and predictors of postsurgical outcome. METHODS: The MELD (Multi-centre Epilepsy Lesion Detection) project collated a retrospective cohort of 580 patients with epilepsy attributed to FCD from 20 epilepsy centers worldwide. Magnetic resonance imaging-based maps of individual FCDs with accompanying demographic, clinical, and surgical information were collected. We mapped the distribution of FCDs, examined for associations between clinical factors and lesion location, and developed a predictive model of postsurgical seizure freedom. RESULTS: FCDs were nonuniformly distributed, concentrating in the superior frontal sulcus, frontal pole, and temporal pole. Epilepsy onset was typically before the age of 10 years. Earlier epilepsy onset was associated with lesions in primary sensory areas, whereas later epilepsy onset was associated with lesions in association cortices. Lesions in temporal and occipital lobes tended to be larger than frontal lobe lesions. Seizure freedom rates varied with FCD location, from around 30% in visual, motor, and premotor areas to 75% in superior temporal and frontal gyri. The predictive model of postsurgical seizure freedom had a positive predictive value of 70% and negative predictive value of 61%. SIGNIFICANCE: FCD location is an important determinant of its size, the age at epilepsy onset, and the likelihood of seizure freedom postsurgery. Our atlas of lesion locations can be used to guide the radiological search for subtle lesions in individual patients. Our atlas of regional seizure freedom rates and associated predictive model can be used to estimate individual likelihoods of postsurgical seizure freedom. Data-driven atlases and predictive models are essential for evidence-based, precision medicine and risk counseling in epilepsy.

Maximum lesions area and orthogonal values accessed from DWI images would be alternative imaging markers for predicting the outcome of acute ischemia in the middle cerebral artery territory.

BACKGROUND: Evaluating the acute ischemic volume on diffusion-weighted imaging (DWI) in the middle cerebral artery (MCA) territory would predict outcome. PURPOSE: To investigate the correlations between maximum area with restricted diffusion (MaxA), the orthogonal diameters (OD) as well as lesion volume on DWI, and to explore the role of MaxA and OD on predicting unfavorable outcome after an acute MCA ischemic stroke. MATERIAL AND METHODS: Sixty consecutive adult patients, including modified Rankin Scale score (mRS) ≤2 (n = 31) and mRS > 2 (n = 29) groups, were retrospectively enrolled. The MaxA and OD of lesions were assessed at the slice containing the largest infarction size on DWI images. We compared the prediction efficiencies of these methods on unfavorable outcomes. RESULTS: The correlation coefficients between the MaxA and infarction volume and OD and infarction volume were 0.982 ( P < 0.001) and 0.952 ( P < 0.001), respectively. The times required for measuring MaxA (150 s [130-160]) and OD (30 s [20-60]) were much shorter than that for infarction volume measurement (1240 s [180-1480]) ( P = 0.001, P = 0.004). With thresholds of ≥57.3 mL for infarction volume, ≥15.2 cm2 for MaxA, and ≥38.1 for the arithmetic product of OD, the AUCs of infarction volume, MaxA, and OD for predicting an unfavorable outcome were 0.818, 0.821, and 0.820, respectively. CONCLUSION: Since they correlated well with the infarction volume, MaxA and OD assessed on DWI were time-saving and achieved comparable diagnostic efficiencies; thus, they may represent alternative imaging markers for predicting unfavorable outcomes of acute ischemic stroke in MCA territory.

Cardiorespiratory fitness and brain volumes in men and women in the FINGER study.

Background: high cardiorespiratory fitness (CRF) is associated with larger brain volumes but data on sex differences in the association of CRF with brain volumes are scarce. We investigated whether the association of CRF with total grey matter (GM) and white matter volumes as well as medial temporal lobe and striatum volumes is different between men and women at increased risk for Alzheimer's disease (AD). Methods: we used baseline data from The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) in which the inclusion criteria were set to select individuals with cognitive performance at the mean level or slightly lower than expected for age according to Finnish population norms. Our sub-study included 39 randomly selected men and 29 women aged 61-75 years. CRF was assessed as peak oxygen consumption (VO2peak) measured in a maximal exercise test on cycle ergometer. Brain structural imaging was performed using a 1.5-T scanner. Results: in men, VO2peak was associated with cortical GM volume (ß = 0.56, P = 0.001) and total GM volume (ß = 0.54, P = 0.001). In women, no associations were found between VO2peak and brain volumes. VO2peak accounted for 23% and 1% of total variance of cortical GM volume as well as 25% and 4% of total variance of total GM volume in men and women, respectively. Conclusion: CRF is associated with cortical GM and total GM volumes in elderly men at increased risk for AD, but not in women.

Comparison of contrast-enhanced ultrasonography with virtual touch tissue quantification in the evaluation of focal liver lesions.

PURPOSE: To evaluate the diagnostic efficacy of virtual touch tissue quantification (VTQ) and contrast-enhanced ultrasonography (CEUS), separately and in combination, in diagnosing malignant focal liver lesions (FLLs). METHODS: Forty-six patients with 55 FLLs (28 benign and 27 malignant) underwent both VTQ and CEUS. The diagnostic values of VTQ and CEUS, alone and in combination, were compared. RESULTS: The diagnostic sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of CEUS were 92.6% (25/27), 96.4% (27/28), 94.5% (52/55), 96.2% (25/26), and 93.1% (27/29), respectively. The diagnostic sensitivity, specificity, accuracy, PPV, and NPV of VTQ with a cutoff of 2.22 m/s were 51.9% (14/27), 85.7% (24/28), 69.1% (38/55), 77.8% (14/18), and 64.9% (24/37), respectively. The diagnostic sensitivity, specificity, accuracy, PPV, and NPV of VTQ and CEUS combined were 96.3% (26/27), 82.1% (23/28), 89.1% (49/55), 83.9% (26/31), and 95.8% (23/24), respectively. Comparing the accuracies of the three methods, the diagnostic values of CEUS and of the combination of CEUS with VTQ were significantly higher than those of VTQ alone (p ≤ 0.01). There was no significant difference between the combination of CEUS with VTQ and CEUS (p = 0.49). CONCLUSIONS: CEUS is superior to VTQ in diagnosing malignant FLLs. Adding VTQ to CEUS did not improve the diagnosis of FLLs. © 2016 Wiley Periodicals, Inc. J Clin Ultrasound 44:347-353, 2016.

Functional magnetic resonance imaging reveals abnormal brain connectivity in EGR3 gene transfected rat model of schizophrenia.

Schizophrenia is characterized by the disorder of "social brain". However, the alternation of connectivity density in brain areas of schizophrenia patients remains largely unknown. In this study, we successfully created a rat model of schizophrenia by the transfection of EGR3 gene into rat brain. We then investigated the connectivity density of schizophrenia susceptible regions in rat brain using functional magnetic resonance imaging (fMRI) in combination with multivariate Granger causality (GC) model. We found that the average signal strength in prefrontal lobe and hippocampus of schizophrenia model group was significantly higher than the control group. Bidirectional Granger causality connection was observed between hippocampus and thalamic in schizophrenia model group. Both connectivity density and Granger causality connection were changed in prefrontal lobe, hippocampus and thalamus after risperidone treatment. Our results indicated that fMRI in combination with GC connection analysis may be used as an important method in diagnosis of schizophrenia and evaluation the effect of antipsychotic treatment. These findings support the connectivity disorder hypothesis of schizophrenia and increase our understanding of the neural mechanisms of schizophrenia.

The EADC-ADNI Harmonized Protocol for manual hippocampal segmentation on magnetic resonance: evidence of validity.

BACKGROUND: An international Delphi panel has defined a harmonized protocol (HarP) for the manual segmentation of the hippocampus on MR. The aim of this study is to study the concurrent validity of the HarP toward local protocols, and its major sources of variance. METHODS: Fourteen tracers segmented 10 Alzheimer's Disease Neuroimaging Initiative (ADNI) cases scanned at 1.5 T and 3T following local protocols, qualified for segmentation based on the HarP through a standard web-platform and resegmented following the HarP. The five most accurate tracers followed the HarP to segment 15 ADNI cases acquired at three time points on both 1.5 T and 3T. RESULTS: The agreement among tracers was relatively low with the local protocols (absolute left/right ICC 0.44/0.43) and much higher with the HarP (absolute left/right ICC 0.88/0.89). On the larger set of 15 cases, the HarP agreement within (left/right ICC range: 0.94/0.95 to 0.99/0.99) and among tracers (left/right ICC range: 0.89/0.90) was very high. The volume variance due to different tracers was 0.9% of the total, comparing favorably to variance due to scanner manufacturer (1.2), atrophy rates (3.5), hemispheric asymmetry (3.7), field strength (4.4), and significantly smaller than the variance due to atrophy (33.5%, P < .001), and physiological variability (49.2%, P < .001). CONCLUSIONS: The HarP has high measurement stability compared with local segmentation protocols, and good reproducibility within and among human tracers. Hippocampi segmented with the HarP can be used as a reference for the qualification of human tracers and automated segmentation algorithms.

Comparing predictors of conversion to Alzheimer's disease using the disease state index.

BACKGROUND: The Disease State Index (DSI) is a method which interprets data originating from multiple different sources, assisting the clinician in the diagnosis and follow-up of dementia diseases. OBJECTIVE: We compared the differences in accuracy in differentiating stable mild cognitive impairment (S-MCI) and progressive MCI (P-MCI) obtained from different data combinations using the DSI. METHODS: We investigated 212 cases with S-MCI and 165 cases with P-MCI from the Alzheimer's Disease Neuroimaging Initiative cohort. Data from neuropsychological tests, cerebrospinal fluid, apolipoprotein E (APOE) genotype, magnetic resonance imaging (MRI) and positron emission tomography (PET) were included. RESULTS: The combination of all parameters gave the highest accuracy (accuracy 0.70, sensitivity 0.71, specificity 0.68). In the different categories, neuropsychological tests (0.65, 0.65, 0.65) and hippocampal volumetry (0.66, 0.66, 0.66) achieved the highest accuracy. CONCLUSION: In addition to neuropsychological testing, MRI is recommended to be included for differentiating S-MCI from P-MCI. APOE genotype, CSF and PET may provide some additional information.

Novel LAMC3 pathogenic variant enriched in Finnish population causes malformations of cortical development and severe epilepsy.

OBJECTIVE: Recessive LAMC3 mutations are recognized to cause epilepsy with cortical malformations characterized by polymicrogyria and pachygyria. The objective of this study was to describe the clinical picture and epilepsy phenotype of four patients with a previously undescribed LAMC3 variant. METHODS: All epilepsy patients treated in Kuopio Epilepsy Center (located in Kuopio, Finland) are offered the possibility to participate in a scientific study investigating biomarkers in epilepsy (Epibiomarker study). We have collected a comprehensive database of the study population, and are currently re-evaluating our database regarding the patients with developmental and/or epileptic encephalopathy (DEE). If the etiology of epilepsy remains unknown in the clinical setting, we are performing whole exome sequencing to recognize the genetic causes. RESULTS: Among our study population of 323 DEE patients we recognized three patients with similar homozygous LAMC3 c.1866del (p.(Phe623Serfs*10)) frameshift variant and one patient with a compound heterozygous mutation where the same frameshift variant was combined with an intronic LAMC3 c.4231-12C>G variant on another allele. All these patients have severe epilepsy and either bilateral agyria-pachygyria or bilateral polymicrogyria in their clinical MRI scanning. Cortical malformations involve the occipital lobes in all our patients. Epilepsy phenotype is variable as two of our patients have DEE with epileptic spasms progressing to Lennox-Gastaut syndrome and intellectual disability. The other two patients have focal epilepsy without marked cognitive deficit. The four patients are unrelated. LAMC3 c.1866del p.(Phe623Serfs*10) frameshift variant is enriched in the Finnish population. SIGNIFICANCE: Only a few patients with epilepsy caused by LAMC3 homozygous or compound heterozygous mutations have been described in the literature. To our knowledge, the variants discovered in our patients have not previously been published. Clinical phenotype appears to be more varied than previously assumed and patients with a milder phenotype and normal cognition have probably remained unrecognized.

Refined imaging features of culprit plaques improve the prediction of recurrence in intracranial atherosclerotic stroke within the middle cerebral artery territory.

Recurrence is a significant adverse outcome of ischemic stroke (IS), particularly in cases of intracranial arteriosclerosis (ICAS). In this study, we investigated the impact of imaging features of culprit plaque using high-resolution magnetic resonance vessel wall imaging (HR-MR-VWI) on the prediction of IS recurrence. A total of 86 patients diagnosed with ICAS-related IS within the middle cerebral artery (MCA) territory were included, of which 23.25% experienced recurrent IS within one year. Our findings revealed significant differences between the recurrence and non-recurrence groups in terms of age (p = 0.007), diabetes mellitus (p = 0.031), hyperhomocysteinemia (p = 0.021), artery-artery embolism (AAE) infarction (p = 0.019), prominent enhancement (p = 0.013), and surface irregularity of the culprit plaque (p = 0.009). Age (HR = 1.063, p = 0.005), AAE infarction (HR = 5.708, p = 0.008), and prominent enhancement of the culprit plaque (HR = 4.105, p = 0.025) were identified as independent risk factors for stroke recurrence. The areas under the receiver operating characteristic curve (AUCs) for predicting IS recurrence using clinical factors, conventional imaging findings, HR-MR-VWI plaque features, and a combination of clinical and conventional imaging models were 0.728, 0.645, 0.705, and 0.814, respectively. Notably, the combination model demonstrated superior predictive performance with an AUC of 0.870. Similarly, AUC of combination model for predicting IS recurrence in validation cohort which enrolled another 37 patients was 0.865. In conclusion, the presence of obvious enhancement in culprit plaque on HR-MR-VWI is a valuable factor in predicting IS recurrence in ICAS-related strokes within the MCA territory. Furthermore, our combination model, incorporating plaque features, exhibited improved prediction accuracy.

Altered intrinsic functional brain architecture in patients with functional constipation: a surface-based network study.

Background: Functional constipation (FCon) is a common functional gastrointestinal disorder (FGID). Studies have indicated a higher likelihood of psychiatric disorders, such as anxiety, depression, sleep disturbances, and impaired concentration, among patients with FCon. However, the underlying pathophysiological mechanisms responsible for these symptoms in FCon patients remain to be fully elucidated. The human brain is a complex network architecture with several fundamental organizational properties. Neurological interactions between gut symptoms and psychiatric issues may be closely associated with these complex networks. Methods: In the present study, a total of 35 patients with FCon and 40 healthy controls (HC) were recruited for a series of clinical examinations and resting-state functional magnetic imaging (RS-fMRI). We employed the surface-based analysis (SBA) approach, utilizing the Schaefer cortical parcellation template and Tikhonov regularization. Graph theoretical analysis (GTA) and functional connectivity (FC) analysis of RS-fMRI were conducted to investigate the aberrant network alterations between the two groups. Additionally, correlation analyses were performed between the network indices and clinical variables in patients with FCon. Results: At the global level, we found altered topological properties and networks in patients with FCon, mainly including the significantly increased clustering coefficient (CP), local efficiency (Eloc), and shortest path length (LP), whereas the decreased global efficiency (Eglob) compared to HC. At the regional level, patients with FCon exhibited increased nodal efficiency in the frontoparietal network (FPN). Furthermore, FC analysis demonstrated several functional alterations within and between the Yeo 7 networks, particularly including visual network (VN), limbic network (LN), default mode network (DMN), and somatosensory-motor network (SMN) in sub-network and large-scale network analysis. Correlation analysis revealed that there were no significant associations between the network metrics and clinical variables in the present study. Conclusion: These results highlight the altered topological architecture of functional brain networks associated with visual perception abilities, emotion regulation, sensorimotor processing, and attentional control, which may contribute to effectively targeted treatment modalities for patients with FCon.

Education increases reserve against Alzheimer's disease--evidence from structural MRI analysis.

INTRODUCTION: The aim of this study was to determine whether years of schooling influences regional cortical thicknesses and volumes in Alzheimer's disease (AD), mild cognitive impairment (MCI), and healthy age-matched controls. METHODS: Using an automated image analysis pipeline, 33 regional cortical thickness and 15 regional volumes measures from MRI images were determined in 121 subjects with MCI, 121 patients with AD, and 113 controls from AddNeuroMed study. Correlations with years of schooling were determined and more highly and less highly educated subjects compared, controlling for intracranial volume, age, gender, country of origin, cognitive status, and multiple testing. RESULTS: After controlling for confounding factors and multiple testing, in the control group, subjects with more education had larger regional cortical thickness in transverse temporal cortex, insula, and isthmus of cingulate cortex than subjects with less education. However, in the AD group, the subjects with more education had smaller regional cortical thickness in temporal gyrus, inferior and superior parietal gyri, and lateral occipital cortex than the subjects with less education. No significant difference was found in the MCI group. CONCLUSION: Education may increase regional cortical thickness in healthy controls, leading to increased brain reserve, as well as helping AD patients to cope better with the effects of brain atrophy by increasing cognitive reserve.

Impaired white matter microstructure associated with severe depressive symptoms in patients with PD.

Depression is a common occurrence in patients with Parkinson's disease (PD); however, its pathophysiology is still unclear. This study assessed the association between the integrity of white matter and depressive symptoms in patients with PD. 67 patients with PD were divided into a non-depressed PD group (ndPD, n = 30) and a depressed PD group (dPD, n = 37). The dPD group was further subdivided into a mild-moderately depressed PD (mdPD, n = 22) and a severely depressed PD group (sdPD, n = 15). Tract-Based Spatial Statistics was used to compare fractional anisotropy (FA) between groups. Region-of-interest analysis was used to explore changes in diffusivity indices in the regions showing FA abnormalities. The sdPD patients exhibited significantly reduced FA in the left superior longitudinal fasciculus, uncinate fasciculus, anterior corona radiata, corticospinal tract, and bilateral inferior fronto-occipital fasciculus when compared with the ndPD patients, but the decreased FA was within a smaller area when compared with the mdPD patients. No significant difference in FA was found between the mdPD and ndPD groups. Among the dPD patients, FA values in the left superior longitudinal fasciculus negatively correlated with BDI scores. Impaired white matter integrity in the prefronto-limbic/temporal circuitry, mainly in the left hemisphere, is associated with severe, but not mild-moderate depressive symptoms in patients with PD.

Multivariate analysis of MRI data for Alzheimer's disease, mild cognitive impairment and healthy controls.

We have used multivariate data analysis, more specifically orthogonal partial least squares to latent structures (OPLS) analysis, to discriminate between Alzheimer's disease (AD), mild cognitive impairment (MCI) and elderly control subjects combining both regional and global magnetic resonance imaging (MRI) volumetric measures. In this study, 117 AD patients, 122 MCI patients and 112 control subjects (from the AddNeuroMed study) were included. High-resolution sagittal 3D MP-RAGE datasets were acquired from each subject. Automated regional segmentation and manual outlining of the hippocampus were performed for each image. Altogether this yielded volumes of 24 different anatomically defined structures which were used for OPLS analysis. 17 randomly selected AD patients, 12 randomly selected control subjects and the 22 MCI subjects who converted to AD at 1-year follow up were excluded from the initial OPLS analysis to provide a small external test set for model validation. Comparing AD with controls we found a sensitivity of 87% and a specificity of 90% using hippocampal measures alone. Combining both global and regional measures resulted in a sensitivity of 90% and a specificity of 94%. This increase in sensitivity and specificity resulted in an increase of the positive likelihood ratio from 9 to 15. From the external test set, the model predicted 82% of the AD patients and 83% of the control subjects correctly. Finally, 73% of the MCI subjects which converted to AD at 1 year follow-up were shown to resemble AD patients more closely than controls. This method shows potential for distinguishing between different patient groups. Combining the different MRI measures together resulted in a significantly better classification than using them separately. OPLS also shows potential for predicting conversion from MCI to AD.

New Enhancement beyond Radiation Field Improves Survival Prediction in Patients with Post-Treatment High-Grade Glioma.

The imaging signs which can accurately predict survival prognosis after standard treatment of high-grade glioma (HGG) are highly desirable. This study aims to explore the role of new enhancement beyond radiation field (NERF) in the survival prediction in patients with post-treatment HGG. The present study included 142 pathologically confirmed HGG patients who had received standard treatment. NERF, as well as other conventional MR findings and clinical variables, were included in univariate and multivariate analyses for evaluating their impactions on progression-free survival (PFS) and overall survival (OS). Univariate analysis showed that histological grade (p=0.008) and NERF (p=0.001) were the prognostic variables for poor PFS, whereas histological grade (p=0.017), NERF (p=0.001), and new subventricular zone enhancement (nSVZE) (p=0.001) were prognostic variables for poor OS. The multivariate analysis showed that NERF (HR 3.93; 95% CI 1.93-8.01; p=0.001) and nSVZE (HR 3.92; 95% CI 1.95-7.89; p=0.001) were the prognostic variables for poor OS. However, only nSVZE was (HR 3.29; 95% CI 2.04-5.28; p=0.001) the prognostic variable for poor PFS. When combining the NERF with the clinical and other MR variables, the highest AUC (0.924) and specificity (0.899) for predicting poor OS were achieved. The location of new developed enhancements relevant to high dose radiation field appears to be the main determinant of their prognostic value. Our results suggest that the new enhancement beyond radiation field can improve the survival prediction in patients with HGG after standard treatment.

FLAIR and ADC Image-Based Radiomics Features as Predictive Biomarkers of Unfavorable Outcome in Patients With Acute Ischemic Stroke.

At present, it is still challenging to predict the clinical outcome of acute ischemic stroke (AIS). In this retrospective study, we explored whether radiomics features extracted from fluid-attenuated inversion recovery (FLAIR) and apparent diffusion coefficient (ADC) images can predict clinical outcome of patients with AIS. Patients with AIS were divided into a training (n = 110) and an external validation (n = 80) sets. A total of 753 radiomics features were extracted from each FLAIR and ADC image of the 190 patients. Interquartile range (IQR), Wilcoxon rank sum test, and least absolute shrinkage and selection operator (LASSO) were used to reduce the feature dimension. The six strongest radiomics features were related to an unfavorable outcome of AIS. A logistic regression analysis was employed for selection of potential predominating clinical and conventional magnetic resonance imaging (MRI) factors. Subsequently, we developed several models based on clinical and conventional MRI factors and radiomics features to predict the outcome of AIS patients. For predicting unfavorable outcome [modified Rankin scale (mRS) > 2] in the training set, the area under the receiver operating characteristic curve (AUC) of ADC radiomics model was 0.772, FLAIR radiomics model 0.731, ADC and FLAIR radiomics model 0.815, clinical model 0.791, and clinical and conventional MRI model 0.782. In the external validation set, the AUCs for the prediction with ADC radiomics model was 0.792, FLAIR radiomics model 0.707, ADC and FLAIR radiomics model 0.825, clinical model 0.763, and clinical and conventional MRI model 0.751. When adding radiomics features to the combined model, the AUCs for predicting unfavorable outcome in the training and external validation sets were 0.926 and 0.864, respectively. Our results indicate that the radiomics features extracted from FLAIR and ADC can be instrumental biomarkers to predict unfavorable clinical outcome of AIS and would additionally improve predictive performance when adding to combined model.

Role of Dynamic Susceptibility Contrast Perfusion MRI in Glioma Progression Evaluation.

Accurately and quickly differentiating true progression from pseudoprogression in glioma patients is still a challenge. This study aims to explore if dynamic susceptibility contrast- (DSC-) MRI can improve the evaluation of glioma progression. We enrolled 65 glioma patients with suspected gadolinium-enhancing lesion. Longitudinal MRI follow-up (mean 590 days, range: 210-2670 days) or re-operation (n = 3) was used to confirm true progression (n = 51) and pseudoprogression (n = 14). We assessed the diagnostic performance of each MRI variable and the different combinations. Our results showed that the relative cerebral blood volume (rCBV) in the true progression group (1.094, 95%CI: 1.135-1.636) was significantly higher than that of the pseudoprogression group (0.541 ± 0.154) (p < 0.001). Among the 18 patients who had serial DSC-MRI, the rCBV of the progression group (0.480, 95%CI: 0.173-0.810) differed significantly from pseudoprogression (-0.083, 95%CI: -1.138-0.620) group (p=0.015). With an rCBV threshold of 0.743, the sensitivity and specificity for discriminating true progression from pseudoprogression were 76.5% and 92.9%, respectively. The Cho/Cr and Cho/NAA ratios of the true progression group (2.520, 95%CI: 2.331-2.773; 2.414 ± 0.665, respectively) were higher than those of the pseudoprogression group (1.719 ± 0.664; 1.499 ± 0.500, respectively) ((p=0.001), (p < 0.001), respectively). The areas under ROC curve (AUCs) of enhancement pattern, MRS, and DSC-MRI for the differentiation were 0.782, 0.881, and 0.912, respectively. Interestingly, when combined enhancement pattern, MRS, and DSC-MRI variables, the AUC was 0.965 and achieved sensitivity 90.2% and specificity 100.0%. Our results suggest that DSC-MRI can significantly improve the diagnostic performance for identifying glioma progression. DSC-MRI combined with conventional MRI may promptly distinguish true gliomas progression from pseudoprogression when the suspected gadolinium-enhancing lesion was found, without the need for a long-term follow-up.