MRI-based Classification Research Articles

Atherosclerosis Might Be Responsible for Branch Artery Disease: Evidence From White Matter Hyperintensity Burden in Acute Isolated Pontine Infarction.

Objective: To investigate an MRI-based etiological classification for acute isolated pontine infarcts and to assess differences in vascular risk factors, clinical characteristics and WMH burden among the etiological subtypes.Methods: All participants from SMART cohort with DWI-proven acute isolated pontine infarcts (AIPI) were included and categorized into 3 groups: large-artery-occlusive disease (LAOD), basilar artery branch disease (BAD), and small vessel disease (SVD), according to basilar artery atherosclerosis severity and lesion extent of the transverse axial plane. The vascular risk factors and 6-month functional outcome was analyzed among 3 groups.Results: Of the 1129 patients enrolled, 175 had AIPI. BAD was the most frequent subtype of AIPI (46.3%), followed by SVD (36.0%) and LAOD (17.7%). Neurological impairment on admission was more severe in the LAOD group, followed by BAD. The BAD group had greater frequencies of female sex, hypertension, diabetes mellitus compared to the SVD group (P < 0.05). NIHSS on admission were significantly higher in the BAD group as compared with the SVD group (P < 0.001), but no difference was found between BAD and LAOD group. Poor outcome (mRS≥3) was found in only 13.7% of patients at 6-month post-stroke and there was no difference among 3 groups. WMH severity was significant higher in the SVD group compared to the BAD group for the deep subcortical region; however, there was no difference for the periventricualr region. There was no significant difference in either DWMH or PVWMH severity between the BAD and LAOD groups.Conclusion: BAD is the most frequent etiology of AIPI followed by SVD and LAOD. WMH burden, vascular risk factors and clinical characteristics in BAD group were more similar to the LAOD group, rather than to the SVD group, suggesting the atherothrombotic nature of BAD.

Long-term results following intra-articular fractures of the medial malleolus in children and adolescents with special emphasis on MRI.

Presently, there is no publication combining clinical follow-up with magnetic resonance imaging (MRI) to determine possible post-traumatic alterations following paediatric intra-articular fractures of the medial malleolus. Therefore, the aims of this study were to retrospectively analyze a cohort of patients with Salter-Harris (SH) III and IV fractures of the medial malleolus and to evaluate their long-term outcome. MRI was used to assess possible changes of the articular surface that cannot be diagnosed on native radiographs. Fifty-four patients with SH III ( n = 38) or IV fractures ( n = 17) of the medial malleolus treated between 2001 and 2011 were invited for a follow-up examination. Clinical outcome was assessed with the Weber score and osteoarthritis with the Kellgren and Lawrence classification. MRI was rated by the Outerbridge classification. Correlations between the clinical and radiological outcomes were calculated. Seventeen patients were recruited for long-term follow-up at a mean of 112 (range, 65-184) months. The Weber score was very good for 5 patients, good for 10 patients and poor for 2 patients; the Kellgren and Lawrence score revealed a favourable grade 0 in 15 patients and grade 1 in 2 patients. The MRI-based Outerbridge classification yielded grade 0 for 12 patients, grade 1 for 1 patient, grade 2 for 2 patients and grade 3 for 1 patient. The Outerbridge score significantly ( p < 0.05) correlated with the Weber score. This study shows excellent and good outcome of SH III and IV fractures of the medial malleolus. Worse clinical outcome correlated with post-traumatic changes of the articular cartilage seen on MRI.

Interobserver and Intraobserver Reliability of an MRI-Based Classification System for Injuries to the Ulnar Collateral Ligament

Background: Despite improvements in understanding biomechanics and surgical options for ulnar collateral ligament (UCL) tears, there remains a need for a reliable classification of UCL tears that has the potential to guide clinical decision making. Purpose: To assess the intra- and interobserver reliability of the newly proposed magnetic resonance imaging (MRI)–based classification for UCL tears. Secondary objectives included assessing the effect of additional views, discrimination between distal and nondistal tears, and correlation of imaging reads with intraoperative findings of the UCL. Study Design: Cohort study (diagnosis); Level of evidence, 2. Methods: Nine fellowship-trained specialists from 7 institutions independently completed 4 surveys consisting of 60 elbow MRI scans with UCL tears using a newly proposed 6-stage classification system. The first and third surveys contained 60 coronal images, while the second and fourth contained the same images with coronal and axial views presented in a random order to assess intraobserver variability via the weighted kappa value and the effect of additional imaging views. Weighted kappa values were also calculated for each of the 4 surveys to acquire interobserver reliability. Reliability analysis was repeated through a 2-group classification analysis for distal and nondistal tears. Observer readings were compared with intraoperative UCL findings. Results: For the newly proposed 6-stage MRI-based classification, intra- and interobserver reliability demonstrated near perfect and substantial agreement, respectively. These values increased only when substratified into the 2-group distal and nondistal tear classification (P < .05). The additional axial view did not statistically improve the agreement within and among readers. When compared with intraoperative findings from 30 elbows, observer readings were accurate for tear grade (partial and complete), proximal location, and distal location but not midsubstance tears. Conclusion: The newly proposed 6-stage MRI-based classification utilizing grade and location of the injury had substantial to near perfect agreement among and within fellowship-trained observers.

Wednesday, September 26, 2018 1:00 PM – 2:00 PM Navigation and Intraoperative Monitoring: 44. A novel MRI-based classification of spinal cord shape and CSF presence at the curve apex to assess risk of intraoperative neuromonitoring data loss with thoracic spinal deformity correction

BACKGROUND CONTEXT Intraoperative loss of neuromonitoring data can be detrimental for both the patient and the surgeon. We propose a classification system to identify spinal cords at risk for data loss with thoracic deformity correction based on the preop axial MRI images at the curve apex. PURPOSE To test the hypothesis that the architecture of the spinal cord and amount of CSF on MRI at the thoracic apex can identify patients at risk for intraoperative loss of neuromonitoring data during spinal deformity correction. STUDY DESIGN/SETTING Review of thoracic deformity patients undergoing surgical correction at a single large academic center. PATIENT SAMPLE Thoracic deformity patients undergoing surgical correction. OUTCOME MEASURES Occurrence of intraoperative loss of trans-cranial neuromonitoring data. METHODS We reviewed 128 consecutive patients undergoing surgical correction of a thoracic deformity with pedicle screw/rod constructs. On preop MRI axial imaging at the apical concavity, three types of spinal cord/CSF architecture were defined. Type 1: circular cord with visible CSF between the cord and the apical concave pedicle. Type 2: circular cord but no visible CSF between it and the concave pedicle. Type 3: cord deformed against the apical concave pedicle, with no intervening CSF. RESULTS Of the 128 pts: 81 (63%) had Type 1; 32 (25%) Type 2; and 12 (11.7%) Type 3 spinal cords. Lower extremity transcranial motor-evoked potentials (TcMEPs) and/or somatosensory evoked potentials (SSEPs) were lost intraoperatively in 21 (16%) cases, with full recovery of data in 20 of those. On regression analysis, a Type 1 spinal cord was protective against intraoperative data loss (odds ratio = 0.17, p=.0003). A Type 2 spinal cord had no association with data loss (OR=0.66, p=.49). Type 3 spinal cord had significantly higher odds of intraoperative data loss (OR=28.3, p CONCLUSIONS This MRI-based spinal cord risk classification scheme identifies patients at risk of losing monitoring data during surgery. Patients with the spinal cord deformed against the apical concave pedicle (Type 3) have 28 times greater odds of losing monitoring data during surgery versus Type 1 (normal cord with adequate CSF) and Type 2 (normal cord without CSF between cord and concave pedicle). FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.

Assessment of Carotid Artery Plaque Components With Machine Learning Classification Using Homodyned-K Parametric Maps and Elastograms

Quantitative ultrasound (QUS) imaging methods, including elastography, echogenicity analysis, and speckle statistical modeling, are available from a single ultrasound (US) radio-frequency data acquisition. Since these US imaging methods provide complementary quantitative tissue information, characterization of carotid artery plaques may gain from their combination. Sixty-six patients with symptomatic ( n = 26 ) and asymptomatic ( n = 40 ) carotid atherosclerotic plaques were included in the study. Of these, 31 underwent magnetic resonance imaging (MRI) to characterize plaque vulnerability and quantify plaque components. US radio-frequency data sequence acquisitions were performed on all patients and were used to compute noninvasive vascular US elastography and other QUS features. Additional QUS features were computed from three types of images: homodyned-K (HK) parametric maps, Nakagami parametric maps, and log-compressed B-mode images. The following six classification tasks were performed: detection of 1) a small area of lipid; 2) a large area of lipid; 3) a large area of calcification; 4) the presence of a ruptured fibrous cap; 5) differentiation of MRI-based classification of nonvulnerable carotid plaques from neovascularized or vulnerable ones; and 6) confirmation of symptomatic versus asymptomatic patients. Feature selection was first applied to reduce the number of QUS parameters to a maximum of three per classification task. A random forest machine learning algorithm was then used to perform classifications. Areas under receiver-operating curves (AUCs) were computed with a bootstrap method. For all tasks, statistically significant higher AUCs were achieved with features based on elastography, HK parametric maps, and B-mode gray levels, when compared to elastography alone or other QUS alone ( ). For detection of a large area of lipid, the combination yielding the highest AUC (0.90, 95% CI 0.80-0.92, ) was based on elastography, HK, and B-mode gray-level features. To detect a large area of calcification, the highest AUC (0.95, 95% CI 0.94-0.96, ) was based on HK and B-mode gray level features. For other tasks, AUCs varied between 0.79 and 0.97. None of the best combinations contained Nakagami features. This study shows the added value of combining different features computed from a single US acquisition with machine learning to characterize carotid artery plaques.

Single-subject classification of presymptomatic frontotemporal dementia mutation carriers using multimodal MRI.

BackgroundClassification models based on magnetic resonance imaging (MRI) may aid early diagnosis of frontotemporal dementia (FTD) but have only been applied in established FTD cases. Detection of FTD patients in earlier disease stages, such as presymptomatic mutation carriers, may further advance early diagnosis and treatment. In this study, we aim to distinguish presymptomatic FTD mutation carriers from controls on an individual level using multimodal MRI-based classification.MethodsAnatomical MRI, diffusion tensor imaging (DTI) and resting-state functional MRI data were collected in 55 presymptomatic FTD mutation carriers (8 microtubule-associated protein Tau, 35 progranulin, and 12 chromosome 9 open reading frame 72) and 48 familial controls. We calculated grey and white matter density features from anatomical MRI scans, diffusivity features from DTI, and functional connectivity features from resting-state functional MRI. These features were applied in a recently introduced multimodal behavioural variant FTD (bvFTD) classification model, and were subsequently used to train and test unimodal and multimodal carrier-control models. Classification performance was quantified using area under the receiver operator characteristic curves (AUC).ResultsThe bvFTD model was not able to separate presymptomatic carriers from controls beyond chance level (AUC = 0.570, p = 0.11). In contrast, one unimodal and several multimodal carrier-control models performed significantly better than chance level. The unimodal model included the radial diffusivity feature and had an AUC of 0.646 (p = 0.021). The best multimodal model combined radial diffusivity and white matter density features (AUC = 0.680, p = 0.005).ConclusionsFTD mutation carriers can be separated from controls with a modest AUC even before symptom-onset, using a newly created carrier-control classification model, while this was not possible using a recent bvFTD classification model. A multimodal MRI-based classification score may therefore be a useful biomarker to aid earlier FTD diagnosis. The exclusive selection of white matter features in the best performing model suggests that the earliest FTD-related pathological processes occur in white matter.

A Novel MRI-Based Classification System for Tibial Eminence Fractures in Pediatric Patients: An Improvement from Meyers and McKeever?

The most common system for classifying pediatric tibial spine fractures is based on X-Rays. We propose an MRI-based system. Our results show a moderate intra-rater and fair inter-rater reliability of the new system. In 1/3 of our cases, assessment using MRI substantially modified the treatment. This new system may be a helpful tool for planning the treatment of tibial spine fractures in children. The Meyers and McKeever Classification (MMC) is typically used to classify pediatric tibial spine fractures on plain radiographs. While MRI images can provide important information about the fractures, there is currently no system for classifying these fractures on MRI. The aims of this study were: (1) propose a new MRI-based system for classifying tibial spine fractures and (2) assess how often using an MRI-based system changes the treatment plan compared to the MMC, which is based on X-Rays. Our hypothesis is that an MRI-based system allows for more precise identification and treatment of tibial spine fractures. The new MRI-based system has three grades.-Grade I fractures are defined as non- or minimally-displaced fractures with < 2 mm of displacement.-Grade II fractures are posterior-hinged fractures with > 2 mm displacement of the anterior aspect of the fracture and < 2 mm displacement of the posterior aspect of the fragment.-Grade III fractures are displaced fractures (> 2 mm of displacement of the posterior aspect of the fragment), any tibial spine fracture that results in meniscal entrapment (where the meniscus or intra-meniscal ligament is underneath a fracture fragment), or any tibial spine fracture extending to the weight-bearing surface of the medial or lateral tibial plateau with > 2 mm of displacement. We considered three scenarios as “changes in treatment”: when use of the MRI-system (1) allowed raters to identify a fracture that had been graded as “no fracture” using X-Rays, (2) changed the treatment from non-operative to operative, and (3) changed the treatment from operative to non-operative. The number of cases that experienced a change was reported as raw counts and percentages. Four independent observers (one attending orthopaedic surgeon, one attending radiologist, a senior orthopaedic surgery resident, and a junior orthopaedic surgery resident) classified the images of 20 patients according to the MMC using plain X-Rays. Each observer graded images at two time points at least one week apart. Next, raters repeated their assessments on MRIs using the MRI-based system. The inter-rater and intra-rater reliability of each system was assessed (Stata 14.0 software). Based on the Altman threshold, there was moderate intra-rater correlation between the observers when using MRI-based classification (avg = 0.60; range = 0.45-0.75) and a fair inter-rater correlation (k = 0.33). Following the evaluation of MRIs with the new system, we observed “changes in treatment” in 33% (26 of 80) of grades after the first grading and 34% (27 of 80) after the second compared to the initial evaluation using X-Rays. In 11 grades (20%), MRI recognized previously unnoticed fractures; in 21 grades (40%), MRI assessment upgraded the lesion to operative treatment; in 21 grades (40%) formerly operative candidates were re-assigned to conservative treatment. There is moderate intra-rater and fair inter-rater reliability of the new system. In one third of the cases, assessment using MRI substantially modified the treatment recommendations. This MRI-based system may become a helpful tool for diagnosing and planning the treatment of tibial spine fractures in children.

Functional outcome analysis of lumbar canal stenosis patients post decompression and posterior stabilization with stenosis grading using magnetic resonance imaging

Lumbar canal stenosis (LCS) is a condition that can potentially cause disability. It often occurs in aging populations. The aim of this study was to analyze the correlation between the clinical outcomes of postoperative patients and classifications that were based on MRI assessments. This prospective cohort study was carried out at Cipto Mangunkusumo General Hospital from January to July 2016 using consecutive sampling. Thirty-eight patient samples were obtained, all of whom were managed with the same surgical technique of decompression and posterior stabilization. The patients were categorized in four types based on MRI examination using the Schizas classification. Pre- and post-treatment (three months and six months) assessments of the patients were conducted according to Visual Analogue Scale (VAS), the Oswestry Disability Index (ODI), the Japanese Orthopedic Association Score (JOA), and the Roland-Morris Disability Questionnaire (RMDQ). The statistical analysis was performed using the statistical program for social science (SPSS) v.19. The average age of the patients in this sample was 58.92 years (range 50–70 years). There were 16 males and 22 females. Most patients were classified as type C (21 subjects) based on MRI examination. The improvement in the clinical scores of male subjects was better than in the female subjects. Significant differences were found in the six-month postoperative VAS (p = 0.003) and three-month postoperative JOA scores (p = 0.029). The results at follow-up showed that the VAS, ODI, JOA and RMDQ scores were improved. There were no statistical differences between the MRI-based classification and the clinical outcomes at preoperative, three and six months postoperative according to VAS (p = 0.451, p = 0.738, p = 0.448), ODI (p = 0.143, p = 0.929, p = 0.796), JOA (p = 0.157, p = 0.876, p = 0.961), and RMDQ (p = 0.065, p = 0.057, p = 0.094). There was clinical improvement after decompression and posterior stabilization in lumbar canal stenosis, which was manifested at three and six month post-operation follow-up in the VAS ODI, JOA and RMDQ scores. There was no association between the degree of LCS and VAS, ODI, JOA and RMDQ scores.

The relevance of feature selection methods to the classification of obsessive-compulsive disorder based on volumetric measures

BackgroundMagnetic resonance images (MRI) show detectable anatomical and functional differences between individuals with obsessive-compulsive disorder (OCD) and healthy subjects. Moreover, machine learning techniques have been proposed as tools to identify potential biomarkers and, ultimately, to support clinical diagnosis. However, few studies to date have investigated feature selection (FS) influences in OCD MRI-based classification. MethodsVolumes of cortical and subcortical structures, from MRI data of 38 OCD patients (split into two groups according symptoms severity) and 36 controls, were submitted to seven feature selection algorithms. FS aims to select the most relevant and less redundant features which discriminate between two classes. Then, a classification step was applied, from which the classification performances before and after different FS were compared. For the performance evaluation, leave-one-subject-out accuracies of Support Vector Machine classifiers were considered. ResultsUsing different FS algorithms, performance improvement was achieved for Controls vs. All OCD discrimination (19.08% of improvement reducing by 80% the amount of features), Controls vs. Low OCD (20.10%, 75%), Controls vs. High OCD (17.32%, 85%) and Low OCD vs. High OCD (10.53%, 75%). Furthermore, all algorithms pointed out classical cortico-striato-thalamo-cortical circuitry structures as relevant features for OCD classification. LimitationsLimitations include the sample size and using only filter approaches for FS. ConclusionsOur results suggest that FS positively impacts OCD classification using machine-learning techniques. Complementarily, FS algorithms were able to select biologically plausible features automatically.

MRI-Based Classification Models in Prediction of Mild Cognitive Impairment and Dementia in Late-Life Depression.

Objective: Late-life depression (LLD) is associated with development of different types of dementia. Identification of LLD patients, who will develop cognitive decline, i.e., the early stage of dementia would help to implement interventions earlier. The purpose of this study was to assess whether structural brain magnetic resonance imaging (MRI) in LLD patients can predict mild cognitive impairment (MCI) or dementia 1 year prior to the diagnosis.Methods: LLD patients underwent brain MRI at baseline and repeated clinical assessment after 1-year. Structural brain measurements were obtained using Freesurfer software (v. 5.1) from the T1W brain MRI images. MRI-based Random Forest classifier was used to discriminate between LLD who developed MCI or dementia after 1-year follow-up and cognitively stable LLD. Additionally, a previously established Random Forest model trained on 185 patients with Alzheimer’s disease (AD) vs. 225 cognitively normal elderly from the Alzheimer’s disease Neuroimaging Initiative was tested on the LLD data set (ADNI model).Results: MCI and dementia diagnoses were predicted in LLD patients with 76%/68%/84% accuracy/sensitivity/specificity. Adding the baseline Mini-Mental State Examination (MMSE) scores to the models improved accuracy/sensitivity/specificity to 81%/75%/86%. The best model predicted MCI status alone using MRI and baseline MMSE scores with accuracy/sensitivity/specificity of 89%/85%/90%. The most important region for all the models was right ventral diencephalon, including hypothalamus. Its volume correlated negatively with the number of depressive episodes. ADNI model trained on AD vs. Controls using SV could predict MCI-DEM patients with 67% accuracy.Conclusion: LDD patients developing MCI and dementia can be discriminated from LLD patients remaining cognitively stable with good accuracy based on baseline structural MRI alone. Baseline MMSE score improves prediction accuracy. Ventral diencephalon, including the hypothalamus might play an important role in preservation of cognitive functions in LLD.

Learning MRI-based classification models for MGMT methylation status prediction in glioblastoma

Background and ObjectiveThe O6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation has been shown to be associated with improved outcomes in patients with glioblastoma (GBM) and may be a predictive marker of sensitivity to chemotherapy. However, determination of the MGMT promoter methylation status requires tissue obtained via surgical resection or biopsy. The aim of this study was to assess the ability of quantitative and qualitative imaging variables in predicting MGMT methylation status noninvasively. MethodsA retrospective analysis of MR images from GBM patients was conducted. Multivariate prediction models were obtained by machine-learning methods and tested on data from The Cancer Genome Atlas (TCGA) database. ResultsThe status of MGMT promoter methylation was predicted with an accuracy of up to 73.6%. Experimental analysis showed that the edema/necrosis volume ratio, tumor/necrosis volume ratio, edema volume, and tumor location and enhancement characteristics were the most significant variables in respect to the status of MGMT promoter methylation in GBM. ConclusionsThe obtained results provide further evidence of an association between standard preoperative MRI variables and MGMT methylation status in GBM.

Haglund’s syndrome: classification system and surgical protocol

Background: Haglund’s deformity is not an uncommon cause of retrocalcaneal pain. Currently, there is neither a classification system nor a standard surgical protocol for this condition. The purpose of this study was to evaluate the outcome after Achilles tendon debridement, partial detachment, and anchor reattachment of the tendon, osteotomy of the Haglund deformity, and excision of the retrocalcaneal bursa. An MRI-based classification system is proposed that relies on the extent of Achilles tendon pathology in the presence of the Haglund deformity. Methods: Fifty heels in 46 patients were surgically treated with the proposed surgical protocol. Patients were evaluated using the Tegner Activity Level and the Foot and Ankle Outcome (FAOS) scores. Results: The mean Tegner Activity Level score improved from a preoperative mean of 3.44 to a mean of 6.56 at 6 mo, then to a mean of 6.84 at 12 mo follow-up. The mean FAOS score improved from 47.60 preoperatively to a mean of 66.50 at 6 mo and to a mean of 79.49 at 12 mo follow-up. Conclusions: Surgical treatment of Haglund’s deformity with tendon debridement, partial detachment, calcaneal osteotomy, and bursal debridement provides good results and functional outcomes that continues to improve over 12 mo follow-up.

Magnetic resonance imaging–based classification for ulnar collateral ligament injuries of the elbow

Magnetic resonance imaging (MRI) arthrography has been considered the gold standard for imaging ulnar collateral ligament (UCL) injuries. No classification system has been described for UCL tears to help discuss and guide treatment options. We propose that an MRI-based UCL classification system would correlate with valgus laxity and help predict surgical management. The MRIs from 240 patients who underwent UCL reconstruction were reviewed and classified according to a system based on severity and location of the UCL injury. Bilateral elbow valgus stress radiographs were used to compare the amount of valgus laxity associated with each grade of UCL injury. Additional data from the imaging review included intraligamentous calcifications, T-sign presence, and a newly described "vacuum sign." The amount of valgus stress opening increased with the classification severity as follows: type I, 0.13 mm; type II, 0.20 mm; type III, 0.63 mm; and type IV, 0.76 mm. We propose a new classification for UCL injuries based on MRI findings that helps predict valgus laxity, improve communication, and guide treatment for UCL pathology in throwing athletes.

Position statement: the epidemiology, pathogenesis and risk factors of osteoarthritis of the knee

Osteoarthritis (OA) represents a large burden on healthcare resources worldwide with continually increasing prevalence. This has led to renewed interest in the causes and pathogenesis of the condition. In recent years, there has been a move away from a simple ‘wear and tear’ model of cartilage to one of a complex inflammatory process involving cellular and extracellular derangements that allow a catabolic state to dominate. Ultimately, OA is now seen as pan-joint disease involving synovium, menisci, ligaments and muscle, in addition to cartilage. There are several classification systems including radiographic, MRI-based, clinical and combined classification systems. As radiographs only detect OA in latter stages, there has been a focus on early diagnosis using MRI and serum biomarkers. New physiological MRI sequences can now measure the proteoglycan content in cartilage and new semiquantitative analyses have been developed to score total knee joint involvement in the disease process. Serum biomarkers can be divided into those that are collagen breakdown products and those that are inflammatory cytokines; these can be used in early detection of OA before radiographic appearances arise. The risk factors for OA include ageing, knee injury, obesity, altered limb alignment, impaired muscle strength, female gender, heavy physical work and genetic susceptibility. Research continues to identify the mechanisms involved that lead to OA development, with possibly unique processes underpinning each risk factor. Our understanding of the pathophysiology of OA will continue to improve in the next few years which should lead to new intervention strategies that target different processes. More informative MRI sequences will continue to be developed and the optimum combination of biomarkers to detect early OA will need to be identified. Genetic studies will continue to identify new susceptibility loci that could be targeted in future therapies.

Multimodal MRI-Based Classification of Trauma Survivors with and without Post-Traumatic Stress Disorder.

Post-traumatic stress disorder (PTSD) is a debilitating psychiatric disorder. It can be difficult to discern the symptoms of PTSD and obtain an accurate diagnosis. Different magnetic resonance imaging (MRI) modalities focus on different aspects, which may provide complementary information for PTSD discrimination. However, none of the published studies assessed the diagnostic potential of multimodal MRI in identifying individuals with and without PTSD. In the current study, we investigated whether the complementary information conveyed by multimodal MRI scans could be combined to improve PTSD classification performance. Structural and resting-state functional MRI (rs-fMRI) scans were conducted on 17 PTSD patients, 20 trauma-exposed controls without PTSD (TEC) and 20 non-traumatized healthy controls (HC). Gray matter volume (GMV), amplitude of low-frequency fluctuations (ALFF), and regional homogeneity were extracted as classification features, and in order to integrate the information of structural and functional MRI data, the extracted features were combined by a multi-kernel combination strategy. Then a support vector machine (SVM) classifier was trained to distinguish the subjects at individual level. The performance of the classifier was evaluated using the leave-one-out cross-validation (LOOCV) method. In the pairwise comparison of PTSD, TEC, and HC groups, classification accuracies obtained by the proposed approach were 2.70, 2.50, and 2.71% higher than the best single feature way, with the accuracies of 89.19, 90.00, and 67.57% for PTSD vs. HC, TEC vs. HC, and PTSD vs. TEC respectively. The proposed approach could improve PTSD identification at individual level. Additionally, it provides preliminary support to develop the multimodal MRI method as a clinical diagnostic aid.

An MRI-based classification scheme to predict passive access of 5 to 50-nm large nanoparticles to tumors.

Nanoparticles are useful tools in oncology because of their capacity to passively accumulate in tumors in particular via the enhanced permeability and retention (EPR) effect. However, the importance and reliability of this effect remains controversial and quite often unpredictable. In this preclinical study, we used optical imaging to detect the accumulation of three types of fluorescent nanoparticles in eight different subcutaneous and orthotopic tumor models, and dynamic contrast-enhanced and vessel size index Magnetic Resonance Imaging (MRI) to measure the functional parameters of these tumors. The results demonstrate that the permeability and blood volume fraction determined by MRI are useful parameters for predicting the capacity of a tumor to accumulate nanoparticles. Translated to a clinical situation, this strategy could help anticipate the EPR effect of a particular tumor and thus its accessibility to nanomedicines.

Individualized differential diagnosis of schizophrenia and mood disorders using neuroanatomical biomarkers.

Magnetic resonance imaging-based markers of schizophrenia have been repeatedly shown to separate patients from healthy controls at the single-subject level, but it remains unclear whether these markers reliably distinguish schizophrenia from mood disorders across the life span and generalize to new patients as well as to early stages of these illnesses. The current study used structural MRI-based multivariate pattern classification to (i) identify and cross-validate a differential diagnostic signature separating patients with first-episode and recurrent stages of schizophrenia (n = 158) from patients with major depression (n = 104); and (ii) quantify the impact of major clinical variables, including disease stage, age of disease onset and accelerated brain ageing on the signature's classification performance. This diagnostic magnetic resonance imaging signature was then evaluated in an independent patient cohort from two different centres to test its generalizability to individuals with bipolar disorder (n = 35), first-episode psychosis (n = 23) and clinically defined at-risk mental states for psychosis (n = 89). Neuroanatomical diagnosis was correct in 80% and 72% of patients with major depression and schizophrenia, respectively, and involved a pattern of prefronto-temporo-limbic volume reductions and premotor, somatosensory and subcortical increments in schizophrenia versus major depression. Diagnostic performance was not influenced by the presence of depressive symptoms in schizophrenia or psychotic symptoms in major depression, but earlier disease onset and accelerated brain ageing promoted misclassification in major depression due to an increased neuroanatomical schizophrenia likeness of these patients. Furthermore, disease stage significantly moderated neuroanatomical diagnosis as recurrently-ill patients had higher misclassification rates (major depression: 23%; schizophrenia: 29%) than first-episode patients (major depression: 15%; schizophrenia: 12%). Finally, the trained biomarker assigned 74% of the bipolar patients to the major depression group, while 83% of the first-episode psychosis patients and 77% and 61% of the individuals with an ultra-high risk and low-risk state, respectively, were labelled with schizophrenia. Our findings suggest that neuroanatomical information may provide generalizable diagnostic tools distinguishing schizophrenia from mood disorders early in the course of psychosis. Disease course-related variables such as age of disease onset and disease stage as well alterations of structural brain maturation may strongly impact on the neuroanatomical separability of major depression and schizophrenia.

Spinal cord injury without radiologic abnormality in children: a systematic review and meta-analysis.

Spinal cord injury in children is associated with severe morbidity and immense socioeconomic burden. In spinal cord injury without radiologic abnormalities (SCIWORA), magnetic resonance imaging (MRI) can detect intramedullary or extramedullary pathologies or show absence of neuroimaging abnormalities. However, the prognostic and therapeutic consequences of specific MRI patterns are unclear. A comprehensive systematic literature search was performed to examine patient characteristics and imaging patterns of pediatric SCIWORA and to evaluate the prognostic value of a MRI-based classification system. MEDLINE, Cochrane Central Register of Controlled Trials, and Google Scholar were searched for studies on SCIWORA in children. Inclusion criteria were (1) traumatic spinal cord injury with acute neurologic deficit, (2) absence of fractures and/or dislocations of the spine, and (3) an immature skeleton or age of less than 18 years. MRI patterns and clinical course were correlated. Forty articles reporting 114 patients were identified. At admission, neurologic deficit assessed by the American Spinal Injury Association impairment scale was A in 28%, B in 17%, C in 31%, and D in 25%. At final follow-up, these were 19%, 6%, 10%, and 16%, respectively. In 43%, no MRI abnormalities (Type I) were detected, and 57% exhibited abnormal scan results (Type II): 6% revealed extraneural (Type IIa), 38% intraneural (Type IIb), and 13% combined abnormalities (Type IIc). At admission and follow-up, American Spinal Injury Association impairment scale differed significantly between the imaging types. This systematic review emphasizes the prognostic value of spinal MRI for children with SCIWORA. It highlights the role of the MRI classification system in improving the comparability and interpretability. Systematic review, level IV.

Invasion of the cavernous sinus space in pituitary adenomas: endoscopic verification and its correlation with an MRI-based classification.

OBJECT An important prognostic factor for the surgical outcome and recurrence of a pituitary adenoma is its invasiveness into parasellar tissue, particularly into the space of the cavernous sinus (CS). The aims of this study were to reevaluate the existing parasellar classifications using an endoscopic technique and to evaluate the clinical and radiological outcomes associated with each grade. METHODS The authors investigated 137 pituitary macroadenomas classified radiologically at least on one side as Grade 1 or higher (parasellar extension) and correlated the surgical findings using an endoscopic technique, with special reference to the invasiveness of the tumor into the CS. In each case, postoperative MRI was performed to evaluate the gross-total resection (GTR) rate and the rate of endocrinological remission (ER) in functioning adenomas. RESULTS The authors found a 16% rate of CS invasion during surgery for these macroadenomas. Adenomas radiologically classified as Grade 1 were found to be invasive in 1.5%, and the GTR/ER rate was 83%/88%. For Grade 2 adenomas, the rate of invasion was 9.9%, and the GTR/ER rate was 71%/60%. For Grade 3 adenomas, the rate of invasion was 37.9%, and the GTR/ER rate was 75%/33%. When the superior compartment of the CS (Grade 3A) was involved, the authors found a rate of invasion that was lower (p < 0.001) than that when the inferior compartment was involved (Grade 3B). The rate of invasion in Grade 3A adenomas was 26.5% with a GTR/ER rate of 85%/67%, whereas for Grade 3B adenomas, the rate of surgically observed invasion was 70.6% with a GTR/ER rate of 64%/0%. All of the Grade 4 adenomas were invasive, and the GTR/ER rate was 0%. A comparison of microscopic and endoscopic techniques revealed no difference in adenomas with Grade 1 or 4 parasellar extension. In Grade 2 adenomas, however, the CS was found by the endoscopic technique to be invaded in 9.9% and by microscopic evaluation to be invaded in 88% (p < 0.001); in Grade 3 adenomas, the difference was 37.9% versus 86%, respectively (p = 0.002). Grade 4 adenomas had a statistically significant lower rate of GTR than those of all the other grades. In case of ER only, Grade 1 adenomas had a statistically significant higher rate of remission than did Grade 3B and Grade 4 adenomas. CONCLUSIONS The proposed classification proved that with increasing grades, the likelihood of surgically observed invasion rises and the chance of GTR and ER decreases. The direct endoscopic view confirmed the low rate of invasion of Grade 1 adenomas but showed significantly lower rates of invasion in Grade 2 and 3 adenomas than those previously found using the microscopic technique. In cases in which the intracavernous internal carotid artery was encased (Grade 4), all the adenomas were invasive and the GTR/ER rate was 0%/0%. The authors suggest the addition of Grades 3A and 3B to distinguish the strikingly different outcomes of adenomas invading the superior CS compartments and those invading the inferior CS compartments.

MRI-Based Classification of the Morphology and Pathology of the Annulus Fibrosus of the Intervertebral Disc

Introduction The high rate of recurrent disc herniation after discectomy is a prevalent problem. An effective biological annular repair could improve the surgical outcome in patients with contained disc herniation but otherwise minor degenerative changes. To estimate the integrity of the annulus fibrosus (AF) and judge different repair options a better understanding of disc morphology and pathology is crucial.1 Axial- and sagittal T2*-mapping are effective to detect early stages of degenerative disc disease and are especially suited to evaluate the interface between the proton-rich nucleus pulposus and the fibers of the annulus fibrosus.2,3 So far there is no classification of AF morphology or pathology as current classifications only describe localization and extend of the hernia itself. Materials and Methods Prospective consecutive case study of 100 patients requiring MRI diagnostic for suspected lumbar disc herniation. Beside the clinical standard MRI protocol with sagittal T1- and T2-weighted images, axial and sagittal T2*-mapping sequences of the affected and adjacent discs were performed using 3T MRI. Three observers used established classification schemes to assess the grade of degeneration (Pfirrmann, Benneker, Watanabe) and herniation (Carragee, Rajaserakan) and also developed a new classification protocol specific for the annulus fibrosus. Results In the preliminary data four types of annulus morphology/pathology could be found: type 1: healthy, uniform thickness of AF, clear demarcation between nucleus pulposus (NP) and AF. Type 2: concentric thinning with bulge, concentric tears, demarcation between NP and AF unclear. Type 3: nonconcentric thinning, ±radial fissures, ±protrusion, and type 4: herniations with further subgroups. Conclusion Based on preliminary axial and sagittal T2* images a new classification system for AF morphology and pathology could be established. Inter- and intraobserver liability is high. Disclosure of Interest None declared References Guterl CC, See EY, Blanquer SB, et al. Challenges and strategies in the repair of ruptured annulus fibrosus. Eur Cell Mater 2013;25:1–21 Stelzeneder D, Welsch GH, Kovács BK, et al. Quantitative T2 evaluation at 3.0T compared to morphological grading of the lumbar intervertebral disc: a standardized evaluation approach in patients with low back pain. Eur J Radiol 2012;81(2):324–330 Hoppe S, Quirbach S, Mamisch TC, Krause FG, Werlen S, Benneker LM. Axial T2 mapping in intervertebral discs: a new technique for assessment of intervertebral disc degeneration. Eur Radiol 2012;22(9):2013–2019