Schizophrenia is a severe neuropsychiatric disease whose diagnosis, unfortunately, lacks an objective diagnostic tool supporting a thorough psychiatric examination of the patient. We took advantage of today's computational abilities, structural magnetic resonance imaging, and modern machine learning methods, such as stacked autoencoders (SAE) and 3D convolutional neural networks (3D CNN), to teach them to classify 52 patients with schizophrenia and 52 healthy controls. The main aim of this study was to explore whether complex feature extraction methods can help improve the accuracy of deep learning-based classifiers compared to minimally preprocessed data. Our experiments employed three commonly used preprocessing steps to extract three different feature types. They included voxel-based morphometry, deformation-based morphometry, and simple spatial normalization of brain tissue. In addition to classifier models, features and their combination, other model parameters such as network depth, number of neurons, number of convolutional filters, and input data size were also investigated. Autoencoders were trained on feature pools of 1000 and 5000 voxels selected by Mann-Whitney tests, and 3D CNNs were trained on whole images. The most successful model architecture (autoencoders) achieved the highest average accuracy of 69.62% (sensitivity 68.85%, specificity 70.38%). The results of all experiments were statistically compared (the Mann-Whitney test). In conclusion, SAE outperformed 3D CNN, while preprocessing using VBM helped SAE improve the results.

• Klíčová slova
• 3D CNN, autoencoders, classification, deep learning, deformation-based morphometry, schizophrenia, voxel-based morphometry,
• Publikační typ
• časopisecké články MeSH

Machine learning (ML) is a growing field that provides tools for automatic pattern recognition. The neuroimaging community currently tries to take advantage of ML in order to develop an auxiliary diagnostic tool for schizophrenia diagnostics. In this letter, we present a classification framework based on features extracted from magnetic resonance imaging (MRI) data using two automatic whole-brain morphometry methods: voxel-based (VBM) and deformation-based morphometry (DBM). The framework employs a random subspace ensemble-based artificial neural network classifier-in particular, a multilayer perceptron (MLP). The framework was tested on data from first-episode schizophrenia patients and healthy controls. The experiments differed in terms of feature extraction methods, using VBM, DBM, and a combination of both morphometry methods. Thus, features of different types were available for model adaptation. As we expected, the combination of features increased the MLP classification accuracy up to 73.12%-an improvement of 5% versus MLP-based only on VBM or DBM features. To further verify the findings, other comparisons using support vector machines in place of MLPs were made within the framework. However, it cannot be concluded that any classifier was better than another.

• MeSH
• diagnóza počítačová metody   MeSH
• lidé MeSH
• magnetická rezonanční tomografie * metody   MeSH
• mozek diagnostické zobrazování   MeSH
• neuronové sítě * MeSH
• rozpoznávání automatizované metody   MeSH
• schizofrenie diagnostické zobrazování   MeSH
• Check Tag
• lidé MeSH
• mužské pohlaví MeSH
• Publikační typ
• časopisecké články MeSH
• práce podpořená grantem MeSH

Cerebellar atrophy is a characteristic sign of late-onset Tay-Sachs disease (LOTS). Other structural neuroimaging abnormalities are inconsistently reported. Our study aimed to perform a detailed whole-brain analysis and quantitatively characterize morphometric changes in LOTS patients. Fourteen patients (8 M/6F) with LOTS from three centers were included in this retrospective study. For morphometric brain analyses, we used deformation-based morphometry, voxel-based morphometry, surface-based morphometry, and spatially unbiased cerebellar atlas template. The quantitative whole-brain morphometric analysis confirmed the finding of profound pontocerebellar atrophy with most affected cerebellar lobules V and VI in LOTS patients. Additionally, the atrophy of structures mainly involved in motor control, including bilateral ventral and lateral thalamic nuclei, primary motor and sensory cortex, supplementary motor area, and white matter regions containing corticospinal tract, was present. The atrophy of the right amygdala, hippocampus, and regions of occipital, parietal and temporal white matter was also observed in LOTS patients in contrast with controls (p < 0.05, FWE corrected). Patients with dysarthria and those initially presenting with ataxia had more severe cerebellar atrophy. Our results show predominant impairment of cerebellar regions responsible for speech and hand motor function in LOTS patients. Widespread morphological changes of motor cortical and subcortical regions and tracts in white matter indicate abnormalities in central motor circuits likely coresponsible for impaired speech and motor function.

• Klíčová slova
• GM2-gangliosidosis, MRI, brain atrophy, late-onset Tay-Sachs disease,
• MeSH
• atrofie patologie   MeSH
• bílá hmota * diagnostické zobrazování   MeSH
• lidé MeSH
• magnetická rezonanční tomografie MeSH
• mozek patologie   MeSH
• retrospektivní studie MeSH
• Tay-Sachsova nemoc * patologie   MeSH
• Check Tag
• lidé MeSH
• Publikační typ
• časopisecké články MeSH
• práce podpořená grantem MeSH
• Research Support, N.I.H., Extramural MeSH

The study is based on a longitudinal cephalometric follow-up of X-ray films of patients with a complete unilateral cleft lip and palate during puberty. Carthesian x, y coordinates of the total of 22 landmarks on the lateral films of 25 patients with the same diagnosis and therapy were established, which served as outcome data for shape analysis. The changes in facial shape experienced by patients between their 10th and 15th year were evaluated by the thin-plate splines (TPS) method. We used the TPS method to find a function which transforms the shape of the face at the age of 10 as well as the square TPS grid into the facial shape at the age of 15 years with the deformed TPS grid. With the help of Geometric PCA for Bookstein's coordinates we found individuals with different shapes who develop abnormally during puberty and whose development is not well predictable. During puberty the face becomes relatively elongated. The most pronounced deformation is caused by the shift of the first permanent molar anteriorly, or a relative shortening of the frontal part of the dental arch. In the anterior direction the position of landmarks changed only in the area of nose; the alveolar process shows retrusion. The position of the mandible in relation to the maxilla change mainly in the inferior direction; the mandibular angle shifts posteriorly. The shape of skeletal profile was compared with a group of patients with the same diagnosis who underwent different therapy.

• MeSH
• dítě MeSH
• kefalometrie metody   MeSH
• kraniofaciální abnormality diagnostické zobrazování   MeSH
• lebka růst a vývoj   MeSH
• lidé MeSH
• lineární modely MeSH
• mladiství MeSH
• následné studie MeSH
• nosní přepážka chirurgie   MeSH
• radiografie MeSH
• rozštěp patra chirurgie   MeSH
• rozštěp rtu chirurgie   MeSH
• studie případů a kontrol MeSH
• Check Tag
• dítě MeSH
• lidé MeSH
• mladiství MeSH
• mužské pohlaví MeSH
• Publikační typ
• časopisecké články MeSH
• práce podpořená grantem MeSH

Although recently conceptualized as a neural node essential for a vast spectrum of associative and cognitive processes, the cerebellum has largely eluded attention in the research of aging, where it is marginalized mainly to structural analyses. In the current cross-sectional study of 67 healthy subjects of various ages (20 to 76 years), we sought to provide a comprehensive, multimodal account of age-related changes in the cerebellum during predictive motor timing, which was previously shown to engage this structure. We combined behavioral assessments of performance with functional MRI and voxel-based morphometry using an advanced method to avoid cerebellar deformation and registration imprecisions inherent to the standard processing at the whole-brain level. Higher age was surprisingly associated with stable behavioral performance during predictive motor timing, despite the massive decrease of infratentorial gray matter volume of a far higher extent than in the supratentorial region, affecting mainly the posterior cerebellar lobe. Nonetheless, this very area showed extensive hyperactivation directly correlated with age. The same region had decreased connectivity with the left caudate and increased connectivity with the left fusiform gyrus, the right pallidum, the hippocampus, and the lingual gyrus. Hence, we propose to extend the scaffolding theory of aging, previously limited mainly to the frontal cortices, to include also the cerebellum, which is likewise suffering from atrophy to a far greater extent than the rest of the brain and is similarly counteracting it by bilateral hyperactivation.

• Klíčová slova
• Cerebellar aging, Functional connectivity, Voxel-based morphometry, fMRI,
• MeSH
• dospělí MeSH
• lidé středního věku MeSH
• lidé MeSH
• mladý dospělý MeSH
• mozeček patologie   fyziologie   MeSH
• pohybová aktivita fyziologie   MeSH
• pozornost fyziologie   MeSH
• průřezové studie MeSH
• senioři MeSH
• stárnutí patologie   fyziologie   MeSH
• zdravé stárnutí patologie   fyziologie   MeSH
• Check Tag
• dospělí MeSH
• lidé středního věku MeSH
• lidé MeSH
• mladý dospělý MeSH
• mužské pohlaví MeSH
• senioři MeSH
• ženské pohlaví MeSH
• Publikační typ
• časopisecké články MeSH

This work explores the design and implementation of an algorithm for the classification of magnetic resonance imaging data for computer-aided diagnosis of schizophrenia. Features for classification were first extracted using two morphometric methods: voxel-based morphometry (VBM) and deformation-based morphometry (DBM). These features were then transformed into a wavelet domain using the discrete wavelet transform with various numbers of decomposition levels. The number of features was then reduced by thresholding and subsequent selection by: Fisher's Discrimination Ratio (FDR), Bhattacharyya Distance, and Variances (Var.). A Support Vector Machine with a linear kernel was used for classification. The evaluation strategy was based on leave-one-out cross-validation.

• Klíčová slova
• Machine learning, neuroimaging, schizophrenia, support vector machines,
• MeSH
• lidé MeSH
• magnetická rezonanční tomografie MeSH
• schizofrenie * MeSH
• support vector machine MeSH
• vlnková analýza MeSH
• Check Tag
• lidé MeSH
• Publikační typ
• časopisecké články MeSH

OBJECTIVE: To date, very few studies have focused on structural changes and their association with cognitive performance in isolated REM sleep behaviour disorder (iRBD). Moreover, the results of these studies are inconclusive. This study aims to evaluate differences in the associations between brain morphology and cognitive tests in iRBD and healthy controls. METHODS: Sixty-three patients with iRBD and thirty-six controls underwent MRI with a 3 T scanner. The cognitive performance was assessed by a comprehensive neuropsychological battery. Based on performance, the iRBD group was divided into two subgroups with (iRBD-MCI) and without mild cognitive impairment (iRBD-NC). The high-resolution T1-weighted images were analysed using an automated atlas segmentation tool, voxel-based (VBM) and deformation-based (DBM) morphometry to identify between-group differences and correlations with cognitive performance. RESULTS: VBM, DBM and the comparison of ROI volumes yielded no significant differences between iRBD and controls. In the iRBD group, significant correlations in VBM were found between several cortical and subcortical structures primarily located in the temporal, parietal, occipital lobe, cerebellum, and basal ganglia and three cognitive tests assessing psychomotor speed and one memory test. Between-group analysis of cognition revealed a significant difference between iRBD-MCI and iRBD-NC in tests including a processing speed component. CONCLUSIONS: iRBD shows deficits in several cognitive tests that correlate with morphological changes, the most prominent of which is in psychomotor speed and visual attention as measured by the TMT-A and associated with the volume of striatum, insula, cerebellum, temporal lobe, pallidum and amygdala.

• Klíčová slova
• Cognition, Mild cognitive impairment, Prodromal, REM sleep without atonia, Synucleinopathy, Trail Making Test,
• MeSH
• kognice MeSH
• kognitivní dysfunkce * diagnostické zobrazování   komplikace   MeSH
• lidé MeSH
• mozeček MeSH
• porucha chování v REM spánku * diagnostické zobrazování   komplikace   MeSH
• spánkový lalok MeSH
• Check Tag
• lidé MeSH
• Publikační typ
• časopisecké články MeSH

Post-traumatic stress disorder (PTSD) is a chronic disease caused by traumatic incidents. Numerous studies have revealed grey matter volume differences in affected individuals. The nature of the disease renders it difficult to distinguish between a priori versus a posteriori changes. To overcome this difficulty, we studied the consequences of a traumatic event on brain morphology in mice before and 4 weeks after exposure to brief foot shocks (or sham treatment), and correlated morphology with symptoms of hyperarousal. In the latter context, we assessed hyperarousal upon confrontation with acoustic, visual, or composite (acoustic/visual/tactile) threats and integrated the individual readouts into a single Hyperarousal Score using logistic regression analysis. MRI scans with subsequent whole-brain deformation-based morphometry (DBM) analysis revealed a volume decrease of the dorsal hippocampus and an increase of the reticular nucleus in shocked mice when compared to non-shocked controls. Using the Hyperarousal Score as regressor for the post-exposure MRI measurement, we observed negative correlations with several brain structures including the dorsal hippocampus. If the development of changes with respect to the basal MRI was considered, reduction in globus pallidus volume reflected hyperarousal severity. Our findings demonstrate that a brief traumatic incident can cause volume changes in defined brain structures and suggest the globus pallidus as an important hub for the control of fear responses to threatening stimuli of different sensory modalities.

• Klíčová slova
• Animal model, Grey matter volume, Hyperarousal, Magnetic resonance imaging, Post-traumatic stress disorder,
• MeSH
• arousal fyziologie   MeSH
• globus pallidus MeSH
• hipokampus MeSH
• magnetická rezonanční tomografie MeSH
• mozek patofyziologie   MeSH
• myši MeSH
• počítačové zpracování obrazu MeSH
• posttraumatická stresová porucha patofyziologie   MeSH
• šedá hmota fyziologie   MeSH
• zvířata MeSH
• Check Tag
• mužské pohlaví MeSH
• myši MeSH
• zvířata MeSH
• Publikační typ
• časopisecké články MeSH
• práce podpořená grantem MeSH

Background Abnormal findings at brain MRI in patients with neurologic Wilson disease (WD) are characterized by signal intensity changes and cerebral atrophy. T2 signal hypointensities and atrophy are largely irreversible with treatment; their relationship with permanent disability has not been systematically investigated. Purpose To investigate associations of regional brain atrophy and iron accumulation at MRI with clinical severity in participants with neurologic WD who are undergoing long-term anti-copper treatment. Materials and Methods Participants with WD and controls were compared in a prospective study performed from 2015 to 2019. MRI at 3.0 T included three-dimensional T1-weighted and six-echo multigradient-echo pulse sequences for morphometry and quantitative susceptibility mapping, respectively. Neurologic severity was assessed with the Unified WD Rating Scale (UWDRS). Automated multi-atlas segmentation pipeline with dual contrast (susceptibility and T1) was used for the calculation of volumes and mean susceptibilities in deep gray matter nuclei. Additionally, whole-brain analysis using deformation and surface-based morphometry was performed. Least absolute shrinkage and selection operator regression was used to assess the association of regional volumes and susceptibilities with the UWDRS score. Results Twenty-nine participants with WD (mean age, 47 years ± 9 [standard deviation]; 15 women) and 26 controls (mean age, 45 years ± 12; 14 women) were evaluated. Whole-brain analysis demonstrated atrophy of the deep gray matter nuclei, brainstem, internal capsule, motor cortex and corticospinal pathway, and visual cortex and optic radiation in participants with WD (P < .05 at voxel level, corrected for family-wise error). The UWDRS score was negatively correlated with volumes of putamen (r = -0.63, P < .001), red nucleus (r = -0.58, P = .001), globus pallidus (r = -0.53, P = .003), and substantia nigra (r = -0.50, P = .006) but not with susceptibilities. Only the putaminal volume was identified as a stable factor associated with the UWDRS score (R2 = 0.38, P < .001) using least absolute shrinkage and selection operator regression. Conclusion Individuals with Wilson disease (WD) had widespread brain atrophy most pronounced in the central structures. The putaminal volume was associated with the Unified WD Rating Scale score and can be used as a surrogate imaging marker of clinical severity. © RSNA, 2021 Supplemental material is available for this article. See also the editorial by Du and Bydder in this issue.

• MeSH
• atrofie MeSH
• hepatolentikulární degenerace diagnostické zobrazování   farmakoterapie   metabolismus   patologie   MeSH
• lidé středního věku MeSH
• lidé MeSH
• magnetická rezonanční tomografie metody   MeSH
• mozek diagnostické zobrazování   metabolismus   patologie   MeSH
• prospektivní studie MeSH
• studie případů a kontrol MeSH
• stupeň závažnosti nemoci MeSH
• železo metabolismus   MeSH
• Check Tag
• lidé středního věku MeSH
• lidé MeSH
• mužské pohlaví MeSH
• ženské pohlaví MeSH
• Publikační typ
• časopisecké články MeSH
• práce podpořená grantem MeSH
• Názvy látek
• železo MeSH

The authors analyze the results of preoperative morphometry of atrial septal defects. They compare the area of the defect and the septal area in different types of defects. Based on the thus obtained relative value, they select either suture of the defect or its correction by means of a patch. They emphasize that during correction defects deformations and reduction of the septal area must be avoided. They do not admit the possibility that the suture of the defect is exposed to traction. Ostium secundum defects the area of which is greater than 40% of the septal area are suited for correction by means of a patch. An important indicator is the transverse dimension of the defect and of the septum. This relation determines the grade of deformity during an inadequately selected suture and the presence of traction mechanisms in the area of the suture and cardiac skeleton. The authors recommend more frequent use of patches also in ostium secundum defects in the cranial lateral and distal part of the atrial defect.

• MeSH
• defekty septa síní diagnóza   chirurgie   MeSH
• dítě MeSH
• kojenec MeSH
• lidé MeSH
• mladiství MeSH
• novorozenec MeSH
• předškolní dítě MeSH
• Check Tag
• dítě MeSH
• kojenec MeSH
• lidé MeSH
• mladiství MeSH
• novorozenec MeSH
• předškolní dítě MeSH
• Publikační typ
• anglický abstrakt MeSH
• časopisecké články MeSH