High-resolution three-dimensional magnetic resonance imaging (3D-MRI) is being increasingly used to delineate morphological changes underlying neuropsychiatric disorders. Unfortunately, artifacts frequently compromise the utility of 3D-MRI yielding irreproducible results, from both type I and type II errors. It is therefore critical to screen 3D-MRIs for artifacts before use. Currently, quality assessment involves slice-wise visual inspection of 3D-MRI volumes, a procedure that is both subjective and time consuming. Automating the quality rating of 3D-MRI could improve the efficiency and reproducibility of the procedure. The present study is one of the first efforts to apply a support vector machine (SVM) algorithm in the quality assessment of structural brain images, using global and region of interest (ROI) automated image quality features developed in-house. SVM is a supervised machine-learning algorithm that can predict the category of test datasets based on the knowledge acquired from a learning dataset. The performance (accuracy) of the automated SVM approach was assessed, by comparing the SVM-predicted quality labels to investigator-determined quality labels. The accuracy for classifying 1457 3D-MRI volumes from our database using the SVM approach is around 80%. These results are promising and illustrate the possibility of using SVM as an automated quality assessment tool for 3D-MRI.

Purpose The purpose of the present study was to identify abnormal areas of regional synchronization in patients with mesial temporal lobe epilepsy and hippocampus sclerosis (mTLE-HS) compared to healthy controls, by applying a relatively novel method, the Regional Homogeneity (ReHo) method to resting state fMRI(RS-fMRI) data. Methods Eyes closed RS-fMRI data were acquired from 10 mTLE-HS patients (4 right-side, 6 left-side) and 15 age and gender matched healthy subjects, and were analyzed by using ReHo. For group analysis, 4 right-side MTLE-HS patients’ functional images were flipped, so that a homogeneous left MTLE-HS group with 10 cases were made. Key Findings Compared to the healthy control group, patients showed significantly increased ReHo in ipsilateral parahippocampal gyrus, midbrain, insula, corpus callosum, bilateral sensorimotor cortex and fronto-parietal subcortical structures, while decreased ReHo was mainly observed in default model network (DMN) (including precuneous and posterior cingulate gyrus, bilateral inferior lateral parietal and mesial prefrontal cortex) and cerebellum in patients relative to the control group. Significance This study identified that ReHo pattern in mTLE-HS patients was altered compared to healthy controls. We consider decreased ReHo in DMN to be responsible for wide functional impairments in cognitive processes. We propose that the increased ReHo in specific regions may compose a network which might be responsible for seizure genesis and propagation.

Neuroimaging science has seen a recent explosion in dataset size driving the need to develop database management with efficient processing pipelines. Multi-center neuroimaging databases consistently receive magnetic resonance imaging (MRI) data with unlabeled or incorrectly labeled contrast. There is a need to automatically identify the contrast of MRI scans to save database-managing facilities valuable resources spent by trained technicians required for visual inspection. We developed a deep learning (DL) algorithm with convolution neural network architecture to automatically infer the contrast of MRI scans based on the image intensity of multiple slices. For comparison, we developed a random forest (RF) algorithm to automatically infer the contrast of MRI scans based on acquisition parameters. The DL algorithm was able to automatically identify the MRI contrast of an unseen dataset with <0.2% error rate. The RF algorithm was able to identify the MRI contrast of the same dataset with 1.74% error rate. Our analysis showed that reduced dataset sizes caused the DL algorithm to lose generalizability. Finally, we developed a confidence measure, which made it possible to detect, with 100% specificity, all MRI volumes that were misclassified by the DL algorithm. This confidence measure can be used to alert the user on the need to inspect the small fraction of MRI volumes that are prone to misclassification. Our study introduces a practical solution for automatically identifying the MRI contrast. Furthermore, it demonstrates the powerful combination of convolution neural networks and DL for analyzing large MRI datasets.

A streamlined scientific workflow system that can track the details of the data processing history is critical for the efficient handling of fundamental routines used in scientific research. In the scientific workflow research community, the information that describes the details of data processing history is referred to as “provenance” which plays an important role in most of the existing workflow management systems. Despite its importance, however, provenance modeling and management is still a relatively new area in the scientific workflow research community. The proper scope, representation, granularity and implementation of a provenance model can vary from domain to domain and pose a number of challenges for an efficient pipeline design. This paper provides a case study on structured provenance modeling and management problems in the neuroimaging domain by introducing the Bio-Swarm-Pipeline. This new model, which is evaluated in the paper through real world scenarios, systematically addresses the provenance scope, representation, granularity, and implementation issues related to the neuroimaging domain. Although this model stems from applications in neuroimaging, the system can potentially be adapted to a wide range of bio-medical application scenarios.

Seizure localization includes neuroimaging like electroencephalogram (EEG), and magnetic resonance imaging (MRI) with limited ability to characterize the epileptogenic network. Temporal clustering analysis (TCA) characterizes epileptogenic network congruent with interictal epileptiform discharges (IED) by clustering together voxels with transient function MRI signals. We generated epileptogenic areas for 12 of 13 epilepsy patients with TCA, congruent with different areas of seizure onset. Resting fMRI scans are non-invasive, and can be acquired quickly (5 min), in patients with different levels of severity and function. Analyzing resting fMRI data using TCA is quick and can be utilized to complement clinical methods to characterize the epileptogenic network.

Micro-Electrocorticograph (μECoG) arrays offer the flexibility to record local field potentials (LFPs) from the surface of the cortex, using high density electrodes that are sub-mm in diameter. Research to date has not provided conclusive evidence for the underlying signal generation of μECoG recorded LFPs, or if μECoG arrays can capture network activity from the cortex. We studied the pervading view of the LFP signal by exploring the spatial scale at which the LFP can be considered elemental. We investigated the underlying signal generation and ability to capture functional networks by implanting, μECoG arrays to record sensory-evoked potentials in four rats. The organization of the sensory cortex was studied by analyzing the sensory-evoked potentials with two distinct modeling techniques: (1) The volume conduction model, that models the electrode LFPs with an electrostatic representation, generated by a single cortical generator, and (2) the dynamic causal model (DCM), that models the electrode LFPs with a network model, whose activity is generated by multiple interacting cortical sources. The volume conduction approach modeled activity from electrodes separated < 1000 μm, with reasonable accuracy but a network model like DCM was required to accurately capture activity > 1500 μm. The extrinsic network component in DCM was determined to be essential for accurate modeling of observed potentials. These results all point to the presence of a sensory network, and that μECoG arrays are able to capture network activity in the neocortex. The estimated DCM network models the functional organization of the cortex, as signal generators for the μECoG recorded LFPs, and provides hypothesis-testing tools to explore the brain.

Background: In this ongoing study we are testing a closed-loop neurological feedback device that can facilitate functional recovery in stroke patients with upper extremity motor deficits. Methods: This device combines Brain Computer Interface (BCI) and functional electrical muscle stimulation (FES), together with tongue stimulation (TS) in order to utilize the subject’s intention-to-move with the stimulated output. FMRI is used to examine the brain plasticity changes secondary to the rehabilitation. Subjects, wearing a 16-channel EEG cap, are first trained to voluntarily modulate beta and mu rhythms as they use motor imagery or execution of left and right hand squeezing task and then trained to use this imagery or execution to control the movement of a cursor, either to the right or the left depending on the presentation of a target rectangle shown on the screen. Once subjects achieve consistent accuracy in doing this task, FES in conjunction with TS is used. The subject is then asked to perform the task with the stimulation of FES and TS linked to their task performance. Results: Two chronic stroke patients (mean age=57, 1 male, more than 1 year postonset) were able to complete the entire 3-week BCI training course and to perform the tasks at a > 70% success rate. Cortical activation recorded by EEG in response to attempted paretic arm movement became concentrated over the contralateral motor areas. Similar changes were confirmed by fMRI measures (Figure 1). Although neither subject showed any improvement on the Action Research Arm Test (ARAT), subjects self-reported increased strength, less spasticity and a greater range of movement in their paretic arm. Conclusion: Our preliminary results indicate that training with the device may lead to brain plasticity changes toward normalization of cortical activation patterns and promote behavioral improvements for stroke patients even in their chronic stage.

Background: People with intellectual disability in Chile face individual and collective barriers to social participation. Lack of knowledge about their rights and tools for effective self-advocacy seem to be key elements that need to be improved to facilitate participation.Method: We present PaísDI, a 16 h long manualised program created by selfadvocates in collaboration with an interdisciplinary team, with four modules: rights and intellectual disability, leadership in intellectual disability, effective communication and financial considerations of social projects. This quasi-experimental study had 349 participants, divided in three groups: people with intellectual disability, relatives and professionals. Feasibility and effectiveness where measured. Results:The program is shown to be viable and effective, especially in its impact on self-perception for self-advocacy activities. Conclusion:The discussion highlights Chile's historic debt in creating policies that promote self-determination, knowledge and the empowerment of people with intellectual disability, to bolster their participation as citizens.