Search

Brain source localization has been consistently implemented over the recent years to elucidate complex brain operations, pairing the high temporal resolution of the EEG with the high spatial estimation of the estimated sources. This review paper aims to present the basic principles of Electrical source imaging (ESI) in the context of the recent progress for solving the forward and the inverse problems, and highlight the advantages and limitations of the different approaches. As such, a synthesis of the current state-of-the-art methodological aspects is provided, offering a complete overview of the present advances with regard to the ESI solutions. Moreover, the new dimensions for the analysis of the brain processes are indicated in terms of clinical and cognitive ESI applications, while the prevailing challenges and limitations are thoroughly discussed, providing insights for future approaches that could help to alleviate methodological and technical shortcomings.

Background: During RT cycles, the tumor response pattern could affect tumor coverage and may lead to organs at risk of overdose. As such, early prediction of significant volumetric changes could therefore reduce potential radiation-related adverse effects. Nevertheless, effective machine learning approaches based on the radiomic features of the clinically used CBCT images to determine the tumor volume variations due to RT not having been implemented so far. Methods: CBCT images from 40 HN cancer patients were collected weekly during RT treatment. From the obtained images, the Clinical Target Volume (CTV) and Parotid Glands (PG) regions of interest were utilized to calculate 104 delta-radiomics features. These features were fed on a feature selection and classification procedure for the early prediction of significant volumetric alterations. Results: The proposed framework was able to achieve 0.90 classification performance accuracy while detecting a small subset of discriminative characteristics from the 1st week of RT. The selected features were further analyzed regarding their effects on temporal changes in anatomy and tumor response modeling. Conclusion: The use of machine learning algorithms offers promising perspectives for fast and reliable early prediction of large volumetric deviations as a result of RT treatment, exploiting hidden patterns in the overall anatomical characteristics.

Global population ageing is a pressing issue with significant humanitarian implications associated with compromised independence, social isolation, frailty, and diminished quality of life affecting millions of older individuals and their families worldwide as well as healthcare and social infrastructures. In this context, this chapter presents a personalized smart-living framework for promoting healthy and successful ageing, combining interdisciplinary knowledge and technologies including internet of things (IoT), artificial intelligence (AI), and affective computing (AC). The proposed intervention tackles the health-related, societal, and emotional impacts of ageing within a society with a rising mean global age aiming to mitigate the effects of decline and reduce the associated risks. The outlined architecture is based on sensor-derived information blended with non-sensor-based computerized modules and voice-based inputs for establishing a smart ecosystem that enables older people to live independently inside and outside their home, maintaining a high cognitive and physical status.

Σε αυτή τη διδακτορική διατριβή, έχει εφαρμοστεί ένα ευρύ φάσμα προσεγγίσεων Μηχανικής Μάθησης (ΜΜ) σε πειράματα που σχετίζονται με τον εγκέφαλο, παρέχοντας τη βάση για αλληλεπιδράσεις Διεπαφών Ανθρώπου-Μηχανής (ΔΑΜ) και Διεπαφών Εγκεφάλου-Υπολογιστή (ΔΕΥ) ρίχνοντας φως σε υποκείμενες γνωστικές πτυχές διαφορετικών καταστάσεων όσον αφορά τη νευροεπιστήμη. Υπό αυτό το πρίσμα, πρέπει να σημειωθεί ότι η εφαρμογή ΔΕΥ απαιτεί τη συνεχή μέτρηση των εγκεφαλικών σημάτων τα οποία να μπορούν να μεταφραστούν και να εισάγουν πληροφορίες στη συνδεδεμένη συσκευή για την εκτέλεση της κάθε εργασίας. Κατ' επέκταση, μη σχετιζόμενα νευρικά ερεθίσματα ενδέχεται να συγχέουν την εσωτερική διαδικασία ταξινόμησης, επομένως είναι εξαιρετικά σημαντικό να ανιχνεύονται συγκεκριμένες νευρικές ιδιότητες με τρόπο που να είναι όσο το δυνατόν πιο διακριτές και με καθολικό τρόπο. Ο στόχος της διατριβής αυτής είναι να θέσει τα θεμέλια για την εφαρμογή ΔΕΥ αξιολογώντας τα καθολικά χαρακτηριστικά διαφορετικών γνωστικών καταστάσεων και συνθηκών στα πλαίσια ΜΜ, ενώ παράλληλα να λαμβάνει υπόψη τις εξελίξεις στις αλγοριθμικές εφαρμογές και των εφαρμογών των υπαρχόντων συστημάτων στην ανάλυση ηλεκτροεγκεφαλογραφικών (ΗΕΓ) καταγραφών.Πιο συγκεκριμένα, προτείνονται προηγμένες προσεγγίσεις Μηχανικής Μάθησης για την επεξεργασία και την ανάλυση των ηλεκτροεγκεφαλογραφικών σημάτων, αξιοποιώντας ιδιότητες σήματος στο πεδίο του χρόνου και / ή συχνότητας. Προς αυτήν την κατεύθυνση, χρησιμοποιήθηκαν τρία πειράματα ηλεκτροεγκεφαλογραφικών καταγραφών υψηλής πυκνότητας προκειμένου να μελετηθούν γνωστικές λειτουργίες υψηλότερης τάξης, αποτελώντας τη βάση για εφαρμογές πραγματικού κόσμου ειδικά στον τομέα των Διεπαφών Εγκεφάλου-Υπολογιστή. Σε αυτήν τη διδακτορική διατριβή, προτείνονται προηγμένες προσεγγίσεις Μηχανικής Μάθησης για την επεξεργασία και την ανάλυση των ηλεκτροεγκεφαλογραφικών σημάτων, αξιοποιώντας ιδιότητες σήματος στο πεδίο του χρόνου και / ή συχνότητας. Προς αυτήν την κατεύθυνση, χρησιμοποιήθηκαν τρία πειράματα ηλεκτροεγκεφαλογραφικών καταγραφών υψηλής πυκνότητας προκειμένου να μελετηθούν γνωστικές λειτουργίες υψηλότερης τάξης, αποτελώντας τη βάση για εφαρμογές πραγματικού κόσμου ειδικά στον τομέα των Διεπαφών Εγκεφάλου-Υπολογιστή. Τα καταγεγραμμένα δεδομένα διερευνήθηκαν χρησιμοποιώντας δυναμικά που σχετίζονται με συμβάντα (Βιωματικά Δυναμικά, ΒΔ), φασματική αποσύνθεση και δίκτυα εγκεφάλου, ενώ αναπτύχθηκαν νέες μεθοδολογίες για τη μοντελοποίηση και ανάλυση των νευρολογικών δομών παρέχοντας έγκυρα αποτελέσματα. Επιπλέον, τα χαρακτηριστικά που εξήχθησαν από κάθε πείραμα συνδυάστηκαν με διαφορετικές τεχνικές ταξινόμησης και επιλογής χαρακτηριστικών (απομονώνοντας ένα μικρό υποσύνολο σημαντικών αλληλεπιδράσεων) και πετυχαίνοντας υψηλή ακρίβεια στη διάκριση μεταξύ των διαφόρων καταστάσεων. Αξίζει να σημειωθεί ότι η αποκρυπτογράφηση της ηλεκτρικής δραστηριότητας του εγκεφάλου με γενικευμένο τρόπο (άσχετο με τη μεταβλητότητα των εργασιών και των θεμάτων) είναι μια σημαντική απαίτηση για την προσαρμογή ενός κοινού πλαισίου σε πραγματικές εφαρμογές Διεπαφών Εγκεφάλου Υπολογιστή. Σε αυτό επικεντρώνεται αυτή η διδακτορική διατριβή, μέσω της επέκτασης της τρέχουσας τεχνολογίας σε νευρογνωστικές προσαρμογές της Μηχανικής Μάθησης. Πιο συγκεκριμένα, το πείραμα 1 (Κεφάλαιο 3) διερευνά τον αντίκτυπο διαφορετικών χρονικών παραθύρων σε ΒΔ μετά το σφάλμα σε ακουστική δοκιμασία με διαφορετικές συνθήκες πολυπλοκότητας. Ως εκ τούτου, χαρακτηριστικά σήματος (όπως πλάτος, λανθάνουσα κατάσταση κ.λπ.) που είναι αντιπροσωπευτικά των συστατικών ΒΔ που προκαλούνται από σφάλματα, χρησιμοποιήθηκαν σε ένα πλαίσιο επιλογής χαρακτηριστικών και ταξινόμησης που εξετάζει την επίδραση των διαμορφώσεων σήματος με την πάροδο του χρόνου. Το πείραμα 2 (Κεφάλαιο 4) αφορά την ανεξάρτητη (από το είδος διαδικασίας) εκτίμηση του υψηλού έναντι του χαμηλού νοητικού φόρτου εργασίας σε δύο διαφορετικά παραδείγματα μνήμης εργασίας. Στο πλαίσιο αυτό, ένα δικτύου εγκεφάλου συνδυάστηκε με φασματικά χαρακτηριστικά εγκαφαλογραφήματος για την εξαγωγή των γενικευμένων χαρακτηριστικών που είναι αντιπροσωπευτικά του νοητικού φόρτου, ενώ διαφορετικοί μέθοδοι ταξινόμησης συνδυάστηκαν με έναν αλγόριθμο επιλογής χαρακτηριστικών για την αξιολόγηση των ιδιοτήτων διακριτότητας των υποσύνολων λειτουργιών. Η περαιτέρω εξέταση των επιλεγμένων χαρακτηριστικών αποκάλυψε κοινά πρότυπα ανεξάρτητα από την εργασία σχετικά με τις φασματικές και τοπολογικές τους ιδιότητες. Τέλος, το πείραμα 3 (Κεφάλαιο 5) διερευνά τους νευρικούς μηχανισμούς του νοητικού φόρτου εργασίας και τη διαμόρφωσή τους σε συνθήκες που προσομοιώνουν σενάρια πραγματικού κόσμου. Σε αυτήν την περίπτωση, πραγματοποιήθηκε ένα καλά ελεγχόμενο πείραμα προσομοίωσης πτήσης σε περιβάλλοντα οθόνης υπολογιστή (2D) και εικονικής πραγματικότητας (3D) με πολλαπλά επίπεδα δυσκολίας. Τα προκύπτοντα δεδομένα στη συνέχεια υποβλήθηκαν σε επεξεργασία χρησιμοποιώντας εύρεση νευρολογικών πηγών και συσσωματώθηκαν σε φλοιώδεις περιοχές κατασκευάζοντας φασματικά διαστημικά δικτύων πηγής. Στη συνέχεια εφαρμόστηκαν αλγόριθμοι μηχανικής μάθησης για να συγκριθούν οι μεταβολές της συνδεσιμότητας και να εντοπιστούν τα πρότυπα αναδιοργάνωσης των εγκεφαλικών δικτύων σε σχέση με κάθε επίπεδο και κατάσταση φόρτου εργασίας.Τα αποτελέσματα παρέχουν νέες γνώσεις στους υποκείμενους γνωστικούς μηχανισμούς, ενώ παρέχουν υποστήριξη σε ειδικούς στους σχετικούς ιατρικούς τομείς, επιτρέποντας στη συνέχεια την αποτελεσματική παρακολούθηση της εγκεφαλικής δραστηριότητας. Επιπλέον, ο συνδυασμός της ανάλυσης που παρέχεται μέσω της Μηχανικής Μάθησης με τη νευροεπιστήμη ανοίγει το δρόμο για εφαρμογές διεπαφής ανθρώπου/εγκεφάλου-μηχανής τόσο σε επιστημονικούς όσο και σε ιατρικούς τομείς. Τα πειραματικά έργα και η ανάλυση στα προαναφερθέντα πειράματα αποτελούν τη θεωρητική βάση για τέτοιες διεπαφές, των οποίων οι επιπτώσεις και η ταξινόμηση συζητούνται επίσης σε αυτό το διδακτορικό (Κεφάλαιο 6).

Epilepsy is one of the most common neurological disorders, with millions affected worldwide, disturbing the normal brain activity and causing abnormal dynamics to be initiated in various regions of the brain. In order, to define the different ictal states and thus to evaluate the overall course of the patient, expert clinicians rely on Electroencephalography (EEG) denoting the differentiated events based on their experience and perception. As such, the implementation of cutting-edge Artificial Intelligence (AI) tools can not only alleviate misdetection but also provide a support bases in order to extract additional information regarding epileptic characteristics. In this chapter, AI tools and techniques that comprise distinct frameworks of epilepsy evaluation over the last decade are described while noting the performance, the methodological aspects and challenges. The related AI applications are further reviewed concerning the principles, parameters, complexity, feature selection and classification approaches and their implications in automated system integration and epileptic attributes identification.

Despite the apparent usefulness of efficient mental workload assessment in various real-world situations, the underlying neural mechanism remains largely unknown, and studies of the mental workload are limited to well-controlled cognitive tasks using a 2D computer screen. In this paper, we investigated functional brain network alterations in a simulated flight experiment with three mental workload levels and compared the reorganization pattern between computer screen (2D) and virtual reality (3D) interfaces. We constructed multiband functional networks in electroencephalogram (EEG) source space, which were further assessed in terms of network efficiency and workload classification performances. We found that increased alpha band efficiencies and beta band local efficiency were associated with elevated mental workload levels, while beta band global efficiency exhibited distinct development trends between 2D and 3D interfaces. Furthermore, using a small subset of connectivity features, we achieved a satisfactory multi-level workload classification accuracy in both interfaces (82% for both 2D and 3D). Further inspection of these discriminative connectivity subsets, we found predominant alpha band connectivity features followed by beta and theta band features with different topological patterns between 2D and 3D interfaces. These findings allow for a more comprehensive interpretation of the neural mechanisms of mental workload in relation to real-world assessment.

In the nascent field of neuroergonomics, mental workload assessment is one of the most important issues and has an apparent significance in real-world applications. Although prior research has achieved efficient single-task classification, scatted studies on cross-task mental workload assessment usually result in unsatisfactory performance. Here, we introduce a data-driven analysis framework to overcome the challenges regarding task-independent workload assessment using a fusion of EEG spectral characteristics and unveil the common neural mechanisms underlying mental workload. Specifically, multi-frequency power spectrum and functional connectivity (FC) were estimated for two workload levels in two working-memory tasks performed by 40 healthy participants, subsequently being fed into a machine learning approach to obtain the importance of each feature vector and evaluate classification performance in a cross-task fashion. Our framework achieved a classification accuracy of 0.94 for task-independent mental workload discrimination. Further investigation of the designated features in terms of their spectral and localization properties revealed task-independent common patterns in the neural mechanisms governing workload. In particular, increased workload was associated with elevated frontal delta and theta power but reduced parietal alpha power, whereas FC exhibited complex frequency- and region-dependent alterations. By implication, the employment of the EEG feature fusion emphasized their utility in serving as promising indicators for different workload conditions applications.

Diagnosis of Autism Spectrum Disorder (ASD) is a complex task that typically relies on the expertise of the clinician due to the lack of specific quantitative biomarkers. As a consequence, automatic categorization of an individual within the ASD taxonomy poses many challenges, usually with controversial results. The implementation of Machine Learning approaches as a diagnostic tool for ASD classification is rapidly growing in the field of neuroscience, holding the potential to enhance discrimination validity among ASD and Typically Developed (TD) individuals, while providing indications in regard to ASD differentiating factors. In this study, various feature selection and classification techniques were employed in order to successfully discern between ASD and TD, using data from large resting-state functional Magnetic Resonance Imaging (rs-fMRI) database. Moreover, we adopt novel features, namely the Haralick texture features and the Kullback-Leibler divergence, combined with already established ones (i.e. static Functional Connectivity and demographics), assessing the most informative global attributes. Our framework succeeded in the identification of a small number of discriminative features, leading to high performance relative to previous works with optimal classification accuracy of 0.725.

Despite the apparent importance of mental fatigue detection, a reliable application is hindered due to the incomprehensive understanding of the neural mechanisms of mental fatigue. In this paper, we investigated the topological alterations of functional brain networks in the theta band (4 - 7 Hz) of electroencephalography (EEG) data from 40 male subjects undergoing two distinct fatigue-inducing tasks: a low-intensity one-hour simulated driving and a high-demanding half-hour sustained attention task [psychomotor vigilance task (PVT)]. Behaviorally, subjects demonstrated a robust mental fatigue effect, as reflected by significantly declined performances in cognitive tasks prior and post these two tasks. Furthermore, characteristic path length presented a positive correlation with task duration, which led to a significant increase between the first and the last five minutes of both tasks, indicating a fatigue-related disruption in information processing efficiency. However, significantly increased clustering coefficient was revealed only in the driving task, suggesting distinct network reorganizations between the two fatigue-inducing tasks. Moreover, high accuracy (92% for driving; 97% for PVT) was achieved for fatigue classification with apparently different discriminative functional connectivity features. These findings augment our understanding of the complex nature of fatigue-related neural mechanisms and demonstrate the feasibility of using functional connectivity as neural biomarkers for applicable fatigue monitoring.

Efficient classification of mental workload, an important issue in neuroscience, is limited, so far to single task, while cross-task classification remains a challenge. Furthermore, network approaches have emerged as a promising direction for studying the complex organization of the brain, enabling easier interpretation of various mental states. In this paper, using two mental tasks (N-back and mental arithmetic), we present a framework for cross- as well as within-task workload discrimination by utilizing multiband electroencephalography (EEG) cortical brain connectivity. In detail, we constructed functional networks in EEG source space in different frequency bands and considering the individual functional connections as classification features, we identified salient feature subsets based on a sequential feature selection algorithm. These connectivity subsets were able to provide accuracy of 87% for cross-task, 88% for N-back task, and 86% for mental arithmetic task. In conclusion, our method achieved to detect a small number of discriminative interactions among brain areas, leading to high accuracy in both within-task and cross-task classifications. In addition, the identified functional connectivity features, the majority of which were detected in frontal areas in theta and beta frequency bands, helped delineate the shared as well as the distinct neural mechanisms of the two mental tasks.

Background Context Although general hypothermia is recognized as a clinically applicable neuroprotective intervention, acute moderate local hypothermia post contusive spinal cord injury (SCI) is being considered a more effective approach. Previously, we have investigated the feasibility and safety of inducing prolonged local hypothermia in the central nervous system of a rodent model. Purpose Here, we aimed to verify the efficacy and neuroprotective effects of 5 and 8 hours of local moderate hypothermia (30±0.5°C) induced 2 hours after moderate thoracic contusive SCI in rats. Study Design Rats were induced with moderate SCI (12.5 mm) at its T8 section. Local hypothermia (30±0.5°C) was induced 2 hours after injury induction with an M-shaped copper tube with flow of cold water (12°C), from the T6 to the T10 region. Experiment groups were divided into 5-hour and 8-hour hypothermia treatment groups, respectively, whereas the normothermia control group underwent no hypothermia treatment. Methods The neuroprotective effects were assessed through objective weekly somatosensory evoked potential (SSEP) and motor behavior (basso, beattie and bresnahan Basso, Beattie and Bresnahan (BBB) scoring) monitoring. Histology on spinal cord was performed until at the end of day 56. All authors declared no conflict of interest. This work was supported by the Singapore Institute for Neurotechnology Seed Fund (R-175-000-121-733), National University of Singapore, Ministry of Education, Tier 1 (R-172-000-414-112.). Results Our results show significant SSEP amplitudes recovery in local hypothermia groups starting from day 14 post-injury onward for the 8-hour treatment group, which persisted up to days 28 and 42, whereas the 5-hour group showed significant improvement only at day 42. The functional improvement plateaued after day 42 as compared with control group of SCI with normothermia. This was supported by both 5-hour and 8-hour improvement in locomotion as measured by BBB scores. Local hypothermia also observed insignificant changes in its SSEP latency, as compared with the control. In addition, 5- and 8-hour hypothermia rats' spinal cord showed higher percentage of parenchyma preservation. Conclusions Early local moderate hypothermia can be induced for extended periods of time post SCI in the rodent model. Such intervention improves functional electrophysiological outcome and motor behavior recovery for a long time, lasting until 8 weeks.

Restoring normal walking abilities following the loss of them is a challenge. Importantly, there is a growing need for a better understanding of brain plasticity and the neural involvements for the initiation and control of these abilities so as to develop better rehabilitation programmes and external support devices. In this paper, we attempt to identify gait-related neural activities by decoding neural signals obtained from electroencephalography (EEG) measurements while subjects performed three types of walking: without exoskeleton (free walking), and with exoskeleton support (zero force and assisting force). An average classification accuracy of 92.0% for training and 73.8% for testing sets was achieved using features extracted from mu and beta frequency bands. Furthermore, we found that mu band features contributed significantly to the classification accuracy and were localized mainly in sensorimotor regions that are associated with the control of the exoskeleton. These findings contribute meaningful insight on the neural dynamics associated with lower limb movements and provide useful information for future developments of orthotic devices and rehabilitation programs.

Concurrent continuous EEG-fMRI is a novel multimodal technique that is finding its way into clinical practice in epilepsy. EEG timeseries are used to identify the timing of interictal epileptiform discharges (IEDs) which is then included in a GLM analysis in fMRI to localize the epileptic onset zone. Nevertheless, there are still some concerns about its reliability concerning BOLD changes correlated with IEDs. Even though IEDs are identified by an experienced neurologist-epiliptologist, the reliability and concordance of the mark-ups is depending on many factors including the level of fatigue, the amount of time that he spent or, in some cases, even the screen that is being used for the display of timeseries. This investigation is aiming to unravel the effect of misidentification or inaccuracy in the mark-ups of IEDs in the fMRI statistical parametric maps. Concurrent EEG-fMRI was conducted in six subjects with various types of epilepsy. IEDs were identified by an experienced neurologist-epiliptologist. Analysis of EEG was performed with EEGLAB and analysis of fMRI was conducted in FSL. Preliminary results revealed lower statistical significance for missing events or larger period of IEDs than the actual ones and the introduction of false positives and false negatives in statistical parametric maps when random events were included in the GLM on top of the IEDs. Our results suggest that mark-ups in EEG for simultaneous EEG-fMRI should be done with caution from an experienced and restful neurologist as it affects the fMRI results in various and unpredicted ways.

Previous neurocognitive assessments in Small Cell Lung Cancer (SCLC) population, highlight the presence of neurocognitive impairments (mainly in attention processing and executive functioning) in this type of cancer. The majority of these studies, associate these deficits with the Prophylactic Cranial Irradiation (PCI) that patients undergo in order to avoid brain metastasis. However, there is not much evidence exploring cognitive impairments induced by chemotherapy in SCLC patients. For this reason, we aimed to investigate the underlying processes that may potentially affect cognition by examining brain functional connectivity in nineteen SCLC patients after chemotherapy treatment, while additionally including fourteen healthy participants as control group. Independent Component Analysis (ICA) is a functional connectivity measure aiming to unravel the temporal correlation between brain regions, which are called brain networks. We focused on two brain networks related to the aforementioned cognitive functions, the Default Mode Network (DMN) and the Task-Positive Network (TPN). Permutation tests were performed between the two groups to assess the differences and control for familywise errors in the statistical parametric maps. ICA analysis showed functional connectivity disruptions within both of the investigated networks. These results, propose a detrimental effect of chemotherapy on brain functioning in the SCLC population.

The detection of an error is the cognitive evaluation of an action outcome that is considered undesired or mismatches an expected response. Brain activity during monitoring of correct and incorrect responses elicits Event Related Potentials (ERPs) revealing complex cerebral responses to deviant sensory stimuli. Development of accurate error detection systems is of great importance both concerning practical applications and in investigating the complex neural mechanisms of decision making. In this study, data are used from an audio identification experiment that was implemented with two levels of complexity in order to investigate neurophysiological error processing mechanisms in actors and observers. To examine and analyse the variations of the processing of erroneous sensory information for each level of complexity we employ Support Vector Machines (SVM) classifiers with various learning methods and kernels using characteristic ERP time-windowed features. For dimensionality reduction and to remove redundant features we implement a feature selection framework based on Sequential Forward Selection (SFS). The proposed method provided high accuracy in identifying correct and incorrect responses both for actors and for observers with mean accuracy of 93% and 91% respectively. Additionally, computational time was reduced and the effects of the nesting problem usually occurring in SFS of large feature sets were alleviated.