Autonomous navigation systems for Mars rovers face significant challenges due to the limited availability of real Martian imagery for training Artificial Intelligence (AI) models. In contrast, a large amount of synthetic images of Martian landscape can be generated and used for training. This paper presents a novel "Sim2Real" approach that integrates synthetic data generation and classification with few-shot learning to enhance real images classification for autonomous Mars rovers. Specifically, a custom Convolutional Neural Network (CNN) is first developed and pre-trained on synthetic Martian landscapes provided by Thales Alenia Space Italy (TAS-I). The pre-trained CNN is then used as a basis to develop the proposed Sim2Real approach. In particular, a few-shot learning strategy is introduced to classify real Mars rover images from an existing available dataset of the National Aeronautics and Space Administration (NASA). Preliminary experimental results demonstrated promising performance when applied to real Martian images, highlighting the potential of few-shot based Sim2Real approaches for planetary exploration.

This paper introduces a hybrid deep learning approach for processing electroencephalographic (EEG) signals to track their dynamics over time in the spatial-spectral domain. This method is particularly valuable whenever the temporal evolution of the brain process under analysis is relevant to classification. A hybrid deep learning model, herein referred to as EEGConViT, processes spatial-temporal stream of data (EEG movies). Specifically, EEGConViT consists of a custom CNN that encodes each frame into a feature vector, which is then augmented with temporal position embeddings with a Transformer model able to capture sequential dependencies. In this work, the application of motor EEG signals was investigated with the proposed EEGConViT. In particular, EEG signals preceding motor execution were processed to assess the ability of the model to predict the upcoming sub-movement of the upper limb. The model was evaluated using a collection of EEG signals from 14 subjects, derived from a publicly available repository. Using a leave-one-subject-out strategy, the model was trained on data from 13 subjects and fine-tuned on the remaining one (cross-subject training and subsequent calibration over the single subject). Results demonstrate that our approach outperforms comparable models in the literature while significantly reducing training time, an essential factor in medical applications, where both classification performance and rapid calibration are critical.

Decoding electroencephalographic (EEG) signals is of key importance in the development of brain–computer interface (BCI) systems. However, high inter-subject variability in EEG signals requires user-specific calibration, which can be time-consuming and limit the application of deep learning approaches, due to general need of large amount of data to properly train these models. In this context, this paper proposes a multidimensional and explainable deep learning framework for fast and interpretable EEG decoding. In particular, EEG signals are projected into the spatial–spectral–temporal domain and processed using a custom three-dimensional (3D) Convolutional Neural Network, here referred to as EEGCubeNet. In this work, the method has been validated on EEGs recorded during motor BCI experiments. Namely, hand open (HO) and hand close (HC) movement planning was investigated by discriminating them from the absence of movement preparation (resting state, RE). The proposed method is based on a global- to subject-specific fine-tuning. The model is globally trained on a population of subjects and then fine-tuned on the final user, significantly reducing adaptation time. Experimental results demonstrate that EEGCubeNet achieves state-of-the-art performance (accuracy of [Formula: see text] and [Formula: see text] for HC versus RE and HO versus RE, binary classification tasks, respectively) with reduced framework complexity and with a reduced training time. In addition, to enhance transparency, a 3D occlusion sensitivity analysis-based explainability method (here named 3D xAI-OSA) that generates relevance maps revealing the most significant features to each prediction, was introduced. The data and source code are available at the following link: https://github.com/AI-Lab-UniRC/EEGCubeNet

Skin cancer is one of the leading causes of mortality worldwide. Early diagnosis can ensure more effective patient treatment and outcomes, but, this is challenging due to the high similarity between different skin lesion types. There is a growing interest in developing Artificial Intelligence (AI)-based systems for automated skin lesion classification. However, current AI models are not transparent, leading to a lack of trust from clinicians who struggle to interpret and validate AI decisions. To this end, in this paper, a fine tuned EfficientNet-B0-based classifier is first developed to classify dermoscopic images of Melanoma (MEL), Nevus (NV) and Seborrheic Keratosis (SK) skin lesions gathered from the International Skin Imaging Collaboration (ISIC) dataset. Next, the explainability of the model is investigated. In particular, a new Trustworthiness Index for eXplainable AI, herein referred to as TIxAI , is proposed. The TIxAI is based on the difference between the relevance degree of the lesion and non-lesion areas, leading to the conclusion that the higher the TIxAI , the more trustworthy the classifier is expected to be. Experimental results support the use of the proposed TIxAI to assess and benchmark the reliability of classifiers also in other real-world applications.

Electroencephalography (EEG) is widely used as a non-invasive technique for the diagnosis of several brain disorders, including Alzheimer's disease and epilepsy. Until recently, diseases have been identified over EEG readings by human experts, which may not only be specific and difficult to find, but are also subject to human error. Despite the recent emergence of machine learning methods for the interpretation of EEGs, most approaches are not capable of capturing the underlying arbitrary non-Euclidean relations between signals in the different regions of the human brain. In this context, Graph Neural Networks (GNNs) have gained attention for their ability to effectively analyze complex relationships within different types of graph-structured data. This includes EEGs, a use case still relatively unexplored. In this paper, we aim to bridge this gap by presenting a study that applies GNNs for the EEG-based detection of Alzheimer's disease and discrimination of two different types of seizures. To this end, we demonstrate the value of GNNs by showing that a single GNN architecture can achieve state-of-the-art performance in both use cases. Through design space explorations and explainability analysis, we develop a graph-based transformer that achieves cross-validated accuracies over 89% and 96% in the ternary classification variants of Alzheimer's disease and epilepsy use cases, respectively, matching the intuitions drawn by expert neurologists. We also argue about the computational efficiency, generalizability and potential for real-time operation of GNNs for EEGs, positioning them as a valuable tool for classifying various neurological pathologies and opening up new prospects for research and clinical practice.

Automatic ship detection from spaceborne systems such as satellites or aircrafts, raises considerable attention in sea surface monitoring because of the several applications in military and civilian field. In this context, processing satellite images on-board would reduce the latency time especially for emergency situations. In this paper, an hardware-oriented (HO) ship detection system based on a customized Convolutional Neural Network (CNN), here referred to as HO-ShipNet, is proposed and tested on a revised version of the “Ships in Satellite Imagery” (SSI) Kaggle dataset, reporting detection accuracy of up to 95%. Furthermore, the explainability of HO-ShipNet is investigated by means of explainable Artificial Intelligence (xAI) techniques (i.e., Local Interpretable Model-Agnostic Explanation (LIME) and Occlusion Sensitivuty Analysis (OSA)), in order to understand the reasoning behind the HO-ShipNet decisions by detecting the most important input features and consequently ensure the trustworthiness of the model itself. Finally, HO-ShipNet is also implemented on the heterogeneous Xilinx xc7z045ffg900-2 SoC Field Programmable Gate Array (FPGA) outperforming state-of-the-art FPGA-based accelerators dealing with high-resolution frames. The promising results encourage the potential deployment of the proposed system for on-board applications.

In this study, a few-shot transfer learning approach was introduced to decode movement intention from electroencephalographic (EEG) signals, allowing to recognize new tasks with minimal adaptation. To this end, a dataset of EEG signals recorded during the preparation of complex sub-movements was created from a publicly available data collection. The dataset was divided into two parts: the source domain dataset (including 5 classes) and the support (target domain) dataset, (including 2 classes) with no overlap between the two datasets in terms of classes. The proposed methodology consists in projecting EEG signals into the space-frequency-time domain, in processing such projections (rearranged in channels × frequency frames) by means of a custom EEG-based deep neural network (denoted as EEGframeNET5), and then adapting the system to recognize new tasks through a few-shot transfer learning approach. The proposed method achieved an average accuracy of 72.45 ± 4.19% in the 5-way classification of samples from the source domain dataset, outperforming comparable studies in the literature. In the second phase of the study, a few-shot transfer learning approach was proposed to adapt the neural system and make it able to recognize new tasks in the support dataset. The results demonstrated the system’s ability to adapt and recognize new tasks with an average accuracy of 80 ± 0.12% in discriminating hand opening/closing preparation and outperforming reported results in the literature. This study suggests the effectiveness of EEG in capturing information related to the motor preparation of complex movements, potentially paving the way for BCI systems based on motion planning decoding. The proposed methodology could be straightforwardly extended to advanced EEG signal processing in other scenarios, such as motor imagery or neural disorder classification.

The present paper introduces a novel method, named AutoEncoder-Filter Bank Common Spatial Patterns (AE-FBCSP), to decode imagined movements from electroencephalography (EEG). AE-FBCSP is an extension of the well-established FBCSP and is based on a global (cross-subject) and subsequent transfer learning subject-specific (intra-subject) approach. A multi-way extension of AE-FBCSP is also introduced in this paper. Features are extracted from high-density EEG (64 electrodes), by means of FBCSP, and used to train a custom AE, in an unsupervised way, to project the features into a compressed latent space. Latent features are used to train a supervised classifier (feed forward neural network) to decode the imagined movement. The proposed method was tested using a public dataset of EEGs collected from 109 subjects. The dataset consists of right-hand, left-hand, both hands, both feet motor imagery and resting EEGs. AE-FBCSP was extensively tested in the 3-way classification (right hand vs left hand vs resting) and also in the 2-way, 4-way and 5-way ones, both in cross- and intra-subject analysis. AE-FBCSP outperformed standard FBCSP in a statistically significant way (p > 0.05) and achieved a subject-specific average accuracy of 89.09% in the 3-way classification. The proposed methodology performed subject-specific classification better than other comparable methods in the literature, applied to the same dataset, also in the 2-way, 4-way and 5-way tasks. One of the most interesting outcomes is that AE-FBCSP remarkably increased the number of subjects that responded with a very high accuracy, which is a fundamental requirement for BCI systems to be applied in practice.

In the Covid-19 era, it is important to have an edge detector for X-ray (XR) images affected by uncertainties with low computational load but with high performance. So, here, a new version of a well-known fuzzy edge detector, in which a new image fuzzification procedure has been formulated, is proposed. The performance were qualitatively/ quantitatively compared with those obtained by Canny’s edge detector (gold standard for this type of problem). In addition, an evolution of the deep fuzzy-neural model named CovNNet, recently proposed by the authors to discriminate chest XR (CXR) images of patients with Covid-19 pneumonia from images of patients with interstitial pneumonias not related to Covid-19 (No-Covid-19), is presented and referred as to Enhanced-CovNNet (ECovNNet). Here, the generalization ability of it is also improved by introducing a regularization based on dropping out some nodes of the network in a random way. ECovNNet processes input CXR images and the corresponding fuzzy CXR images (processed through the proposed enhanced-fuzzy edge detector) and extracts relevant CXR/fuzzy features, subsequently combined in a single array named CXR and fuzzy features vector. The latter is used as input to an Autoencoder-(AE)-based classifier to perform the binary classification: Covid-19 and No-Covid-19, reporting accuracy rate up to 81%. Finally, the work is completed with some interesting physico-mathematical results.

In this study a Deep Learning (DL) based-Brain-Computer Interface (BCI) system able to automatically detect and decode voluntary eye blinks from the analysis of electroen-cephalographic (EEG) signals is proposed for controlling, in principle, an external device by means of ocular movements. To this end, a Convolutional Neural Network (CNN) is developed to classify EEG recordings related to natural (or involuntary) blinks, forced (or voluntary) blinks and baseline (no blinks) category. The proposed system achieved an impressive average classification performance: accuracy rate up to 99.4% +/- 1.3%. However, the core of the present study was to investigate the explainability and interpretability of the proposed CNN with the ultimate aim of explore which segments of the EEG signal is the most relevant in the voluntary/involuntary blink discrimination process. To this end, explainable Artificial Intelligence (xAI) techniques were applied. Specifically, the Gradient-weighted Class Activation Mapping (Grad-CAM) and the Local Interpretable Model Agnostic Explanation (LIME) algorithms were used. xAI allowed us to visually identify the most relevant EEG areas especially for the voluntary and involuntary blink detection. Indeed, limited to the analyzed dataset, for natural blinks, the discriminating region was the interval ranged from the temporal instant the eye was closed till the following instants of the reopening (vice-versa for voluntary blinks). The baseline (no blink), on the other hand, was characterized by a low activation threshold throughout the EEG segment.

This paper proposes a generative model and transfer learning powered system for classification of Scanning Electron Microscope (SEM) images of defective nanofibers (D-NF) and nondefective nanofibers (ND-NF) produced by electrospinning (ES) process. Specifically, a conditional-Generative Adversarial Network (c-GAN) is developed to generate synthetic D-NF/ND-NF SEM images. A transfer learning-oriented strategy is also proposed. First, a Convolutional Neural Network (CNN) is pre-trained on real images. The transfer-learned CNN is trained on synthetic SEM images and validated on real ones, reporting accuracy rate up to 95.31%. The achieved encouraging results endorse the use of the proposed generative model in industrial applications as it could reduce the number of needed laboratory ES experiments that are costly and time consuming.

Being able to decode the subject's intention to move is still a major challenge in the field of Brain Computer Interfaces (BCI). Even more, decoding the intention to perform movements from the motor preparation phase is a still largely unexplored topic, as most of the efforts have been focused so far on motor imagery. The present paper deals with BCIs based on electroencephalography (EEG), the best candidate for future systems meant for widespread use, with the goal of decoding the preparation of hand open/close movement from the EEG recordings of the subject. To this end, a dataset of EEG signals recorded in the 1s frame preceding the onset of movement are extracted from a publicly available database. Epochs are properly pre-filtered between 0.5 and 32 Hz and labelled as pre-hand closing (HC), pre-hand opening (HO) or resting (RE) epochs. A system for motion planning decoding, based on a custom Convolutional Neural Network (CNN) and named “Motion Planning Neural Network” MPnnet, is designed, trained and tested over the constructed dataset, achieving a mean HC-RE and HO-RE accuracy of$90.77 \pm 5.56\%$and of$92.48 \pm 4.3\%$, respectively. MPnnet matched the performance of more complex systems proposed in the past, allowing to skip the inverse problem solution step and showing to be able to self-learn relevant features directly from scalp EEG signals.

In this paper, a hybrid-domain deep learning (DL)-based neural system is proposed to decode hand movement preparation phases from electroencephalographic (EEG) recordings. The system exploits information extracted from the temporal-domain and time-frequency-domain, as part of a hybrid strategy, to discriminate the temporal windows (i.e. EEG epochs) preceding hand sub-movements (open/close) and the resting state. To this end, for each EEG epoch, the associated cortical source signals in the motor cortex and the corresponding time-frequency (TF) maps are estimated via beamforming and Continuous Wavelet Transform (CWT), respectively. Two Convolutional Neural Networks (CNNs) are designed: specifically, the first CNN is trained over a dataset of temporal (T) data (i.e. EEG sources), and is referred to as T-CNN; the second CNN is trained over a dataset of TF data (i.e. TF-maps of EEG sources), and is referred to as TF-CNN. Two sets of features denoted as T-features and TF-features, extracted from T-CNN and TF-CNN, respectively, are concatenated in a single features vector (denoted as TTF-features vector) which is used as input to a standard multi-layer perceptron for classification purposes. Experimental results show a significant performance improvement of our proposed hybrid-domain DL approach as compared to temporal-only and time-frequency-only-based benchmark approaches, achieving an average accuracy of [Formula: see text]%.

The goal of this paper is to develop a Brain Computer Interface (BCI) based on voluntary eye blinks decoding. In particular, the study was focused on the signals generated in the cortex by eye blinking, which can be collected by frontopolar scalp Electroencephalographic (EEG) sensors. Normally, EEG recording systems meant for clinical applications are expensive and cannot be used in large-scale user-friendly applications. Thanks to a prototype made by the STMicroelectronics company, based on an Open Source EEG project, a low-cost EEG recording system was created in this work. The goal is to develop an algorithm that can detect and discriminate between voluntary (forced) and involuntary (natural) blinking so that, in the future, an EEG-based BCI system that is able to control a device through eye movements could be developed, which would be of great use for all people with motor disabilities who can control eye movements. The proposed algorithm is based on a one-dimensional (1D) Convolutional Neural Network (CNN) architecture. Frontopolar EEG signals were collected during the execution of voluntary and spontaneous blinks by four healthy subjects. A dataset of EEG epochs of including blinks was constructed and used to train and validate the proposed CNN. The proposed system allowed to discriminate the blinks performed by the subjects (voluntary vs. involuntary) with an average accuracy of 97.92%.

In this paper, a novel deep learning based approach is proposed for the automatic classification of Electroencephalographic (EEG) signals of subjects diagnosed with the dementia of Alzheimer's disease (AD), Mild Cognitive Impairment (MCI) and Healthy Control (HC). Specifically, a custom Convolutional Neural Network (CNN) is designed to receive as input AD/MCI/HC EEG segments (epochs) of the same temporal width, and perform 2-way classification tasks: AD vs. HC, AD vs. MCI, MCI vs. HC. Our proposed architecture, termed EEG-CNN, is shown to exhibit remarkable abilities to self-learn relevant features directly from the EEG traces, avoiding the need for hand-crafted feature extraction engineering. Comparative experimental results demonstrate the promising performance of EEG-CNN, which is based on an analysis of the EEG time series only, reporting accuracies of 85.78 ± 2.18%, 69.03 ± 1.33%, 85.34 ± 1.86% in AD vs. HC, AD vs. MCI and MCI vs. HC classifications, respectively.

In the last decades, the use of Machine Learning (ML) algorithms have been widely employed to aid clinicians in the difficult diagnosis of neurological disorders, such as Alzheimer's disease (AD). In this context, here, a data-driven ML system for classifying Electroencephalographic (EEG) segments (i.e. epochs) of patients affected by AD, Mild Cognitive Impairment (MCI) and Healthy Control (HC) individuals, is introduced. Specifically, the proposed ML system consists of evaluating the average Time-Frequency Map (aTFM) related to a 19-channels EEG epoch and extracting some statistical coefficients (i.e. mean, standard deviation, skewness, kurtosis and entropy) from the main five conventional EEG sub-bands (or EEG-rhythms: delta, theta, alpha1, alpha2, beta). Afterwards, the time-frequency features vector is fed into an Autoeconder (AE), a Multi-Layer Perceptron (MLP), a Logistic Regression (LR) and a Support Vector Machine (SVM) based classifier to perform the 2-ways EEG epoch-classification tasks: AD vs HC and AD vs MCI. The performances of the proposed approach have been evaluated on a dataset of 189 EEG signals (63 AD, 63 MCI and 63 HC), recorded during an eye-closed resting condition at IRCCS Centro Neurolesi Bonino Pulejo of Messina (Italy). Experimental results reported that the 1-hidden layer MLP (MLP1) outperformed all the other developed learning systems as well as recently proposed state-of-the-art methods, achieving accuracy rate up to 95.76 % ± 0.0045 and 86.84% ± 0.0098 in AD vs HC and AD vs MCI classification, respectively.

Alzheimer's disease (AD) is a neurodegenerative disorder that causes a loss of connections between neurons. The goal of this paper is to construct a complex network model of the brain-electrical activity, using high-density EEG (HD-EEG) recordings, and to compare the network organization in AD, mild cognitive impaired (MCI), and healthy control (CNT) subjects. The HD-EEG of 16 AD, 16 MCI, and 12 CNT was recorded during an eye-closed resting state. The permutation disalignment index (PDI) was used to describe the dissimilarity between EEG signals and to construct the connection matrices of the network model. The three groups were found to have significantly different (p <; 0.001) characteristic path length (λ), average clustering coefficient (CC), and the global efficiency (GE). This is the first time that HD-EEG signals of AD, MCI, and CNT have been compared and that PDI has been used to discriminate between the three groups. Considering the large amount of data originating from HD-EEG acquisition, compared to standard EEG, the aim of this paper is also to assess that compression did not alter the results of the complex network analysis. Compressive sensing was adopted to compress and reconstruct the HD-EEG signals with minimal information loss, achieving an average structural similarity index of 0.954 (AD), 0.957 (MCI), and 0.959 (CNT). When applied to the reconstructed HD-EEG, complex network analysis provided a substantially unaltered performance, compared to the analysis of the original signals: λ, CC, and GE of the three groups were indeed still significantly different (p <; 0.001).

In this paper, a novel electroencephalographic (EEG)-based method is introduced for the quantification of brain-electrical connectivity changes over a longitudinal evaluation of mild cognitive impaired (MCI) subjects. In the proposed method, a dissimilarity matrix is constructed by estimating the coupling strength between every pair of EEG signals, Hierarchical clustering is then applied to group the related electrodes according to the dissimilarity estimated on pairs of EEG recordings. Subsequently, the connectivity density of the electrodes network is calculated. The technique was tested over two different coupling strength descriptors: wavelet coherence (WC) and permutation Jaccard distance (PJD), a novel metric of coupling strength between time series introduced in this paper. Twenty-five MCI patients were enrolled within a follow-up program that consisted of two successive evaluations, at time T0 and at time T1, three months later. At T1, four subjects were diagnosed to have converted to Alzheimer's Disease (AD). When applying the PJD-based method, the converted patients exhibited a significantly increased PJD (p < 0.05), i.e., a reduced overall coupling strength, specifically in delta and θ bands and in the overall range (0.5-32 Hz). In addition, in contrast to stable MCI patients, converted patients exhibited a network density reduction in every subband (delta, θ, alpha, and beta). When WC was used as coupling strength descriptor, the method resulted in a less sensitive and specific outcome. The proposed method, mixing nonlinear analysis to a machine learning approach, appears to provide an objective evaluation of the connectivity density modifications associated to the MCI-AD conversion, just processing noninvasive EEG signals.

In this paper, the need of novel methods to extract diagnostic information from the Electroencephalographic (EEG) recordings of epileptic patients was addressed. A novel method, based on Wavelet Coherence (WC) between EEG signals and Hierarchical Clustering (HC), was proposed to estimate the EEG network connectivity density in Childhood Absence Epilepsy (CAE) patients. The EEG recordings of four patients affected by CAE were partitioned into non overlapping windows and WC was estimated window by window. The behaviour of WC was analysed over the time, for every couple of EEG electrodes. The ictal states (seizures) resulted associated to increased WC levels, thus reflecting an increased synchronization between electrodes during the seizure. A WC-based dissimilarity index was then defined and HC was fed with the dissimilarity indices between every pair of electrodes with the aim of finding possible correlations between changes in electrode clustering and changes in the brain state. For every window under analysis, a dendrogram was constructed, the corresponding set of electrode clusters was determined and the subsequent network density values were calculated. Seizures resulted typically associated to increased network density, reflecting an increased connectivity during the ictal states.

OBJECTIVE: In this work, we introduce Permutation Disalignment Index (PDI) as a novel nonlinear, amplitude independent, robust to noise metric of coupling strength between time series, with the aim of applying it to electroencephalographic (EEG) signals recorded longitudinally from Alzheimer's Disease (AD) and Mild Cognitive Impaired (MCI) patients. The goal is to indirectly estimate the connectivity between the cortical areas, through the quantification of the coupling strength between the corresponding EEG signals, in order to find a possible matching with the disease's progression. METHOD: PDI is first defined and tested on simulated interacting dynamic systems. PDI is then applied to real EEG recorded from 8 amnestic MCI subjects and 7 AD patients, who were longitudinally evaluated at time [Formula: see text]0 and 3 months later (time [Formula: see text]1). At time [Formula: see text]1, 5 out of 8 MCI patients were still diagnosed MCI (stable MCI) whereas the remaining 3 exhibited a conversion from MCI to AD (prodromal AD). PDI was compared to the Spectral Coherence and the Dissimilarity Index. RESULTS: Limited to the size of the analyzed dataset, both Coherence and PDI resulted sensitive to the conversion from MCI to AD, even though only PDI resulted specific. In particular, the intrasubject variability study showed that the three patients who converted to AD exhibited a significantly ([Formula: see text]) increased PDI (reduced coupling strength) in delta and theta bands. As regards Coherence, even though it significantly decreased in the three converted patients, in delta and theta bands, such a behavior was also detectable in one stable MCI patient, in delta band, thus making Coherence not specific. From the Dissimilarity Index point of view, the converted MCI showed no peculiar behavior. CONCLUSIONS: PDI significantly increased, in delta and theta bands, specifically in the MCI subjects who converted to AD. The increase of PDI reflects a reduced coupling strength among the brain areas, which is consistent with the expected connectivity reduction associated to AD progression.

A novel technique of quantitative EEG for differentiating patients with early-stage Creutzfeldt-Jakob disease (CJD) from other forms of rapidly progressive dementia (RPD) is proposed. The discrimination is based on the extraction of suitable features from the time-frequency representation of the EEG signals through continuous wavelet transform (CWT). An average measure of complexity of the EEG signal obtained by permutation entropy (PE) is also included. The dimensionality of the feature space is reduced through a multilayer processing system based on the recently emerged deep learning (DL) concept. The DL processor includes a stacked auto-encoder, trained by unsupervised learning techniques, and a classifier whose parameters are determined in a supervised way by associating the known category labels to the reduced vector of high-level features generated by the previous processing blocks. The supervised learning step is carried out by using either support vector machines (SVM) or multilayer neural networks (MLP-NN). A subset of EEG from patients suffering from Alzheimer's Disease (AD) and healthy controls (HC) is considered for differentiating CJD patients. When fine-tuning the parameters of the global processing system by a supervised learning procedure, the proposed system is able to achieve an average accuracy of 89%, an average sensitivity of 92%, and an average specificity of 89% in differentiating CJD from RPD. Similar results are obtained for CJD versus AD and CJD versus HC.

Alzheimer's disease (AD) is a degenerative neurological disorder characterized by a loss of functional connections between different areas of the brain. AD is considered a cortical dementia, thus Electroencephalography (EEG) has been used as a tool for diagnosing AD for the last two decades. Often, the hallmarks of EEG abnormality in AD patients are a shift of the power spectrum to lower frequencies and reduced coherences among cortical regions, however, it is still mostly unknown how these abnormalities evolve together with the disease progression. In this paper we proposed a longitudinal study of the EEG of three AD patients in order to study the disease progression, from the coherence point of view, over the four major EEG sub-bands: delta, theta, alpha and beta. The EEG was recorded at time T0 and then after three months (time T1). We proposed a coherence based hierarchical clustering method to group the electrodes together according to their mutual pairwise coherence, in order to evaluate how the brain connectivity changed along with the disease in the spectral domain. The results provide an in-depth view of the structure of electrode interconnection of every single patient in every sub-band at time T0 and time T1. This study endorsed the commonly shared belief that coherence reduces over time but it revealed that coherence spatial distribution changes in a different way, from patient to patient. The results also showed that a patient-specific brain connectivity analysis is possible and that a personalized analysis of the disease's progression might provide valuable diagnostic information. In the near future, the study will be extended to a larger dataset in order to validate the method statistically.

Motivated by the dramatic rise of neurological disorders, we propose a SNN technique to model electroen-cephalography (EEG) data collected from people affected by Alzheimer's Disease (AD) and people diagnosed with mild cognitive impairment (MCI). An evolving spatio-temporal data machine (eSTDM), named the NeuCube architecture, is used to analyse changes of neural activity across different brain regions. The model developed allows for studying AD progression and for predicting whether a patient diagnosed with MCI is more likely to develop AD.

Epilepsy is the most diffuse brain disorder that can affect people's lives even on its early stage. In this paper, we used for the first time the spiking neural networks (SNN) framework called NeuCube for the analysis of electroencephalography (EEG) data recorded from a person affected by Absence Epileptic (AE), using permutation entropy (PE) features. Our results demonstrated that the methodology constitutes a valuable tool for the analysis and understanding of functional changes in the brain in term of its spiking activity and connectivity. Future applications of the model aim at personalised modelling of epileptic data for the analysis and the event prediction.

In this paper, we address the issue of dealing with huge amounts of data from recordings of an Electroencephalogram (EEG) in epileptic patients. In particular, the attention is focused on the development of tools to support the neurophysiologists in the time consuming and challenging task of reviewing the EEG to identify critical events that are worth of inspection for diagnostic purposes. A novel methodology is proposed for the automatic estimation of descriptors of EEG complexity and the subsequent classification of critical events. Based on the estimation of Permutation Entropy (PE) profiles from the EEG traces, the methodology relies on Learning Vector Quantization (LVQ) to cluster the electrodes in a competitive way according to their PE levels and to classify the cerebral state accordingly. An absence seizure EEG of 15.5 minutes was processed and a 93.94% sensitivity together with a 100% specificity were obtained.

A complex network approach is combined with time dynamics in order to conduct a space-time analysis applicable to longitudinal studies aimed to characterize the progression of Alzheimer's disease (AD) in individual patients. The network analysis reveals how patient-specific patterns are associated with disease progression, also capturing the widespread effect of local disruptions. This longitudinal study is carried out on resting electroence phalography (EEGs) of seven AD patients. The test is repeated after a three months' period. The proposed methodology allows to extract some averaged information and regularities on the patients' cohort and to quantify concisely the disease evolution. From the functional viewpoint, the progression of AD is shown to be characterized by a loss of connected areas here measured in terms of network parameters (characteristic path length, clustering coefficient, global efficiency, degree of connectivity and connectivity density). The differences found between baseline and at follow-up are statistically significant. Finally, an original topographic multiscale approach is proposed that yields additional results.

Electroencephalogram (EEG) is a non-invasive diagnostic tool in clinical neurophysiology, especially with respect to epilepsy. The epileptic status is characterized by reduced complexity. New markers, based on nonlinear dynamics, like Permutation Entropy (PE) have been developed to measure EEG complexity. In this paper, Multiscale Permutation Entropy (MPE) complexity measure is proposed as a potentially useful framework for detecting epileptic events in EEG data and to distinguish healthy controls from patients. The achieved results show that: 1) MPE is able to discriminate between the two categories; 2) the use of multiple scales may substantially improve the specificity of the diagnosis. This is shown through an SVM-based classification network with three different kernels. The use of the SVM approach is also useful to infer clues about the extracted features.

Epileptic seizures are thought to be generated and to evolve through an underlying anomaly of synchronization in the activity of groups of neuronal populations. The related dynamic scenario of state transitions is revealed by detecting changes in the dynamical properties of Electroencephalography (EEG) signals. The recruitment procedure ending with the crisis can be explored through a spatial-temporal plot from which to extract suitable descriptors that are able to monitor and quantify the evolving synchronization level from the EEG tracings. In this paper, a spatial-temporal analysis of EEG recordings based on the concept of permutation entropy (PE) is proposed. The performance of PE are tested on a database of 24 patients affected by absence (generalized) seizures. The results achieved are compared to the dynamical behavior of the EEG of 40 healthy subjects. Being PE a feature which is dependent on two parameters, an extensive study of the sensitivity of the performance of PE with respect to the parameters' setting was carried out on scalp EEG. Once the optimal PE configuration was determined, its ability to detect the different brain states was evaluated. According to the results here presented, it seems that the widely accepted model of "jump" transition to absence seizure should be in some cases coupled (or substituted) by a gradual transition model characteristic of self-organizing networks. Indeed, it appears that the transition to the epileptic status is heralded before the preictal state, ever since the interictal stages. As a matter of fact, within the limits of the analyzed database, the frontal-temporal scalp areas appear constantly associated to PE levels higher compared to the remaining electrodes, whereas the parieto-occipital areas appear associated to lower PE values. The EEG of healthy subjects neither shows any similar dynamic behavior nor exhibits any recurrent portrait in PE topography.

The genesis of epileptic seizures is nowadays still mostly unknown. The hypothesis that most of scientist share is that an abnormal synchronization of different groups of neurons seems to trigger a recruitment mechanism that leads the brain to the seizure in order to reset this abnormal condition. If this is the case, a gradual transformation of the characteristics of the EEG can be hypothesized. It is therefore necessary to find a parameter that is able to measure the synchronization level in the EEG and, since the spatial dimension has to be taken into account if we aim to find out how the different areas in the brain recruit each other to develop the seizure, a spatio-temporal analysis of this parameter has to be carried out. In the present paper, a spatio-temporal analysis of EEG synchronization in 24 patients affected by absence seizure is proposed and the results are hereby reported and compared to the results obtained with a group of 40 healthy subjects. The spatio-temporal analysis is based on Permutation Entropy (PE). We found out that, ever since the interictal stages, fronto-temporal areas appear constantly associated to PE levels that are higher compared to the rest of the brain, whereas the parietal/occipital areas appear associated to low-PE. The brain of healthy subjects seems to behave in a different way because we could not see a recurrent behaviour of PE topography.

In this paper, we speculate that abstract art can become an useful paradigm for both studying the relationship between neuroscience and art, and as a benchmark problem for the researches on Autonomous Machine Learning (AML) in brain-like computation. In particular, we are considering the case of some Kandinskij's oeuvres. There, it seems to see a deliberate willingness of introducing some effects today's hugely studied in the neuroscience, namely, for the retrieval of mental visual images or the neural correlates underlying visual tasks. The genial use of colours, geometry and vague forms generates very complex pictures that, we claim, excite preferentially mid-hierarchic levels of the bottom-up/top-down architecture of the brain, widely recognized as a possible framework for implementing AML. We introduce a quantitative metric for confirming the intuitive and psychological ranking of complexity given to paintings and pictures, the Artistic Complexity. The paintings of the artist are analysed, by selecting appropriately the oeuvres in order to point out different aspects of the topic. The concept of non-extensive Tsallis entropy is also introduced in an information-theoretic perspective, to cope with long-range interactions, as is done in spectral analysis of the human brain EEG. fMRI experimentations are sought to justify our speculations.

Despite their high specific stiffness and strength, carbon fiber reinforced polymers, stacked at different fiber orientations, are susceptible to interlaminar damages. They may occur in the form of micro-cracks and voids, and leads to a loss of performance. Within this framework, ultrasonic tests can be exploited in order to detect and classify the kind of defect. The main object of this work is to develop the evolution of a previous heuristic approach, based on the use of Support Vector Machines, proposed in order to recognize and classify the defect starting from the measured ultrasonic echoes. In this context, a real-time approach could be exploited to solve real industrial problems with enough accuracy and realistic computational efforts. Particularly, we discuss the cross wavelet transform and wavelet coherence for examining relationships in time-frequency domains between. For our aim, a software package has been developed, allowing users to perform the cross wavelet transform, the wavelet coherence and the Fuzzy Inference System. Since the ill-posedness of the inverse problem, Fuzzy Inference has been used to regularize the system, implementing a data-independent classifier. Obtained results assure good performances of the implemented classifier, with very interesting applications.

Works in literature analyse neuronal electrical source distribution on human brain cortex from noninvasive measurements of electric potentials at multichannel scalp electrodes (Electroencephalography). In this paper a new technique, based on entropy, is introduced to study the synchronization of the electric activity of neuronal sources in the brain. An abnormal synchronization (abnormal interactions between the electrodes) can result in an epileptic seizure and Renyi's entropy topography and its spatial clustering are proposed to measure this interaction and to provide a spatial picture of its distribution over the cortex. Three EEG dataset from patients affected by partial epilepsy were analysed. Entropy showed a very steady spatial distribution and a clear relationship with the region of seizure onset. Entropy mapping was compared with the standard power mapping that was much less stable and selective. A SOM based spatial clustering of entropy topography showed that the critical electrodes were coupled together long time before the seizure onset.

Results in literature show that the convergence of the Short-Term Maximum Lyapunov Exponent (STLmax) time series, extracted from intracranial EEG recorded from patients affected by intractable temporal lobe epilepsy, is linked to the seizure onset. When the STLmax profiles of different electrode sites converge (high entrainment) a seizure is likely to occur. In this paper Renyipsilas Mutual information (MI) is introduced in order to investigate the independence between pairs of electrodes involved in the epileptogenesis. A scalp EEG recording and an intracranial EEG recording, including two seizures each, were analysed. STLmax was estimated for each critical electrode and then MI between couples of STLmax profiles was measured. MI showed sudden spikes that occurred 8 to 15 min before the seizure onset. Thus seizure onset appears related to a burst in MI: this suggests that seizure development might restore the independence between STLmax of critical electrode sites.

Electroencephalographic (EEG) recordings are often contaminated by the artifacts, signals that have non-cerebral origin and that might mimic cognitive or pathologic activity and therefore distort the analysis of EEG. In this paper the issue of artifact extraction from Electroencephalographic data is addressed and a new technique for EEG artifact removal, based on the joint use of Wavelet transform and Independent Component Analysis (WICA), is presented and compared to two other techniques based on ICA and wavelet denoising. An artificial artifact-laden EEG dataset was created mixing a real EEG with a set of synthesized artifacts. This dataset was processed by WICA and the two other methods. The proposed technique had the best artifact separation performance for every kind of artifact also allowing for the minimum information loss.

Digital hardware implementations of neural networks demand the efficient computation of the neurons activation function. In this paper, two new circuits to implement a programmable tanh-like activation function and its derivative are presented. The function exhibits learning and generalization abilities perfectly comparable with those achieved by the typical activation functions but, what is more, can be easily implemented in hardware through only binary shift operations. The first derivative, contrary to the classical solutions, requires a simple constant coefficient multiplier resulting in a great advantage in terms of area occupancy and performance. The accuracy analysis carried out by the average and maximum error, the high computational speed and the small amount of hardware resources point out that the proposed approach is fully competitive with the most recent implementations of activation functions capable of learning.

This paper is concerned with the application of novel techniques of data interpretation for reconstructing plasma shape in Tokamak reactors for nuclear fusion applications. In particular, Artificial Neural Networks have been taken into account to estimate the distance of the plasma boundary from the fist wall of the vacuum vessel in ITER configuration. In addition, a comparison with Principal Component Analysis and Functional Parameterization is presented. Finally, in order to reduce the computational complexity, non linear techniques for ranking sensors is exploited.

This paper presents an efficient approach for the prediction of sunspot-related time series commonly used for monitoring solar activity, namely the Yearly Sunspot Number and the R12 Index. The method consists in matching a "derectification" procedure of sunspot data with the use of nonlinear dynamics tools in order to design neural network based predictors with close-to-optimal performance. In fact, whereas the "de-rectification" process allows to obtain time series that can be modeled by neural structures much better than the original datasets, the incorporation of the prior knowledge extracted by using nonlinear dynamics into neural networks generates models able to fully capture the chaotic dynamics of solar activity. The proposed approach produces prediction results that outperform the most accurate methods existing in literature both for short and medium-term forecasting horizons.

The electroencephalogram (EEG) analysis provides a functional tool to verify a qualitative clinical check. In this paper some techniques, i.e. principal component analysis (PCA) and independent component analysis (ICA), are implemented in order to extrapolate in EEG signals very few dominant components that contain almost all the information necessary to have an adequate knowledge of the brain activity. To obtain that, the compression ability of PCA is mixed with the statistical independence property of the ICA. The achieved results show that in most cases of cerebral death diagnosis, in which the EEG analysis is performed when the brain activity is very reduced, even few components are enough to depict the complete brain activity.

One of the aims of biomedical signal processing is to extract some features from the data in order to make diagnosis and to understand the biological phenomena but, often, a preprocessing step is essential because some unwelcome signals, the artifacts, are superimposed to the useful signals we want to analyse. Automatic artifact detection is a key topic, because we aim to automatically analyse and extract features from the data. In literature, independent component analysis (ICA) has been exploited for artifact isolation and the joint use of some high order statistics, kurtosis and Shannon's entropy has been exploited to automatically detect the artifacts. In this paper we propose the joint use of kurtosis and Renyi's entropy as a new tool for automatic detection and we show that it outperforms the other tool thanks to the features of the Renyi's entropy.

In this paper a new approach to design efficient neural networks based predictors of noise-free chaotic time series is proposed. Using tools of chaos theory, we can provide helpful indications to appropriately design the architectures of time delay neural networks in a very rapid fashion. After that, by combining an efficient data pre-processing with Bayesian learning, we train neural models that are able to fully capture the dynamics of the underlying systems creating powerful predictors of chaotic time series. We test on several benchmarks the proposed approach achieving results comparable or even better than those of many recurrent neural networks. We also prove that the existing local models lose their well-known advantage when compared to our method, with the benefit of using a much smaller number of parameters.

Recent works have shown that artifacts removal in biomedical signals, like electromyographic (EMG) or electroencephalographic (EEG) recordings, can be performed by using discrete wavelet transform (DWT) or independent component analysis (ICA). Often, the removal of some artifacts is very hard because they are superimposed on the recordings and they corrupt biomedical signals also in frequency domain. In these cases DWT and ICA methods cannot perform artifacts cancellation. We present a method based on the joint use of wavelet transform and independent component analysis. We show the obtained results and the comparisons among the proposed method, DWT and ICA techniques. In this preliminary study, a user interface is needed to identify the artifact.

We exploit the potentials of Bayesian neural networks combined with the advantages of a VLSI implementation in order to design a stand-alone predictor system of air pollutants time series. The area under study is Villa San Giovanni, a small town located in front of the Messina Strait (Italy), whose harbor represents the main link to reach Sicily island by cars and trucks. Neural networks are powerful tools to predict air pollutants time series, but almost always they run by software programs on PC or workstations, which make difficult their use when are present constraints such as portability, low power dissipation, limited physical size. In this cases, SRAM based field programmable gate arrays (FPGAs) represent a suitable platform to realize these models, since their reprogrammability offers the possibility to rapidly change the parameters of the network if a new training is needed. The achieved results have highlighted the efficient design of the hardware network, obtained also using a new circuit to compute the activation function of the neurons.

The main effort of this paper is directed towards the characterization of multi-channel muscle contractions recordings measured by surface electromyography (sEMG) for both research and clinical purposes. In particular, an attempt is made in order to describe the kind of modifications in the spectrum and related frequency content of the sEMG data when the forces produced by muscles are varying. Recent works have proposed time-frequency analysis as a powerful tool to investigate some parameters that relate to the progression of muscle fatigue. A different method of time-frequency characterization by means of unsupervised learning processing is here proposed: the growing neural gas (GNG) algorithm is used. The advantage of the proposed method with respect to traditional methods, that make use of the mean or median frequency, seems the complete description of the frequency content of the signal. The obtained results are in agreement with physiologic studies of muscle activity.

The paper proposes the use of independent component analysis (ICA), an unsupervised learning technique, in order to process raw surface electromyographic (sEMG) data by reducing the typical "cross-talk" effect on the electric interference pattern measured by the surface sensors. The ICA is implemented by means of a multi-layer NN scheme. The basic tool is the wavelet decomposition that allows us to detect and analyse time-varying signals. An auto-associative NN that exploits wavelet coefficients an input vector is also used as simple detector of non-stationary based on a measure of reconstruction error. In addition, Morlet wavelets have been exploited for classification problems.

Artificial neural networks are widely used as predictor systems for the pollutants time series. In recent years, the dynamic system theory has also been exploited to find the optimal sampling time interval and the minimum embedding dimension of environmental time series in order to get helpful information and to implement appropriately the forecasting networks. In this paper, we present a novel approach to predict the concentration level of air pollutants in the area of the Messina Strait, whose harbor represents the unique link to reach Sicily Island from Europe by cars and trucks. By coupling feedforward neural networks with Cao's method, we predict the level of carbon monoxide and hydrocarbons from one to ten hours ahead with an accuracy of more than 90%.

A neural-based signal processing system that exploits radial basis function neural network (RBFNN) is proposed to solve the problem of detecting and locating circular holes in conducting plates by means of nondestructive eddy currents testing. The capabilities of basic multilayer perceptron and radial basis function (RBF) schemes are first investigated. Since the achieved performance revealed insufficient, a two-step procedure is then analyzed: in the first step, an RBFNN is used to estimate the distances between the hole's center and the eddy current magnetic sensors; a least square algorithm is then exploited in order to locate the hole starting from the previously estimated distances. The performance of the proposed system are tested on a database of simulated experiments based on the a priori knowledge of the corresponding boundary value direct problem solution, by taking advantage of the closed-form analytical expression of the solution in order to generate a wide range of possible sensor-hole configurations. Both noiseless and noisy measurements are taken into account for assessing the system robustness. The main result achieved is discussed.

A data fusion system based on independent component analysis (ICA) has been applied to remove the negative effects produced by the lift-off variations of eddy current (EC) sensors, in the context of crack detection and recognition. Due to the sensor drift effect, EC magnitude and phase measurements are unavoidably affected by the lift-off noise that, in some cases, has a power content whose level is more than comparable to the defect related signal level. A set of measurements carried out on the specimen under inspection are sent as input to a neural network trained to perform the ICA of the input data: each sample measurement can be interpreted as a linear combination of quasi-independent signals related to measurement noise, lift-off noise and flaw presence. The basic work hypothesis is that the lift-off signal is present in multiple views of the specimen in the form of correlated noise. As a consequence, the ICA will be able to fuse the knowledge provided by magnitude and phase signals in different measuring contexts, in order to extract the lift-off noise from the input signals and to separate it from the signal related to the crack.

We propose a radial basis function neural network (RBFNN) approach to the identification of holes in conducting plates, in the context of an eddy current testing (ECT) signal processing system. The system aims to localise holes in the specimen under inspection by using a two-stage approach, namely, a RBFNN followed by a least squares post-processing block. The RBFNN stage estimates the distances between the hole and the sensor probes; the least squares stage identifies the hole on the basis of the distances computed by the previous neural block. The efficacy of the proposed approach is tested on artificial data and compared with different approaches based on a feedforward multilayer perceptron (MLP) and on a radial basis function neural network. The robustness of the system to the introduction of white Gaussian noise on the simulated data is also successfully tested.

The problem of extracting relevant information about a defective specimen from external noninvasive measurements is hardly solvable without exploiting recent advances in signal processing. The usual two-step NDT/NDE problem (detection and reconstruction of the defect) can indeed be interpreted as a pattern recognition most in which the feature extraction aspect is by far the most interesting from the scientific viewpoint Various relevant feature extraction techniques are compared in this work aiming to finding the most advantageous mapping that reduces the dimensionality of the input patterns while presenting the relevant information content about the defect. The preprocessing analysis is also shown to yield peculiar advantages with respect to mere noise filtering of raw data.

This paper proposes the use of Fuzzy Neural Network approaches for the early detection of disruption in tokamak plasmas. Neural Networks can be used for classifying plasma shots and defecting disruptive shots as well as for estimating the time left before disrupting. The use of fuzzy logic concept is suggested because it offers a framework for embodying expert knowledge about predicting the onset of disruption. Moreover, learning approaches allow to tune the model expressed in terms of fuzzy statements. The proposed method appears to be a step forward with respect to more conventional NN approach.