Editorial Topical Collection: "Explainable and Augmented Machine Learning for Biosignals and Biomedical Images".

Machine learning (ML) is a well-known subfield of artificial intelligence (AI) that aims at developing algorithms and statistical models able to empower computer systems to automatically adapt to a specific task through experience or learning from data [...].

A Novel Hierarchical Extreme Machine-Learning-Based Approach for Linear Attenuation Coefficient Forecasting.

The development of reinforced polymer composite materials has had a significant influence on the challenging problem of shielding against high-energy photons, particularly X-rays and γ-rays in industrial and healthcare facilities. Heavy materials' shielding characteristics hold a lot of potential for bolstering concrete chunks. The mass attenuation coefficient is the main physical factor that is utilized to measure the narrow beam γ-ray attenuation of various combinations of magnetite and mineral powders with concrete. Data-driven machine learning approaches can be investigated to assess the gamma-ray shielding behavior of composites as an alternative to theoretical calculations, which are often time- and resource-intensive during workbench testing. We developed a dataset using magnetite and seventeen mineral powder combinations at different densities and water/cement ratios, exposed to photon energy ranging from 1 to 1006 kiloelectronvolt (KeV). The National Institute of Standards and Technology (NIST) photon cross-section database and software methodology (XCOM) was used to compute the concrete's γ-ray shielding characteristics (LAC). The XCOM-calculated LACs and seventeen mineral powders were exploited using a range of machine learning (ML) regressors. The goal was to investigate whether the available dataset and XCOM-simulated LAC can be replicated using ML techniques in a data-driven approach. The minimum absolute error (MAE), root mean square error (RMSE), and R2score were employed to assess the performance of our proposed ML models, specifically a support vector machine (SVM), 1d-convolutional neural network (CNN), multi-Layer perceptrons (MLP), linear regressor, decision tree, hierarchical extreme machine learning (HELM), extreme learning machine (ELM), and random forest networks. Comparative results showed that our proposed HELM architecture outperformed state-of-the-art SVM, decision tree, polynomial regressor, random forest, MLP, CNN, and conventional ELM models. Stepwise regression and correlation analysis were further used to evaluate the forecasting capability of ML techniques compared to the benchmark XCOM approach. According to the statistical analysis, the HELM model showed strong consistency between XCOM and predicted LAC values. Additionally, the HELM model performed better in terms of accuracy than the other models used in this study, yielding the highest R2score and the lowest MAE and RMSE.

A fuzzy-enhanced deep learning approach for early detection of Covid-19 pneumonia from portable chest X-ray images.

The Covid-19 pandemic is the defining global health crisis of our time. Chest X-Rays (CXR) have been an important imaging modality for assisting in the diagnosis and management of hospitalised Covid-19 patients. However, their interpretation is time intensive for radiologists. Accurate computer aided systems can facilitate early diagnosis of Covid-19 and effective triaging. In this paper, we propose a fuzzy logic based deep learning (DL) approach to differentiate between CXR images of patients with Covid-19 pneumonia and with interstitial pneumonias not related to Covid-19. The developed model here, referred to as CovNNet, is used to extract some relevant features from CXR images, combined with fuzzy images generated by a fuzzy edge detection algorithm. Experimental results show that using a combination of CXR and fuzzy features, within a deep learning approach by developing a deep network inputed to a Multilayer Perceptron (MLP), results in a higher classification performance (accuracy rate up to 81%), compared to benchmark deep learning approaches. The approach has been validated through additional datasets which are continously generated due to the spread of the virus and would help triage patients in acute settings. A permutation analysis is carried out, and a simple occlusion methodology for explaining decisions is also proposed. The proposed pipeline can be easily embedded into present clinical decision support systems.

Permutation Entropy-Based Interpretability of Convolutional Neural Network Models for Interictal EEG Discrimination of Subjects with Epileptic Seizures vs. Psychogenic Non-Epileptic Seizures.

The differential diagnosis of epileptic seizures (ES) and psychogenic non-epileptic seizures (PNES) may be difficult, due to the lack of distinctive clinical features. The interictal electroencephalographic (EEG) signal may also be normal in patients with ES. Innovative diagnostic tools that exploit non-linear EEG analysis and deep learning (DL) could provide important support to physicians for clinical diagnosis. In this work, 18 patients with new-onset ES (12 males, 6 females) and 18 patients with video-recorded PNES (2 males, 16 females) with normal interictal EEG at visual inspection were enrolled. None of them was taking psychotropic drugs. A convolutional neural network (CNN) scheme using DL classification was designed to classify the two categories of subjects (ES vs. PNES). The proposed architecture performs an EEG time-frequency transformation and a classification step with a CNN. The CNN was able to classify the EEG recordings of subjects with ES vs. subjects with PNES with 94.4% accuracy. CNN provided high performance in the assigned binary classification when compared to standard learning algorithms (multi-layer perceptron, support vector machine, linear discriminant analysis and quadratic discriminant analysis). In order to interpret how the CNN achieved this performance, information theoretical analysis was carried out. Specifically, the permutation entropy (PE) of the feature maps was evaluated and compared in the two classes. The achieved results, although preliminary, encourage the use of these innovative techniques to support neurologists in early diagnoses.

A Machine Learning Approach Involving Functional Connectivity Features to Classify Rest-EEG Psychogenic Non-Epileptic Seizures from Healthy Controls.

Until now, clinicians are not able to evaluate the Psychogenic Non-Epileptic Seizures (PNES) from the rest-electroencephalography (EEG) readout. No EEG marker can help differentiate PNES cases from healthy subjects. In this paper, we have investigated the power spectrum density (PSD), in resting-state EEGs, to evaluate the abnormalities in PNES affected brains. Additionally, we have used functional connectivity tools, such as phase lag index (PLI), and graph-derived metrics to better observe the integration of distributed information of regular and synchronized multi-scale communication within and across inter-regional brain areas. We proved the utility of our method after enrolling a cohort study of 20 age- and gender-matched PNES and 19 healthy control (HC) subjects. In this work, three classification models, namely support vector machine (SVM), linear discriminant analysis (LDA), and Multilayer perceptron (MLP), have been employed to model the relationship between the functional connectivity features (rest-HC versus rest-PNES). The best performance for the discrimination of participants was obtained using the MLP classifier, reporting a precision of 85.73%, a recall of 86.57%, an F1-score of 78.98%, and, finally, an accuracy of 91.02%. In conclusion, our results hypothesized two main aspects. The first is an intrinsic organization of functional brain networks that reflects a dysfunctional level of integration across brain regions, which can provide new insights into the pathophysiological mechanisms of PNES. The second is that functional connectivity features and MLP could be a promising method to classify rest-EEG data of PNES form healthy controls subjects.

Compressibility of High-Density EEG Signals in Stroke Patients.

Stroke is a critical event that causes the disruption of neural connections. There is increasing evidence that the brain tries to reorganize itself and to replace the damaged circuits, by establishing compensatory pathways. Intra- and extra-cellular currents are involved in the communication between neurons and the macroscopic effects of such currents can be detected at the scalp through electroencephalographic (EEG) sensors. EEG can be used to study the lesions in the brain indirectly, by studying their effects on the brain electrical activity. The primary goal of the present work was to investigate possible asymmetries in the activity of the two hemispheres, in the case one of them is affected by a lesion due to stroke. In particular, the compressibility of High-Density-EEG (HD-EEG) recorded at the two hemispheres was investigated since the presence of the lesion is expected to impact on the regularity of EEG signals. The secondary objective was to evaluate if standard low density EEG is able to provide such information. Eighteen patients with unilateral stroke were recruited and underwent HD-EEG recording. Each EEG signal was compressively sensed, using Block Sparse Bayesian Learning, at increasing compression rate. The two hemispheres showed significant differences in the compressibility of EEG. Signals acquired at the electrode locations of the affected hemisphere showed a better reconstruction quality, quantified by the Structural SIMilarity index (SSIM), than the EEG signals recorded at the healthy hemisphere (p < 0.05), for each compression rate value. The presence of the lesion seems to induce an increased regularity in the electrical activity of the brain, thus an increased compressibility.

Information Theoretic-Based Interpretation of a Deep Neural Network Approach in Diagnosing Psychogenic Non-Epileptic Seizures.

The use of a deep neural network scheme is proposed to help clinicians solve a difficult diagnosis problem in neurology. The proposed multilayer architecture includes a feature engineering step (from time-frequency transformation), a double compressing stage trained by unsupervised learning, and a classification stage trained by supervised learning. After fine-tuning, the deep network is able to discriminate well the class of patients from controls with around 90% sensitivity and specificity. This deep model gives better classification performance than some other standard discriminative learning algorithms. As in clinical problems there is a need for explaining decisions, an effort has been carried out to qualitatively justify the classification results. The main novelty of this paper is indeed to give an entropic interpretation of how the deep scheme works and reach the final decision.

A Few-Shot Transfer Learning Approach for Motion Intention Decoding from Electroencephalographic Signals.

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.

An explainable Artificial Intelligence approach to study MCI to AD conversion via HD-EEG processing.

An explainable Artificial Intelligence (xAI) approach is proposed to longitudinally monitor subjects affected by Mild Cognitive Impairment (MCI) by using high-density electroencephalography (HD-EEG). To this end, a group of MCI patients was enrolled at IRCCS Centro Neurolesi Bonino Pulejo of Messina (Italy) within a follow-up protocol that included two evaluations steps: T0 (first evaluation) and T1 (three months later). At T1, four MCI patients converted to Alzheimer's Disease (AD) and were included in the analysis as the goal of this work was to use xAI to detect individual changes in EEGs possibly related to the degeneration from MCI to AD. The proposed methodology consists in mapping segments of HD-EEG into channel-frequency maps by means of the power spectral density. Such maps are used as input to a Convolutional Neural Network (CNN), trained to label the maps as "T0" (MCI state) or "T1" (AD state). Experimental results reported high intra-subject classification performance (accuracy rate up to 98.97% (95% confidence interval: 98.68-99.26)). Subsequently, the explainability of the proposed CNN is explored via a Grad-CAM approach. The procedure detected which EEG-channels (i.e., head region) and range of frequencies (i.e., sub-bands) were more active in the progression to AD. The xAI analysis showed that the main information is included in the delta sub-band and that, limited to the analyzed dataset, the highest relevant areas are: the left-temporal and central-frontal lobe for Sb01, the parietal lobe for Sb02, the left-frontal lobe for Sb03 and the left-frontotemporal region for Sb04.

AutoEncoder Filter Bank Common Spatial Patterns to Decode Motor Imagery From EEG.

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.

A Conditional Generative Adversarial Network and Transfer Learning-Oriented Anomaly Classification System for Electrospun Nanofibers.

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.

A Hybrid-Domain Deep Learning-Based BCI For Discriminating Hand Motion Planning From EEG Sources.

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]%.

A deep CNN approach to decode motor preparation of upper limbs from time-frequency maps of EEG signals at source level.

A system that can detect the intention to move and decode the planned movement could help all those subjects that can plan motion but are unable to implement it. In this paper, motor planning activity is investigated by using electroencephalographic (EEG) signals with the aim to decode motor preparation phases. A publicly available database of 61-channels EEG signals recorded from 15 healthy subjects during the execution of different movements (elbow flexion/extension, forearm pronation/supination, hand open/close) of the right upper limb was employed to generate a dataset of EEG epochs preceding resting and movement's onset. A novel system is introduced for the classification of premovement vs resting and of premovement vs premovement epochs. For every epoch, the proposed system generates a time-frequency (TF) map of every source signal in the motor cortex, through beamforming and Continuous Wavelet Transform (CWT), then all the maps are embedded in a volume and used as input to a deep CNN. The proposed system succeeded in discriminating premovement from resting with an average accuracy of 90.3% (min 74.6%, max 100%), outperforming comparable methods in the literature, and in discriminating premovement vs premovement with an average accuracy of 62.47%. The achieved results encourage to investigate motor planning at source level in the time-frequency domain through deep learning approaches.

A novel multi-modal machine learning based approach for automatic classification of EEG recordings in dementia.

Electroencephalographic (EEG) recordings generate an electrical map of the human brain that are useful for clinical inspection of patients and in biomedical smart Internet-of-Things (IoT) and Brain-Computer Interface (BCI) applications. From a signal processing perspective, EEGs yield a nonlinear and nonstationary, multivariate representation of the underlying neural circuitry interactions. In this paper, a novel multi-modal Machine Learning (ML) based approach is proposed to integrate EEG engineered features for automatic classification of brain states. EEGs are acquired from neurological patients with Mild Cognitive Impairment (MCI) or Alzheimer's disease (AD) and the aim is to discriminate Healthy Control (HC) subjects from patients. Specifically, in order to effectively cope with nonstationarities, 19-channels EEG signals are projected into the time-frequency (TF) domain by means of the Continuous Wavelet Transform (CWT) and a set of appropriate features (denoted as CWT features) are extracted from δ, θ, α1, α2, ß EEG sub-bands. Furthermore, to exploit nonlinear phase-coupling information of EEG signals, higher order statistics (HOS) are extracted from the bispectrum (BiS) representation. BiS generates a second set of features (denoted as BiS features) which are also evaluated in the five EEG sub-bands. The CWT and BiS features are fed into a number of ML classifiers to perform both 2-way (AD vs. HC, AD vs. MCI, MCI vs. HC) and 3-way (AD vs. MCI vs. HC) classifications. As an experimental benchmark, a balanced EEG dataset that includes 63 AD, 63 MCI and 63 HC is analyzed. Comparative results show that when the concatenation of CWT and BiS features (denoted as multi-modal (CWT+BiS) features) is used as input, the Multi-Layer Perceptron (MLP) classifier outperforms all other models, specifically, the Autoencoder (AE), Logistic Regression (LR) and Support Vector Machine (SVM). Consequently, our proposed multi-modal ML scheme can be considered a viable alternative to state-of-the-art computationally intensive deep learning approaches.

Permutation Jaccard Distance-Based Hierarchical Clustering to Estimate EEG Network Density Modifications in MCI Subjects.

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.