Reconstructing images using brain signals of imagined visuals may provide an augmented vision to the disabled, leading to the advancement of Brain-Computer Interface (BCI) technology. The recent progress in deep learning has boosted the study area of synthesizing images from brain signals using Generative Adversarial Networks (GAN). In this work, we have proposed a framework for synthesizing the images from the brain activity recorded by an electroencephalogram (EEG) using small-size EEG datasets. This brain activity is recorded from the subject’s head scalp using EEG when they ask to visualize certain classes of Objects and English characters. We use a contrastive learning method in the proposed framework to extract features from EEG signals and synthesize the images from extracted features using conditional GAN. We modify the loss function to train the GAN, which enables it to synthesize 128 × 128 images using a small number of images. Further, we conduct ablation studies and experiments to show the effectiveness of our proposed framework over other state-of-the-art methods using the small EEG dataset.

Naturalistic music typically contains repetitive musical patterns that are present throughout the song. These patterns form a signature, enabling effortless song recognition. We investigate whether neural responses corresponding to these repetitive patterns also serve as a signature, enabling recognition of later song segments on learning initial segments. We examine EEG encoding of naturalistic musical patterns employing the NMED-T and MUSIN-G datasets. Experiments reveal that (a) training machine learning classifiers on the initial 20s song segment enables accurate prediction of the song from the remaining segments; (b) β and γ band power spectra achieve optimal song classification, and (c) listener-specific EEG responses are observed for the same stimulus, characterizing individual differences in music perception.

We examine user and song identification from neural (EEG) signals. Owing to perceptual subjectivity in human-media interaction, music identification from brain signals is a challenging task. We demonstrate that subjective differences in music perception aid user identification, but hinder song identification. In an attempt to address intrinsic complexities in music identification, we provide empirical evidence on the role of enjoyment in song recognition. Our findings reveal that considering song enjoyment as an additional factor can improve EEG-based song recognition.

Learning a sequence of movements is akin to the acquisition of a motor skill. We investigated event-related potentials (ERPs) changes, particularly the error-related negativity (ERN) and P200 components, as participants learned a motor sequence using a serial reaction time task. Unlike previous studies that investigated error-related negativity for only incorrect motor responses, we tracked ERN changes for all responses. We found that ERN decreased significantly from early to later stages of learning. We also observed a significant change in the P200 component associated with increased selective attention to relevant stimuli as learning occurred. Scalp topography showed the differences in neural activity during motor sequence learning in the frontal, central, and parietal regions. We then employed machine learning to identify the best predictors of motor learning stages. Using random forest, we found the most discriminating pattern for early versus late learning phases. The combination of three electrodes, including ‘C3-C4-P3’, obtained the maximum accuracy of 82% while classifying EEG signals corresponding to early and late stages of learning. Our study demonstrates that ERN and P200 signals can serve as temporal neural markers for motor skill learning.

Modern neuroscience has shown that the brain is profoundly rhythmic and that frequencies of neural rhythms are responsive to frequencies of musical rhythms. We collected Electroencephalography (EEG) response on 12 naturalistic music stimuli (songs), from 20 participants. We retrieved the tempo and its sub-harmonics from our stimuli (songs), and further used this information to predict the beats in the brain response using Machine Learning techniques. We observed a hierarchy of beats in each of the songs, with a specific beat frequency to be dominant (i.e. higher in magnitude) than others. This led us to form three groups of songs and their brain responses, with each of the groups indicating the frequency of a beat that dominated in the hierarchy of beat structure of that song. We used small segments of 1, 3 and 5 seconds of brain responses, rather than the entire song duration. We further created two sets for classification of the three groups of brain responses and utilized two spatial filtering techniques: Mean across electrodes (ME) and first principal component (PC1), and a Dense method using data from all electrodes. This was followed by feature extraction using band power. We developed univariate and multivariate models for classification to demonstrate the significance of each frequency band which represent beat frequencies. The dense method outperformed ME and PC1. Features related to eighth note generated maximum discrimination between classes. We also observed a positive correlation between window length and rate of correct prediction. Accuracy from one second to five seconds window improved significantly in both the sets. We achieved maximum accuracy of 70% and 56% accuracies for binary and ternary classification respectively, which is 20% above chance-level accuracy. Random Forest and kNN performed better than SVM. This work contributes to the growing body of knowledge to understand the underlying neural mechanism of rhythm processing in the brain.

With an increasing effort from the scientific community to quantify the effects of meditation practices, various approaches have been used to establish neural correlates of experienced meditators. Numerous techniques have been studied to identify meditation-related changes in brain signals, including network analysis, synchronization metrics, and spectral features. Here, we employ nonlinear measures to explore the dynamical aspects of electroencephalography (EEG) signals in participants who had completed an 8-week Mindfulness Based Stress Reduction training program via pre-post intervention changes. Three nonlinear complexity measures comprising Detrended Fluctuation Analysis (DFA), Higuchi fractal dimension (FD), and Katz FD are implemented to measure the complexity of EEG signals. We examine theta and alpha frequency bands with a specific focus on four sub-bands, along with four anterior and posterior regions. Random Forest (RF) with SHAP explainability method is used to generate the results. RF models are trained for classification on features extracted from pre- and post-session. Our findings reveal that (a) Higuchi FD exhibits a decline post-training session and delivers the best classifying results; (b) the alpha wave contributes the most in the left-right frontal and parietal regions; (c) and overall, the right hemisphere has greater involvement. These findings add to the growing body of knowledge about the neural correlates of meditation. This research’s implications for designing neurotechnology products that improve attention, awareness, and kindness have been discussed.

PySPH is an open-source, Python-based, framework for particle methods in general and Smoothed Particle Hydrodynamics (SPH) in particular. PySPH allows a user to define a complete SPH simulation using pure Python. High-performance code is generated from this high-level Python code and executed on either multiple cores, or on GPUs, seamlessly. It also supports distributed execution using MPI. PySPH supports a wide variety of SPH schemes and formulations. These include, incompressible and compressible fluid flow, elastic dynamics, rigid body dynamics, shallow water equations, and other problems. PySPH supports a variety of boundary conditions including mirror, periodic, solid wall, and inlet/outlet boundary conditions. The package is written to facilitate reuse and reproducibility. This article discusses the overall design of PySPH and demonstrates many of its features. Several example results are shown to demonstrate the range of features that PySPH provides.

This paper presents results from a rapid-response industry-academia collaboration for virtual screening of chemical, natural and virtual drug ligands towards identifying potential therapeutics for COVID-19. Compared to resource-intensive traditional approaches of either conducting high- throughput screening in a lab or in-silico molecular dynamics simulations on supercomputers, we have developed an open- source framework that leverages artificial intelligence (AI) to accurately and quickly predict the binding potential of a drug ligand with a target protein. We have trained a novel molecular-highway graph neural network architecture using the entirety of the BindingDB database to predict the probability of a drug ligand binding to a protein target. Our approach achieves a prodigious 98.3% accuracy with its predictions. Through this paper, we disseminate our source code and use the AI model to screen both public (ChEMBL, DrugBank) and proprietary databases. Compared to other AI-based methods, our approach outperforms the state-of-the-art on the following metrics - (i) number of molecules currently undergoing active clinical trials, (ii) number of antiviral drugs correctly identified, (iii) accuracy despite not needing active-site priors, and (iv) ability to screen more compounds in unit time.

EEG oscillatory correlates of expert meditators have been studied in the time-frequency domain. Machine Learning techniques are required to expand the understanding of oscillatory signatures. In this work, we propose a methodological pipeline to develop machine learning models for the classification between expert and nonexpert meditative state. We carried out this study utilizing the online repository consisting of EEG dataset of 24 meditators that categorized as 12 experts and 12 nonexperts meditators. The pipeline consists of four stages that include feature engineering, machine learning classifiers, feature selection, and visualization. We decomposed signals using five wavelet families consisting of Haar, Biorthogonal(1.3-6.8), Daubechies( orders 2-10), Coiflet(orders 1-5), and Symlet(2-8), followed by feature extraction using relative entropy and power. We classified the meditative state between expert and non-expert meditators employing twelve classifiers to build machine learning models. Wavelet coefficients d8 shows the maximum classification accuracy in all the wavelet families. Wavelet orders Bior3.5 and Coif3 produce the maximum classification performance with the detail coefficient d8 using relative power. We have successfully classified the meditative state between expert and non-expert with 100% accuracy using d5,d6,d7,d8,a8 coefficients. Multi-Layer Perceptron and Quadratic Discriminant Analysis attain the highest accuracy. We have figured out the most discriminating channels during classification and reported 20 channels involving frontal, central and parietal regions. We plot the high dimensional structure of data by utilizing two feature reduction techniques PCA and t-SNE.

The H2020 European research project Safe-Guarding Home IoT Environments with Personalised Real-time Risk Control (GHOST) aims to develop a cyber-security layer on IoT smart home installations. The proposed system analyses packet-level data flows for building patterns of communications between IoT devices and external entities. To ensure non-repudiation, integrity and authentication of the data captured, they are stored in a Blockchain, a distributed ledger network, as digitally-signed transactions. Since the data can potentially include sensitive user information, it is imperative to promote trust by informing users about the operating principles of the network as well as to request the acceptance of a consent form by them. This paper presents the design and implementation of a Forms of Consent application, a Distributed Application that interacts with a set of Smart Contracts deployed on a private Ethereum network. The application is being developed as part of the GHOST project.

The European research project GHOST challenges the traditional cyber security solutions for the Internet of Things (IoT) sector by exploiting novel technologies, such as blockchain, to provide resilience and integrity of decision making on the communication exchange in a smart home context. When it comes to novel cyber security solutions for extremely heterogeneous environments like IoT and smart homes, the key focus is typically given to the understanding of network activities and elimination of suspicious traffic. The GHOST project adds an extra dimension to this approach by integrating blockchain technology at its core decision mechanism. On a daily basis, each GHOST installation is encountering malicious behaviour and suspicious IoT communications, where easy information sharing with other installations, as well as decentralised decision making, are mandatory features for the efficient protection of the end-user. GHOST's Smart Contracts (SC) are designed to tackle in an easy, yet productive way, the reporting on suspicious IP addresses which the IoT devices in a smart home are trying to communicate with. Two variations of blacklisting smart contracts are presented in this paper, covering a diverse spectrum of possible attack vectors while closely following the Privacy by Design (PbD) principles. A reputation scoring scheme for malicious IPs reporting is integrated in the SC, uncovering the implementation details on the penalisation of existing entries in case of malicious behaviour of reporting devices.

The aim of the proposed work is to introduce a secure and interoperable Demand Response (DR) management platform that will assist Aggregators (or other relevant Stakeholders involved in DR business scenarios) in their decision making mechanisms over their portfolios of prosumers. This novel architecture incorporates multiple strategies and policies provided from energy market stakeholders, establishing a more modular and future-proof DR solution. By employing an innovative multi-agent decision making system and self-learning algorithms to enable aggregation, segmentation and coordination of several diverse clusters, consisting of supply and demand assets, a fully autonomous design will be delivered. This DR framework is further fortified in terms of data security by not only implementing cutting-edge blockchain infrastructure, but also by making use of Smart Contracts and Decentralized Applications (dApps) which will further secure and facilitate Aggregators-to-Prosumers transactions. The blockchain technologies will be combined with well-known open protocols (i.e. OpenADR) towards also supporting interoperability in terms of information exchange.

Prediction Markets are the markets designed and operated to mine and aggregate the information scattered among the traders. Recently, some researchers have started exploring the application of prediction markets in the information security domain. The information security prediction market will facilitate trading of contracts to hedge the financial impact of the risks associated with the underlying information security events, such as discovery of a vulnerability in a piece of software. However, prediction markets differ in their objectives and requirements, and therefore information security prediction markets need to be carefully engineered to meet the specific requirements. The contribution of this paper is the identification of a set of design requirements for an information security prediction market, and associated performance criteria. We present five categories of design requirements: Contracts, Trading Process, Participants and Incentives, Clearing House, and Market Management for the information security prediction market. Furthermore, we present six performance measures: Information Elicitation, Transparency, Efficiency, Transaction Cost, Liquidity, and Manipulation Resistance for the performance assessment of information security prediction market.