A Deep Learning Model for Correlation Analysis between Electroencephalography Signal and Speech Stimuli.

In recent years, the use of electroencephalography (EEG) has grown as a tool for diagnostic and brain function monitoring, being a simple and non-invasive method compared with other procedures like histological sampling. Typically, in order to extract functional brain responses from EEG signals, prolonged and repeated stimuli are needed because of the artifacts generated in recordings which adversely impact the stimulus-response analysis. To mitigate the artifact effect, correlation analysis (CA) methods are applied in the literature, where the predominant approaches focus on enhancing stimulus-response correlations through the use of linear analysis methods like canonical correlation analysis (CCA). This paper introduces a novel CA framework based on a neural network with a loss function specifically designed to maximize correlation between EEG and speech stimuli. Compared with other deep learning CA approaches (DCCAs) in the literature, this framework introduces a single multilayer perceptron (MLP) network instead of two networks for each stimulus. To validate the proposed approach, a comparison with linear CCA (LCCA) and DCCA was performed, using a dataset containing the EEG traces of subjects listening to speech stimuli. The experimental results show that the proposed method improves the overall Pearson correlation by 10.56% compared with the state-of-the-art DCCA method.

A CNN-based image detector for plant leaf diseases classification.

Identifying diseases from images of plant leaves is one of the most important research areas in precision agriculture. The aim of this paper is to propose an image detector embedding a resource constrained convolutional neural network (CNN) implemented in a low cost, low power platform, named OpenMV Cam H7 Plus, to perform a real-time classification of plant disease. The CNN network so obtained has been trained on two specific datasets for plant diseases detection, the ESCA-dataset and the PlantVillage-augmented dataset, and implemented in a low-power, low-cost Python programmable machine vision camera for real-time image acquisition and classification, equipped with a LCD display showing to the user the classification response in real-time. Experimental results show that this CNN-based image detector can be effectively implemented on the chosen constrained-resource system, achieving an accuracy of about 98.10%/95.24% with a very low memory cost (718.961 KB/735.727 KB) and inference time (122.969 ms/125.630 ms) tested on board for the ESCA and the PlantVillage-augmented datasets respectively, allowing the design of a portable embedded system for plant leaf diseases classification. Source files are available at https://doi.org/10.17605/OSF.IO/UCM8D.

EEG-Based Alzheimer's Disease Recognition Using Robust-PCA and LSTM Recurrent Neural Network.

The use of electroencephalography (EEG) has recently grown as a means to diagnose neurodegenerative pathologies such as Alzheimer's disease (AD). AD recognition can benefit from machine learning methods that, compared with traditional manual diagnosis methods, have higher reliability and improved recognition accuracy, being able to manage large amounts of data. Nevertheless, machine learning methods may exhibit lower accuracies when faced with incomplete, corrupted, or otherwise missing data, so it is important do develop robust pre-processing techniques do deal with incomplete data. The aim of this paper is to develop an automatic classification method that can still work well with EEG data affected by artifacts, as can arise during the collection with, e.g., a wireless system that can lose packets. We show that a recurrent neural network (RNN) can operate successfully even in the case of significantly corrupted data, when it is pre-filtered by the robust principal component analysis (RPCA) algorithm. RPCA was selected because of its stated ability to remove outliers from the signal. To demonstrate this idea, we first develop an RNN which operates on EEG data, properly processed through traditional PCA; then, we use corrupted data as input and process them with RPCA to filter outlier components, showing that even with data corruption causing up to 20% erasures, the RPCA was able to increase the detection accuracy by about 5% with respect to the baseline PCA.

Synthetic image dataset of shaft junctions inside wind turbines in presence or absence of oil leaks.

This paper presents a dataset of images generated via 3D graphics rendering. The dataset is composed by pictures of the junction between the high-speed shaft and the external bracket of the power generator inside a wind turbine cabin, in presence and absence of oil leaks. Oil leak occurrence is an anomaly that can verify in a zone of interest of the junction. Since the wind turbines industry is becoming more and more important, turbines maintenance is growing in importance accordingly. In this context a dataset, as we propose, can be used, for example, to design machine learning algorithms for predictive maintenance. The renderings have been produced, from various framings and various leaks shapes and colors, using the rendering engine Keyshot9. Subsequent preprocessing has been performed with Matlab, including images grayscale conversion and image binarization. Finally, data augmentation has been implemented in Python, and it can be easily extended/customized for realizing any further processing. The Matlab and Python source codes are also provided. To the authors' knowledge, there are no other public available datasets on this topic.

Energy and Performance Analysis of Lossless Compression Algorithms for Wireless EMG Sensors.

Electromyography (EMG) sensors produce a stream of data at rates that can easily saturate a low-energy wireless link such as Bluetooth Low Energy (BLE), especially if more than a few EMG channels are being transmitted simultaneously. Compressing data can thus be seen as a nice feature that could allow both longer battery life and more simultaneous channels at the same time. A lot of research has been done in lossy compression algorithms for EMG data, but being lossy, artifacts are inevitably introduced in the signal. Some artifacts can usually be tolerable for current applications. Nevertheless, for some research purposes and to enable future research on the collected data, that might need to exploit various and currently unforseen features that had been discarded by lossy algorithms, lossless compression of data may be very important, as it guarantees no extra artifacts are introduced on the digitized signal. The present paper aims at demonstrating the effectiveness of such approaches, investigating the performance of several algorithms and their implementation on a real EMG BLE wireless sensor node. It is demonstrated that the required bandwidth can be more than halved, even reduced to 1/4 on an average case, and if the complexity of the compressor is kept low, it also ensures significant power savings.

Dataset from PPG wireless sensor for activity monitoring.

We introduce a dataset to provide insights about the photoplethysmography (PPG) signal captured from the wrist in presence of motion artifacts and the accelerometer signal, simultaneously acquired from the same wrist. The data presented were collected by the electronics research team of the Department of Information Engineering, Polytechnic University of Marche, Ancona, Italy. This article describes data recorded from 7 subjects and includes 105 PPG signals (15 for each subject) and the corresponding 105 tri-axial accelerometer signals measured with a sampling frequency of 400 Hz. These data can be reused for testing machine learning algorithms for human activity recognition.

Dataset from spirometer and sEMG wireless sensor for diaphragmatic respiratory activity monitoring.

We introduce a dataset to provide insights into the relationship between the diaphragm surface electromyographic (sEMG) signal and the respiratory air flow. The data presented had been originally collected for a research project jointly developed by the Department of Information Engineering and the Department of Industrial Enginering and Mathematical Sciences, Polytechnic University of Marche, Ancona, Italy. This article describes data recorded from 8 subjects, and includes 8 air flow and 8 surface electromyographic (sEMG) signals for diaphragmatic respiratory activity monitoring, measured with a sampling frequency of 2 kHz.

A Comparative Study of Computational Methods for Compressed Sensing Reconstruction of EMG Signal.

Wearable devices offer a convenient means to monitor biosignals in real time at relatively low cost, and provide continuous monitoring without causing any discomfort. Among signals that contain critical information about human body status, electromyography (EMG) signal is particular useful in monitoring muscle functionality and activity during sport, fitness, or daily life. In particular surface electromyography (sEMG) has proven to be a suitable technique in several health monitoring applications, thanks to its non-invasiveness and ease to use. However, recording EMG signals from multiple channels yields a large amount of data that increases the power consumption of wireless transmission thus reducing the sensor lifetime. Compressed sensing (CS) is a promising data acquisition solution that takes advantage of the signal sparseness in a particular basis to significantly reduce the number of samples needed to reconstruct the signal. As a large variety of algorithms have been developed in recent years with this technique, it is of paramount importance to assess their performance in order to meet the stringent energy constraints imposed in the design of low-power wireless body area networks (WBANs) for sEMG monitoring. The aim of this paper is to present a comprehensive comparative study of computational methods for CS reconstruction of EMG signals, giving some useful guidelines in the design of efficient low-power WBANs. For this purpose, four of the most common reconstruction algorithms used in practical applications have been deeply analyzed and compared both in terms of accuracy and speed, and the sparseness of the signal has been estimated in three different bases. A wide range of experiments are performed on real-world EMG biosignals coming from two different datasets, giving rise to two different independent case studies.

Human activity monitoring system based on wearable sEMG and accelerometer wireless sensor nodes.

BACKGROUND: The human activity monitoring technology is one of the most important technologies for ambient assisted living, surveillance-based security, sport and fitness activities, healthcare of elderly people. The activity monitoring is performed in two steps: the acquisition of body signals and the classification of activities being performed. This paper presents a low-cost wearable wireless system specifically designed to acquire surface electromyography (sEMG) and accelerometer signals for monitoring the human activity when performing sport and fitness activities, as well as in healthcare applications. RESULTS: The proposed system consists of several ultralight wireless sensing nodes that are able to acquire, process and efficiently transmit the motion-related (biological and accelerometer) body signals to one or more base stations through a 2.4 GHz radio link using an ad-hoc communication protocol designed on top of the IEEE 802.15.4 physical layer. A user interface software for viewing, recording, and analysing the data was implemented on a control personal computer that is connected through a USB link to the base stations. To demonstrate the capability of the system of detecting the user's activity, data recorded from a few subjects were used to train and test an automatic classifier for recognizing the type of exercise being performed. The system was tested on four different exercises performed by three people, the automatic classifier achieved an overall accuracy of 85.7% combining the features extracted from acceleration and sEMG signals. CONCLUSIONS: A low cost wireless system for the acquisition of sEMG and accelerometer signals has been presented for healthcare and fitness applications. The system consists of wearable sensing nodes that wirelessly transmit the biological and accelerometer signals to one or more base stations. The signals so acquired will be combined and processed in order to detect, monitor and recognize human activities.

Classifier Level Fusion of Accelerometer and sEMG Signals for Automatic Fitness Activity Diarization.

The human activity diarization using wearable technologies is one of the most important supporting techniques for ambient assisted living, sport and fitness activities, healthcare of elderly people. The activity diarization is performed in two steps: the acquisition of body signals and the classification of activities being performed. This paper presents a technique for data fusion at classifier level of accelerometer and sEMG signals acquired by using a low-cost wearable wireless system for monitoring the human activity when performing sport and fitness activities, as well as in healthcare applications. To demonstrate the capability of the system of diarizing the user's activities, data recorded from a few subjects were used to train and test the automatic classifier for recognizing the type of exercise being performed.

An Investigation on the Accuracy of Truncated DKLT Representation for Speaker Identification With Short Sequences of Speech Frames.

Speaker identification plays a crucial role in biometric person identification as systems based on human speech are increasingly used for the recognition of people. Mel frequency cepstral coefficients (MFCCs) have been widely adopted for decades in speech processing to capture the speech-specific characteristics with a reduced dimensionality. However, although their ability to decorrelate the vocal source and the vocal tract filter make them suitable for speech recognition, they greatly mitigate the speaker variability, a specific characteristic that distinguishes different speakers. This paper presents a theoretical framework and an experimental evaluation showing that reducing the dimension of features by applying the discrete Karhunen-Loève transform (DKLT) to the log-spectrum of the speech signal guarantees better performance compared to conventional MFCC features. In particular with short sequences of speech frames, with typical duration of less than 2 s, the performance of truncated DKLT representation achieved for the identification of five speakers are always better than those achieved with the MFCCs for the experiments we performed. Additionally, the framework was tested on up to 100 TIMIT speakers with sequences of less than 3.5 s showing very good recognition capabilities.