A Digital Architecture for the Real-Time Tracking of Wearing off Phenomenon in Parkinson's Disease.

Levodopa administration is currently the most common treatment to alleviate Parkinson's Disease (PD) symptoms. Nevertheless, prolonged use of Levodopa leads to a wearing-off (WO) phenomenon, causing symptoms to reappear. To build a personalized treatment plan aiming to manage PD and its symptoms effectively, there is a need for a technological system able to continuously and objectively assess the WO phenomenon during daily life. In this context, this paper proposes a WO tracker able to exploit neuromuscular data acquired by a dedicated wireless sensor network to discriminate between a Levodopa benefit phase and the reappearance of symptoms. The proposed architecture has been implemented on a heterogeneous computing platform, that statistically analyzes neural and muscular features to identify the best set of features to train the classifier model. Eight models among shallow and deep learning approaches are analyzed in terms of performance, timing and complexity metrics to identify the best inference engine. Experimental results on five subjects experiencing WO, showed that, in the best case, the proposed WO tracker can achieve an accuracy of ~84%, providing the inference in less than 41 ms. It is possible by employing a simple fully-connected neural network with 1 hidden layer and 32 units.

An Embedded Framework for Fully Autonomous Object Manipulation in Robotic-Empowered Assisted Living.

Most of the humanoid social robots currently diffused are designed only for verbal and animated interactions with users, and despite being equipped with two upper arms for interactive animation, they lack object manipulation capabilities. In this paper, we propose the MONOCULAR (eMbeddable autONomous ObjeCt manipULAtion Routines) framework, which implements a set of routines to add manipulation functionalities to social robots by exploiting the functional data fusion of two RGB cameras and a 3D depth sensor placed in the head frame. The framework is designed to: (i) localize specific objects to be manipulated via RGB cameras; (ii) define the characteristics of the shelf on which they are placed; and (iii) autonomously adapt approach and manipulation routines to avoid collisions and maximize grabbing accuracy. To localize the item on the shelf, MONOCULAR exploits an embeddable version of the You Only Look Once (YOLO) object detector. The RGB camera outcomes are also used to estimate the height of the shelf using an edge-detecting algorithm. Based on the item's position and the estimated shelf height, MONOCULAR is designed to select between two possible routines that dynamically optimize the approach and object manipulation parameters according to the real-time analysis of RGB and 3D sensor frames. These two routines are optimized for a central or lateral approach to objects on a shelf. The MONOCULAR procedures are designed to be fully automatic, intrinsically protecting sensitive users' data and stored home or hospital maps. MONOCULAR was optimized for Pepper by SoftBank Robotics. To characterize the proposed system, a case study in which Pepper is used as a drug delivery operator is proposed. The case study is divided into: (i) pharmaceutical package search; (ii) object approach and manipulation; and (iii) delivery operations. Experimental data showed that object manipulation routines for laterally placed objects achieves a best grabbing success rate of 96%, while the routine for centrally placed objects can reach 97% for a wide range of different shelf heights. Finally, a proof of concept is proposed here to demonstrate the applicability of the MONOCULAR framework in a real-life scenario.

A Cybersecure P300-Based Brain-to-Computer Interface against Noise-Based and Fake P300 Cyberattacks.

In a progressively interconnected world where the Internet of Things (IoT), ubiquitous computing, and artificial intelligence are leading to groundbreaking technology, cybersecurity remains an underdeveloped aspect. This is particularly alarming for brain-to-computer interfaces (BCIs), where hackers can threaten the user's physical and psychological safety. In fact, standard algorithms currently employed in BCI systems are inadequate to deal with cyberattacks. In this paper, we propose a solution to improve the cybersecurity of BCI systems. As a case study, we focus on P300-based BCI systems using support vector machine (SVM) algorithms and EEG data. First, we verified that SVM algorithms are incapable of identifying hacking by simulating a set of cyberattacks using fake P300 signals and noise-based attacks. This was achieved by comparing the performance of several models when validated using real and hacked P300 datasets. Then, we implemented our solution to improve the cybersecurity of the system. The proposed solution is based on an EEG channel mixing approach to identify anomalies in the transmission channel due to hacking. Our study demonstrates that the proposed architecture can successfully identify 99.996% of simulated cyberattacks, implementing a dedicated counteraction that preserves most of BCI functions.

A Single-Trial P300 Detector Based on Symbolized EEG and Autoencoded-(1D)CNN to Improve ITR Performance in BCIs.

In this paper, we propose a breakthrough single-trial P300 detector that maximizes the information translate rate (ITR) of the brain-computer interface (BCI), keeping high recognition accuracy performance. The architecture, designed to improve the portability of the algorithm, demonstrated full implementability on a dedicated embedded platform. The proposed P300 detector is based on the combination of a novel pre-processing stage based on the EEG signals symbolization and an autoencoded convolutional neural network (CNN). The proposed system acquires data from only six EEG channels; thus, it treats them with a low-complexity preprocessing stage including baseline correction, windsorizing and symbolization. The symbolized EEG signals are then sent to an autoencoder model to emphasize those temporal features that can be meaningful for the following CNN stage. This latter consists of a seven-layer CNN, including a 1D convolutional layer and three dense ones. Two datasets have been analyzed to assess the algorithm performance: one from a P300 speller application in BCI competition III data and one from self-collected data during a fluid prototype car driving experiment. Experimental results on the P300 speller dataset showed that the proposed method achieves an average ITR (on two subjects) of 16.83 bits/min, outperforming by +5.75 bits/min the state-of-the-art for this parameter. Jointly with the speed increase, the recognition performance returned disruptive results in terms of the harmonic mean of precision and recall (F1-Score), which achieve 51.78 ± 6.24%. The same method used in the prototype car driving led to an ITR of ~33 bit/min with an F1-Score of 70.00% in a single-trial P300 detection context, allowing fluid usage of the BCI for driving purposes. The realized network has been validated on an STM32L4 microcontroller target, for complexity and implementation assessment. The implementation showed an overall resource occupation of 5.57% of the total available ROM, ~3% of the available RAM, requiring less than 3.5 ms to provide the classification outcome.

High-Specificity Digital Architecture for Real-Time Recognition of Loss of Balance Inducing Fall.

Falls are a significant cause of loss of independence, disability and reduced quality of life in people with Parkinson's disease (PD). Intervening quickly and accurately on the postural instability could strongly reduce the consequences of falls. In this context, the paper proposes and validates a novel architecture for the reliable recognition of losses of balance situations. The proposed system addresses some challenges related to the daily life applicability of near-fall recognition systems: the high specificity and system robustness against the Activities of Daily Life (ADL). In this respect, the proposed algorithm has been tested on five different tasks: walking steps, sudden curves, chair transfers via the timed up and go (TUG) test, balance-challenging obstacle avoidance and slip-induced loss of balance. The system analyzes data from wireless acquisition devices that capture electroencephalography (EEG) and electromyography (EMG) signals. The collected data are sent to two main units: the muscular unit and the cortical one. The first realizes a binary ON/OFF pattern from muscular activity (10 EMGs) and triggers the cortical unit. This latter unit evaluates the rate of variation in the EEG power spectrum density (PSD), considering five bands of interest. The neuromuscular features are then sent to a logical network for the final classification, which distinguishes among falls and ADL. In this preliminary study, we tested the proposed model on 9 healthy subjects (aged 26.3 ± 2.4 years), even if the study on PD patients is under investigation. Experimental validation on healthy subjects showed that the system reacts in 370.62 ± 60.85 ms with a sensitivity of 93.33 ± 5.16%. During the ADL tests the system showed a specificity of 98.91 ± 0.44% in steady walking steps recognition, 99.61 ± 0.66% in sudden curves detection, 98.95 ± 1.27% in contractions related to TUG tests and 98.42 ± 0.90% in the obstacle avoidance protocol.

Field Programmable Gate Array-Embedded Platform for Dynamic Muscle Fiber Conduction Velocity Monitoring.

This paper proposes a novel architecture of a wearable Field Programmable Gate Array (FPGA)-based platform to dynamically monitor Muscle Fiber Conduction Velocity (MFCV). The system uses a set of wireless sensors for the detection of muscular activation: four surface electromyography electrodes (EMGs) and two footswitches. The beginning of movement (trigger) is set by sensors (footswitches) detecting the feet position. The MFCV value extraction exploits an iterative algorithm, which compares two 1-bit digitized EMG signals. The EMG electrode positioning is ensured by a dedicated procedure. The architecture is implemented on FPGA board (Altera Cyclone V), which manages an external Bluetooth module for data transmission. The time spent for data elaboration is 63.5 ms ± 0.25 ms, matching real-time requirements. The FPGA-based MFCV estimator has been validated during regular walking and in the fatigue monitoring context. Six healthy subjects contributed to experimental validation. In the gait analysis, the subjects showed MFCV evaluation of about 7.6 m/s ± 0.36 m/s, i.e., <0.1 m/s, a typical value for healthy subjects. Furthermore, in agreement with current research methods in the field, in a fatigue evaluation context, the extracted data showed an MFCV descending trend with the increment of the muscular effort time (Rested: MFCV = 8.51 m/s; Tired: 4.60 m/s).

Spatio-Temporal Optimization of Perishable Goods' Shelf Life by a Pro-Active WSN-Based Architecture.

The waste in the perishable goods supply-chain has prompted many global organizations (e.g., FAO and WHO), to develop the Hazard Analysis and Critical Control Points (HACCP) protocol that ensures a high degree of food quality, minimizing the losses in all the stages of the farm-to-fork chain. It has been proven that good warehouse management practices improve the average life of perishable goods. The advances in wireless sensors network (WSN) technology offers the possibility of a "smart" storage organization. In this paper, a low cost reprogrammable WSN-based architecture for functional warehouse management is proposed. The management is based on the continuous monitoring of environmental parameters (i.e., temperature, light exposure and relative humidity), and on their combination to extract a spatial real-time prediction of the product shelf life. For each product, the quality decay is computed by using a 1st order kinetic Arrhenius model to the whole storage site area. It strives to identify, in a way compatible with the other products' shelf lives, the position within the warehouse that maximizes the food expiration date. The shelf life computing and the "first-expired first-out" logistic problem are entrusted to a Raspberry Pi-based central unit, which manages a set of automated pallet transporters for the displacement of products, according to the computed shelf lives. The management unit supports several commercial light/temperature/humidity sensor solutions, implementing ZigBee, Bluetooth and HTTP-request interfaces. A proof of concept of the presented pro-active WSN-based architecture is also shown. Comparing the proposed monitoring system for the storage of e.g., agricultural products, with a typical one, the experimental results show an improvement of the expected expiration date of about 1.2 ± 0.5 days, for each pallet, when placed in a non-refrigerated environment. In order to stress the versatility of the WSN solution, a section is dedicated to the implemented system user interfaces that highlight detecting critical situations and allow timely automatic or human interventions, minimizing the latter.