Real-Time Myocardial Infarction Detection Approaches with a Microcontroller-Based Edge-AI Device.

Myocardial Infarction (MI), commonly known as heart attack, is a cardiac condition characterized by damage to a portion of the heart, specifically the myocardium, due to the disruption of blood flow. Given its recurring and often asymptomatic nature, there is the need for continuous monitoring using wearable devices. This paper proposes a single-microcontroller-based system designed for the automatic detection of MI based on the Edge Computing paradigm. Two solutions for MI detection are evaluated, based on Machine Learning (ML) and Deep Learning (DL) techniques. The developed algorithms are based on two different approaches currently available in the literature, and they are optimized for deployment on low-resource hardware. A feasibility assessment of their implementation on a single 32-bit microcontroller with an ARM Cortex-M4 core was examined, and a comparison in terms of accuracy, inference time, and memory usage was detailed. For ML techniques, significant data processing for feature extraction, coupled with a simpler Neural Network (NN) is involved. On the other hand, the second method, based on DL, employs a Spectrogram Analysis for feature extraction and a Convolutional Neural Network (CNN) with a longer inference time and higher memory utilization. Both methods employ the same low power hardware reaching an accuracy of 89.40% and 94.76%, respectively. The final prototype is an energy-efficient system capable of real-time detection of MI without the need to connect to remote servers or the cloud. All processing is performed at the edge, enabling NN inference on the same microcontroller.

Innovative Photonic Sensors for Safety and Security, Part II: Aerospace and Submarine Applications.

The employability of photonics technology in the modern era's highly demanding and sophisticated domain of aerospace and submarines has been an appealing challenge for the scientific communities. In this paper, we review our main results achieved so far on the use of optical fiber sensors for safety and security in innovative aerospace and submarine applications. In particular, recent results of in-field applications of optical fiber sensors in aircraft monitoring, from a weight and balance analysis to vehicle Structural Health Monitoring (SHM) and Landing Gear (LG) monitoring, are presented and discussed. Moreover, underwater fiber-optic hydrophones are presented from the design to marine application.

Innovative Photonic Sensors for Safety and Security, Part I: Fundamentals, Infrastructural and Ground Transportations.

Our group, involving researchers from different universities in Campania, Italy, has been working for the last twenty years in the field of photonic sensors for safety and security in healthcare, industrial and environment applications. This is the first in a series of three companion papers. In this paper, we introduce the main concepts of the technologies employed for the realization of our photonic sensors. Then, we review our main results concerning the innovative applications for infrastructural and transportation monitoring.

Innovative Photonic Sensors for Safety and Security, Part III: Environment, Agriculture and Soil Monitoring.

In order to complete this set of three companion papers, in this last, we focus our attention on environmental monitoring by taking advantage of photonic technologies. After reporting on some configurations useful for high precision agriculture, we explore the problems connected with soil water content measurement and landslide early warning. Then, we concentrate on a new generation of seismic sensors useful in both terrestrial and under water contests. Finally, we discuss a number of optical fiber sensors for use in radiation environments.

Liquid Resin Infusion Process Validation through Fiber Optic Sensor Technology.

In the proposed work, a fiber-optic-based sensor network was employed for the monitoring of the liquid resin infusion process. The item under test was a panel composed by a skin and four stringers, sensorized in such a way that both the temperature and the resin arrival could be monitored. The network was arranged with 18 Fiber Bragg Gratings (FBGs) working as temperature sensors and 22 fiber optic probes with a modified front-end in order to detect the resin presence. After an in-depth study to find a better solution to install the sensors without affecting the measurements, the system was investigated using a commercial Micron Optics at 0.5 Hz, with a passive split-box connected in order to be able to sense all the sensors simultaneously. The obtained results in terms of resin arrival detection at different locations and the relative temperature trend allowed us to validate an infusion process numerical model, giving us better understanding of what the actual resin flow was and the time needed to dry preform filling during the infusion process.

Multichannel Approach for Arrayed Waveguide Grating-Based FBG Interrogation Systems.

In this manuscript, an optically passive fiber Bragg grating (FBG) interrogation system able to perform high-frequency measurement is proposed. The idea is mainly based on the use of an arrayed waveguide grating (AWG) device which is used to discriminate the fiber optic sensor (FOS) wavelength encoded response under test in function of its output channels. As made clear by the theoretical model studied in the proposed manuscript, the Bragg wavelength shift can be detected as in linear dependence with the proposed interrogation function which changes with the voltage produced by two (or more) adjacent AWG output channels. To prove the feasibility of the system, some experimental analyses are conducted with a custom electrical module characterized by high-speed and low-noise operational amplifiers. As static measurements, three FBGs with different full width at half maximum (FWHM) have been monitored under wide-range wavelength variation; while, as dynamic measurement, one FBG, glued onto a metal plate, in order to sense the vibration at low and high frequency, was detected. The output signals have been processed by a digital acquisition (DAQ) board and a graphical user interface (GUI). The presented work highlights the characteristics of the proposed idea as competitor among the entire class of interrogation systems currently used. This is because here, the main device, that is the AWG, is passive and reliable, without the need to use modulation signals, or moving parts, that affect the speed of the system. In addition, the innovative multi-channel detection algorithm allows the use of any type of FOS without the need to have a perfectly match of spectra. Moreover, it is also characterized by a high dynamic range without loss of sensitivity.

The MUSHA underactuated hand for robot-aided minimally invasive surgery.

BACKGROUND: Keyhole surgery is characterized by loss of dexterity of surgeon's movements because of the limited workspace, nonintuitive motor skills of the surgical systems, and loss of tactile sensation that may lead to tissue damage and bad execution of the tasks. METHODS: In this paper, a three-fingered underactuated miniature tool for robot-aided laparoscopic surgery is presented. The design is conceived to realize a closed-hand configuration allowing the insertion of the tool into the abdominal cavity through the trocar in one step and to reach different grasping as well as pushing/holding configurations once in the cavity. RESULTS: Aiming to replicate human hand dexterity and versatility, different solutions for the kinematic structure of the hand are analyzed using quality indices to evaluate the manipulability and stability of the grasp. Furthermore, a first prototype of fingertip force sensor based on fiber Bragg grating (FBG) technology has been realized and tested. The design choices of the prototype are described and discussed with the aid of experiments. CONCLUSIONS: The whole concept and the need for such anthropomorphic tool are discussed with surgeons to highlight constraints and potentials in surgical tasks. The feedback by expert surgeons is used to provide specifications and improvements to the kinematics and mechanical design. The investigations of different designs allow identifying the optimal solution to improve grasping and manipulation capabilities. The tests on FBG sensors led to the conclusion that this technology guarantees good performance and can be a good solution for applications in surgical robotics.