scholarly journals From Bits of Data to Bits of Knowledge—An On-Board Classification Framework for Wearable Sensing Systems

Sensors ◽  
2020 ◽  
Vol 20 (6) ◽  
pp. 1655
Author(s):  
Pawel Zalewski ◽  
Letizia Marchegiani ◽  
Atis Elsts ◽  
Robert Piechocki ◽  
Ian Craddock ◽  
...  
Keyword(s):  
Healthcare Provision ◽  
Raw Data ◽  
Sensing Systems ◽  
High Classification Accuracy ◽  
Classification Framework ◽  
Wearable Systems ◽  
Order Of Magnitude ◽  
Promising Solution ◽  
Energy Constrained ◽  
Learning Frameworks

Wearable systems constitute a promising solution to the emerging challenges of healthcare provision, feeding machine learning frameworks with necessary data. In practice, however, raw data collection is expensive in terms of energy, and therefore imposes a significant maintenance burden to the user, which in turn results in poor user experience, as well as significant data loss due to improper battery maintenance. In this paper, we propose a framework for on-board activity classification targeting severely energy-constrained wearable systems. The proposed framework leverages embedded classifiers to activate power-hungry sensing elements only when they are useful, and to distil the raw data into knowledge that is eventually transmitted over the air. We implement the proposed framework on a prototype wearable system and demonstrate that it can decrease the energy requirements by one order of magnitude, yielding high classification accuracy that is reduced by approximately 5%, as compared to a cloud-based reference system.

scholarly journals Recent Advances in Wearable Devices for Non-Invasive Sensing

2021 ◽  
Vol 11 (3) ◽  
pp. 1235
Author(s):  
Su Min Yun ◽  
Moohyun Kim ◽  
Yong Won Kwon ◽  
Hyobeom Kim ◽  
Mi Jung Kim ◽  
...  
Keyword(s):  
Health Monitoring ◽  
Data Transmission ◽  
Wearable Sensors ◽  
Wearable Devices ◽  
Wireless Power ◽  
Non Invasive ◽  
Sensing Systems ◽  
Recent Advances ◽  
Wearable Systems ◽  
Precise Diagnosis

The development of wearable sensors is aimed at enabling continuous real-time health monitoring, which leads to timely and precise diagnosis anytime and anywhere. Unlike conventional wearable sensors that are somewhat bulky, rigid, and planar, research for next-generation wearable sensors has been focused on establishing fully-wearable systems. To attain such excellent wearability while providing accurate and reliable measurements, fabrication strategies should include (1) proper choices of materials and structural designs, (2) constructing efficient wireless power and data transmission systems, and (3) developing highly-integrated sensing systems. Herein, we discuss recent advances in wearable devices for non-invasive sensing, with focuses on materials design, nano/microfabrication, sensors, wireless technologies, and the integration of those.

scholarly journals The Effect of Measurement Trends in Belt Breathing Sensors

2021 ◽  
Vol 6 (1) ◽  
pp. 84
Author(s):  
Erik Vanegas ◽  
Raúl Igual ◽  
Inmaculada Plaza
Keyword(s):  
Experimental Study ◽  
Wearable Sensors ◽  
Sensor Performance ◽  
Sensing System ◽  
Post Process ◽  
Sensing Systems ◽  
Piezoresistive Sensor ◽  
Wearable Systems ◽  
Respiration Signal

Sensors for respiratory monitoring can be classified into wearable and non-wearable systems. Wearable sensors can be worn in several positions, the chest being one of the most effective. In this paper, we have studied the performance of a new piezoresistive breathing sensing system to be worn on the chest with a belt. One of the main problems of belt-attached sensing systems is that they present trends in measurements due to subject movements or differences in subject constitution. These trends affect sensor performance. To mitigate them, it is possible to post-process the data to remove trends in measurements, but relevant data from the respiration signal may be lost. In this study, two different detrending methods are applied to respiration signals. After conducting an experimental study with 21 subjects who breathed in different positions with a chest piezoresistive sensor attached to a belt, detrending method 2 proved to be better at improving the quality of respiration signals.

A method for constructing a network traffic classifier based on deep learning

2021 ◽  
Author(s):  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Network Traffic ◽  
Short Term Memory ◽  
Short Term ◽  
Raw Data ◽  
Term Memory ◽  
High Classification Accuracy ◽  
Data Flows ◽  
Long Short Term Memory

An overview and analysis of methods for constructing of network traffic classifiers is given and the advantage of deep learning methods is shown. Based on a comparative analysis, the methods of deep learning with a teacher is selected. A method based on the use of a multilayer neural network of long short-term memory (LSTM) is considered. The structure of a deep network, at the input of which raw data flows fed, divided into sessions is created. Based on the selected classes of applications, it is experimentally proven that the developed neural network of long short-term memory, to the input of which raw data is supplied, allows to obtain a high classification accuracy. Keywords network traffic; deep learning; long short-term memory neural network; raw data

scholarly journals Detection of the Complete ECG Waveform with Woven Textile Electrodes

Biosensors ◽  
2021 ◽  
Vol 11 (9) ◽  
pp. 331
Author(s):  
Katya Arquilla ◽  
Laura Devendorf ◽  
Andrea K. Webb ◽  
Allison P. Anderson
Keyword(s):  
Psychological State ◽  
Monitoring Systems ◽  
Single Individual ◽  
Full Waveform ◽  
Sensing Systems ◽  
Textile Electrodes ◽  
Wearable Systems ◽  
Human Participants ◽  
Electrocardiogram Ecg ◽  
Woven Textile

Wearable physiological monitoring systems are becoming increasingly prevalent in the push toward autonomous health monitoring and offer new modalities for playful and purposeful interaction within human computer interaction (HCI). Sensing systems that can be integrated into garments and, therefore, daily activities offer promising pathways toward ubiquitous integration. The electrocardiogram (ECG) signal is commonly monitored in healthcare and is increasingly utilized as a method of determining emotional and psychological state; however, the complete ECG waveform with the P, Q, R, S, and T peaks is not commonly used, due to the challenges associated with collecting the full waveform with wearable systems. We present woven textile electrodes as an option for garment-integrated ECG monitoring systems that are capable of capturing the complete ECG waveform. In this work, we present the changes in the peak detection performance caused by different sizes, patterns, and thread types with data from 10 human participants. These testing results provide empirically-derived guidelines for future woven textile electrodes, present a path forward for assessing design decisions, and highlight the importance of testing novel wearable sensor systems with more than a single individual.

scholarly journals Wearable systems for shoulder kinematics assessment: a systematic review

2019 ◽  
Vol 20 (1) ◽  
Author(s):  
Arianna Carnevale ◽  
Umile Giuseppe Longo ◽  
Emiliano Schena ◽  
Carlo Massaroni ◽  
Daniela Lo Presti ◽  
...  
Keyword(s):  
Inertial Sensors ◽  
Wearable Sensors ◽  
Clinical Information ◽  
Clinical Settings ◽  
Novel Technologies ◽  
Motor Abilities ◽  
Wearable Systems ◽  
Comprehensive Search ◽  
Promising Solution ◽  
Shoulder Kinematics

Abstract Background Wearable sensors are acquiring more and more influence in diagnostic and rehabilitation field to assess motor abilities of people with neurological or musculoskeletal impairments. The aim of this systematic literature review is to analyze the wearable systems for monitoring shoulder kinematics and their applicability in clinical settings and rehabilitation. Methods A comprehensive search of PubMed, Medline, Google Scholar and IEEE Xplore was performed and results were included up to July 2019. All studies concerning wearable sensors to assess shoulder kinematics were retrieved. Results Seventy-three studies were included because they have fulfilled the inclusion criteria. The results showed that magneto and/or inertial sensors are the most used. Wearable sensors measuring upper limb and/or shoulder kinematics have been proposed to be applied in patients with different pathological conditions such as stroke, multiple sclerosis, osteoarthritis, rotator cuff tear. Sensors placement and method of attachment were broadly heterogeneous among the examined studies. Conclusions Wearable systems are a promising solution to provide quantitative and meaningful clinical information about progress in a rehabilitation pathway and to extrapolate meaningful parameters in the diagnosis of shoulder pathologies. There is a strong need for development of this novel technologies which undeniably serves in shoulder evaluation and therapy.

scholarly journals Polyhedral-Based Compilation Framework for In-Memory Neural Network Accelerators

2022 ◽  
Vol 18 (1) ◽  
pp. 1-23
Author(s):  
Jianhui Han ◽  
Xiang Fei ◽  
Zhaolin Li ◽  
Youhui Zhang
Keyword(s):  
Neural Network ◽  
Case Studies ◽  
Memory Architecture ◽  
Polyhedral Model ◽  
Order Of Magnitude ◽  
Promising Solution ◽  
Memory Bottleneck ◽  
High Level ◽  
Programming Interfaces ◽  
Compilation Framework

Memristor-based processing-in-memory architecture is a promising solution to the memory bottleneck in the neural network ( NN ) processing. A major challenge for the programmability of such architectures is the automatic compilation of high-level NN workloads, from various operators to the memristor-based hardware that may provide programming interfaces with different granularities. This article proposes a source-to-source compilation framework for such memristor-based NN accelerators, which can conduct automatic detection and mapping of multiple NN operators based on the flexible and rich representation capability of the polyhedral model. In contrast to previous studies, it implements support for pipeline generation to exploit the parallelism in the NN loads to leverage hardware resources for higher efficiency. The evaluation based on synthetic kernels and NN benchmarks demonstrates that the proposed framework can reliably detect and map the target operators. Case studies on typical memristor-based architectures also show its generality over various architectural designs. The evaluation further demonstrates that compared with existing polyhedral-based compilation frameworks that do not support the pipelined execution, the performance can upgrade by an order of magnitude with the pipelined execution, which emphasizes the necessity of our improvement.

A controller for safe, convenient operation of LaB6 emitters on electron-beam instruments

1983 ◽  
Vol 41 ◽  
pp. 408-409
Author(s):  
W. J. Abramson ◽  
H. W. Estry ◽  
L. F. Allard
Keyword(s):  
Electron Beam ◽  
Control System ◽  
Thermal Shock ◽  
Appropriate Care ◽  
Electron Guns ◽  
Tungsten Filaments ◽  
Order Of Magnitude ◽  
Main Components

LaB6 emitters are becoming increasingly popular as direct replacements for tungsten filaments in the electron guns of modern electron-beam instruments. These emitters offer order of magnitude increases in beam brightness, and, with appropriate care in operation, a corresponding increase in source lifetime. They are, however, an order of magnitude more expensive, and may be easily damaged (by improper vacuum conditions and thermal shock) during saturation/desaturation operations. These operations typically require several minutes of an operator's attention, which becomes tedious and subject to error, particularly since the emitter must be cooled during sample exchanges to minimize damage from random vacuum excursions. We have designed a control system for LaBg emitters which relieves the operator of the necessity for manually controlling the emitter power, minimizes the danger of accidental improper operation, and makes the use of these emitters routine on multi-user instruments.Figure 1 is a block schematic of the main components of the control system, and Figure 2 shows the control box.

Novel magneto-optical recording materials: Bi-substituted garnet films

1991 ◽  
Vol 49 ◽  
pp. 776-777
Author(s):  
Takao Suzuki ◽  
Hossein Nuri
Keyword(s):  
Grain Size ◽  
Amorphous Film ◽  
Nitrogen Atmosphere ◽  
Optical Recording ◽  
Garnet Film ◽  
Media Noise ◽  
Garnet Films ◽  
Order Of Magnitude ◽  
A New Technique ◽  
A Current

For future high density magneto-optical recording materials, a Bi-substituted garnet film ((BiDy)3(FeGa)5O12) is an attractive candidate since it has strong magneto-optic effect at short wavelengths less than 600 nm. The signal in read back performance at 500 nm using a garnet film can be an order of magnitude higher than a current rare earth-transition metal amorphous film. However, the granularity and surface roughness of such crystalline garnet films are the key to control for minimizing media noise.We have demonstrated a new technique to fabricate a garnet film which has much smaller grain size and smoother surfaces than those annealed in a conventional oven. This method employs a high ramp-up rate annealing (Γ = 50 ~ 100 C/s) in nitrogen atmosphere. Fig.1 shows a typical microstruture of a Bi-susbtituted garnet film deposited by r.f. sputtering and then subsequently crystallized by a rapid thermal annealing technique at Γ = 50 C/s at 650 °C for 2 min. The structure is a single phase of garnet, and a grain size is about 300A.

Atomic-level imaging of the surface structure of Ag2O

1984 ◽  
Vol 42 ◽  
pp. 656-657
Author(s):  
William Krakow
Keyword(s):  
High Vacuum ◽  
Atomic Level ◽  
Ultra High Vacuum ◽  
Research Groups ◽  
Resolution Electron ◽  
Surface Reconstructions ◽  
Order Of Magnitude ◽  
High Resolution Electron Microscope ◽  
Saturated Structure ◽  
Transmission And Reflection

In recent years electron microscopy has been used to image surfaces in both the transmission and reflection modes by many research groups. Some of this work has been performed under ultra high vacuum conditions (UHV) and apparent surface reconstructions observed. The level of resolution generally has been at least an order of magnitude worse than is necessary to visualize atoms directly and therefore the detailed atomic rearrangements of the surface are not known. The present author has achieved atomic level resolution under normal vacuum conditions of various Au surfaces. Unfortunately these samples were exposed to atmosphere and could not be cleaned in a standard high resolution electron microscope. The result obtained surfaces which were impurity stabilized and reveal the bulk lattice (1x1) type surface structures also encountered by other surface physics techniques under impure or overlayer contaminant conditions. It was therefore decided to study a system where exposure to air was unimportant by using a oxygen saturated structure, Ag2O, and seeking to find surface reconstructions, which will now be described.

Near-field scanning optical microscopy

1986 ◽  
Vol 44 ◽  
pp. 642-643
Author(s):  
E. Betzig ◽  
A. Harootunian ◽  
M. Isaacson ◽  
A. Lewis
Keyword(s):  
Near Field ◽  
Optical Microscope ◽  
Visible Radiation ◽  
Field Methods ◽  
Optical Elements ◽  
Optical Region ◽  
Order Of Magnitude ◽  
Nondestructive Imaging ◽  
Scanning Optical Microscopy ◽  
Near Field Scanning

In general, conventional methods of optical imaging are limited in spatial resolution by either the wavelength of the radiation used or by the aberrations of the optical elements. This is true whether one uses a scanning probe or a fixed beam method. The reason for the wavelength limit of resolution is due to the far field methods of producing or detecting the radiation. If one resorts to restricting our probes to the near field optical region, then the possibility exists of obtaining spatial resolutions more than an order of magnitude smaller than the optical wavelength of the radiation used. In this paper, we will describe the principles underlying such "near field" imaging and present some preliminary results from a near field scanning optical microscope (NS0M) that uses visible radiation and is capable of resolutions comparable to an SEM. The advantage of such a technique is the possibility of completely nondestructive imaging in air at spatial resolutions of about 50nm.

Sign in / Sign up

Export Citation Format

Share Document