A5062152110- Context-Aware Activity Recognition Systems
- Microwave Imaging and Scattering Analysis
- Antenna Design and Analysis
- Electromagnetic Simulation and Numerical Methods
- Wireless Body Area Networks
- Geophysical Methods and Applications
- Microwave Engineering and Waveguides
- Photonic and Optical Devices
- Photonic Crystals and Applications
- IoT and Edge/Fog Computing
- Antenna Design and Optimization
- Electromagnetic Scattering and Analysis
- Advanced SAR Imaging Techniques
- Non-Invasive Vital Sign Monitoring
- Indoor and Outdoor Localization Technologies
- Advanced Antenna and Metasurface Technologies
- Parkinson's Disease Mechanisms and Treatments
- Mobile Health and mHealth Applications
- Anomaly Detection Techniques and Applications
- IoT-based Smart Home Systems
- Bluetooth and Wireless Communication Technologies
- Energy Harvesting in Wireless Networks
- Time Series Analysis and Forecasting
- Dementia and Cognitive Impairment Research
- Human Pose and Action Recognition
University of Bristol
2016-2025
NHS Digital
2023
University of Bath
2021
Team (Italy)
2018
Sciences, Philosophie, Histoire
2018
Toshiba (United Kingdom)
2012-2017
Toshiba (Japan)
2013-2015
University of Oxford
2015
Cabot (United States)
2014
At Bristol
2005
In this contribution, an ultrawideband (UWB) microwave system for breast cancer detection is presented. The based on a novel hemispherical real-aperture antenna array, which employed in multi-static radar-based system. array consists of 16 UWB aperture-coupled stacked-patch antennas located section hemisphere. radar designed to be used with realistic three-dimensional (3D) phantoms, have been developed, as well real patients during initial clinical trials. Images are formed using two...
There's a widely known need to revise current forms of healthcare provision. Of particular interest are sensing systems in the home, which have been central several studies. This article presents an overview this rapidly growing body work, as well implications for machine learning, with aim uncovering gap between state art and broad needs services ambient assisted living. Most approaches address specific concerns, typically result solutions that aren't able support full-scale data analysis...
This paper presents an improved antenna array for radar-based breast cancer imaging. The improvement was achieved by increasing the number of antennas in to 31 elements, as well improving design itself. Using experimental setup, with homogeneous curved phantoms, we have demonstrated substantial imaging new array. system is also able detect 7 mm-diameter tumor phantoms any location within breast, even close 4 mm from skin layer. Additionally, shown good results low-contrast scenarios, where...
This letter presents, for the first time, experimental work on microwave breast cancer imaging using inhomogeneous phantoms. A recently designed 31-antenna array is used in experiments. The system operates full ultrawideband frequency range, between 3 and 10 GHz. To verify performance of our system, new phantoms with interior were developed. For three different presented this work, contrast spherical phantom tumors surrounding materials ranges from 5:1 to 1.6:1. Our results show that biggest...
A microwave imaging system has been developed as a clinical diagnostic tool operating in the 3- to 8-GHz region using multistatic data collection. total of 86 patients recruited from symptomatic breast care clinic were scanned with prototype design. The resultant three-dimensional images have compared "blind" available ultrasound and mammogram determine detection rate. Images show location strongest signal, this corresponded both older younger women, sensitivity [Formula: see text], which...
A wide-slot UWB antenna is presented for intended use in the detection scheme being developed at University of Bristol, based on principle synthetically focused radar using a fully populated static array. The antenna's measured and simulated, input radiation characteristics are compared to an existing, stacked patch that has been designed same purpose. results this study show excellent performance across required frequency range. Compared stacked-patch used our previous array, can be 3 times...
This paper presents a flexible 2.45-GHz wireless power harvesting wristband that generates net dc output from -24.3-dBm RF input. is the lowest reported system sensitivity for systems comprising rectenna and impedance-matching management. A complete has been implemented comprising: fabric antenna, rectifier on rigid substrate, contactless electrical connection between subsystems, electronics impedance matching. Various materials are electrically characterized at 2.45 GHz using two-line...
This paper serves as a survey and empirical evaluation of the state-of-the-art in activity recognition methods using accelerometers. The is particularly focused on long-term real-world settings. In these environments, data collection not trivial matter; thus, there are performance trade-offs between prediction accuracy, which sole system objective, keeping maintenance overhead at minimum levels. We examine research that has selection activities, features extracted from accelerometer data,...
This paper presented the clinical results of breast cancer detection using a radar-based UWB microwave system developed at University Bristol. Additionally, overview and some experimental laboratory are as well. For result shown in this contribution, we compare images obtained standard X-ray mammography system. The has apparently successfully detected tumor correct position, confirmed on image, although compression suffered by during makes precise positional determination impossible.
A novel wideband microwave radar imaging method is presented to detect regions of significant dielectric contrast within the breast. Clutter reduction paramount any algorithm, especially with clinical patient data where tissue composition breast inhomogeneous. Time-domain data-adaptive methods have been previously applied in a narrowband manner for when received signal spectral content was wideband. In this study, time-domain adaptive approach perform focusing across spectrum reduce clutter....
Cardiovascular diseases are the leading cause of death in U.K., motivating use long term wearable devices to monitor heart out-of-the-clinic settings. While a wide number rate measuring now available, they principally based upon photoplethysmography rather than electrocardiogram (ECG) and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">stand-alone</i> integrated with Internet-of-Things infrastructures which collect combine information from...
Multistatic radar apertures record scattering at a number of receivers when the target is illuminated by single transmitter, providing more information than its monostatic counterpart per transmission angle. This paper considers well-known problem detecting tumor targets within breast phantoms using multistatic radar. To accurately image potentially cancerous size breast, significant channels are required in order to adequately calibrate-out unwanted skin reflections, increase immunity...
Smart health home systems and assisted living architectures rely on severely energy-constrained sensing devices, such as wearable sensors, for the generation of data their reliable wireless communication to a central location. However, need recharging battery regularly constitutes maintenance burden that hinders long-term cost-effectiveness these systems, especially health-oriented applications target people in need, elderly or chronically ill. These generate raw is processed into knowledge...
Ambient Assisted Living (AAL) systems based on sensor technologies are seen as key enablers to an ageing society. However, most approaches in this space do not provide a truly generic ambient - one that is only capable of assisting people with diverse medical conditions, but can also recognise the habits healthy habitants, well those developing conditions. The recognition Activities Daily (ADL) understanding and provisioning appropriate efficient care. ADL particularly difficult achieve...
Quantitative assessment of the quality motion is increasingly in demand by clinicians healthcare and rehabilitation monitoring patients. We study compare performances different pose representations HMM models dynamics movement for online human motion. In a general sense, our framework builds model normal from skeleton-based samples healthy individuals. It encapsulates body using robust manifold representation first-order Markovian assumption. then assess deviations it via continuous measure....
Deep clustering has increasingly been demonstrating superiority over conventional shallow algorithms. algorithms usually combine representation learning with deep neural networks to achieve this performance, typically optimizing a and non-clustering loss. In such cases, an autoencoder is connected network, the final jointly learned by both network. Instead, we propose learn autoencoded embedding then search further for underlying manifold. For simplicity, cluster algorithm, rather than...
This paper presents a comprehensive dataset intended to evaluate passive Human Activity Recognition (HAR) and localization techniques with measurements obtained from synchronized Radio-Frequency (RF) devices vision-based sensors. The consists of RF data including Channel State Information (CSI) extracted WiFi Network Interface Card (NIC), Passive Radar (PWR) built upon Software Defined Radio (SDR) platform, Ultra-Wideband (UWB) signals acquired via commercial off-the-shelf hardware. It also...
A patch antenna is presented which has been designed to radiate frequencies in the range 4–9.5 GHz into human breast tissue. The shown by means of previously unpublished simulation and practical measurements possess a wide input bandwidth, radiation patterns that remain largely consistent over band interest good front-to-back ratio. Consideration also given antenna's ability pulse, this respect it found be suitable for proposed application.
This paper presents the development of a 60-element Ultra-WideBand (UWB) radar system for breast cancer detection and its use in clinical trials. The new operates frequency range 4-8GHz is an improvement team's previous designs both terms number measurements made (which increased by factor approximately 4) acquisition speed. 60-antenna has undergone extensive Clinical Trial Breast Care Centre at Frenchay Hospital, Bristol. rapid data improved accuracy images while also providing experience...
Healthcare professionals currently lack the means to gather unbiased and quantitative multi-modal data about long-term behaviors of patients in their home environments. SPHERE is a platform non-medical sensors for behavior monitoring residential environments that aims overcome this major limitation healthcare provision by using inherently cost-efficient scalable technologies Internet Things (IoT). One SPHERE's key tasks help bring next-generation low-power wireless networking sensing from...
A breast cancer detection technique using multi-static radar is proposed. For the first time, images are produced this technique, backscatter data from an anatomically realistic 2D MRI-derived FDTD model of breast. Successful a 2 mm-diameter tumour demonstrated, although clarity dependent on mitigating antenna mutual coupling and skin reflections.
Wearable technologies are valuable tools that can encourage people to monitor their own well-being and facilitate timely health interventions. In this paper, we present SPW-2; a low-profile versatile wearable sensor employs two ultra low power accelerometers an optional gyroscope. Designed