Sensor-enabled RFID system for monitoring arm activity: reliability and validity

Development and Testing of a Daily Activity Recognition System for Post-Stroke Rehabilitation

Those who survive the initial incidence of a stroke experience impacts on daily function. As a part of the rehabilitation process, it is essential for clinicians to monitor patients' health status and recovery progress accurately and consistently; however, little is known about how patients function in their own homes. Therefore, the goal of this study was to develop, train, and test an algorithm within an ambient, in-home depth sensor system that can classify and quantify home activities of individuals post-stroke. We developed the Daily Activity Recognition and Assessment System (DARAS). A daily action logger was implemented with a Foresite Healthcare depth sensor. Daily activity data were collected from seventeen post-stroke participants' homes over three months. Given the extensive amount of data, only a portion of the participants' data was used for this specific analysis. An ensemble network for activity recognition and temporal localization was developed to detect and segment the clinically relevant actions from the recorded data. The ensemble network, which learns rich spatial-temporal features from both depth and skeletal joint data, fuses the prediction outputs from a customized 3D convolutional-de-convolutional network, customized region convolutional 3D network, and a proposed region hierarchical co-occurrence network. The per-frame precision and per-action precision were 0.819 and 0.838, respectively, on the test set. The outcomes from the DARAS can help clinicians to provide more personalized rehabilitation plans that benefit patients.

An Assistive Technology Solution for User Activity Monitoring Exploiting Passive RFID

Population ageing is having a direct influence on serious health issues, including hampered mobility and physical decline. Good habits in performing physical activities, in addition to eating and drinking, are essential to improve the life quality of the elderly population. Technological solutions, aiming at increasing awareness or providing reminders to eat/drink regularly, can have a significant impact in this scenario. These solutions enable the possibility to constantly monitor deviations from users' normal behavior, thus allowing reminders to be provided to users/caregivers. In this context, this paper presents a radio-frequency identification (RFID) system to monitor user's habits, such as the use of food, beverages, and/or drugs. The device was optimized to fulfill specifications imposed by the addressed application. The approach could be extended for the monitoring of home appliances, environment exploitation, and activity rate. Advantages of the approach compared to other solutions, e.g., based on cameras, are related to the low level of invasiveness and flexibility of the adopted technology. A major contribution of this paper is related to the wide investigation of system behavior, which is aimed to define the optimal working conditions of the system, with regards to the power budget, user (antenna)-tag reading range, and the optimal inter-tag distance. To investigate the performance of the system in tag detection, experiments were performed in a scenario replicating a home environment. To achieve this aim, specificity and sensitivity indexes were computed to provide an objective evaluation of the system performance. For the case considered, if proper conditions are meet, a specificity value of 0.9 and a sensitivity value of 1 were estimated.

A fuzzy logic-based warning system for patients classification

Typically acute deterioration in sick people is preceded by subtle changes in the physiological parameters such as pulse and blood pressure. The Modified Early Warning Score is a scoring system developed to assist hospital staff in gauging these physiological changes and identifying patients in need of urgent medical care to avoid catastrophic deterioration. This work discusses the design and implementation of an equivalent warning system that utilizes fuzzy logic techniques to categorize patients' status. The system is implemented and tested in Rashid Centre for Diabetes and Research in UAE. Results are compared with those obtained using the Modified Early Warning System that is currently used in practice. We demonstrate that the implemented system provides reliable results that are in agreement with the current Modified Early Warning Score system, with the added benefit of a scoring scheme that provides a better insight into the status or medical condition of each patient.

A Review of Passive RFID Tag Antenna-Based Sensors and Systems for Structural Health Monitoring Applications

In recent few years, the antenna and sensor communities have witnessed a considerable integration of radio frequency identification (RFID) tag antennas and sensors because of the impetus provided by internet of things (IoT) and cyber-physical systems (CPS). Such types of sensor can find potential applications in structural health monitoring (SHM) because of their passive, wireless, simple, compact size, and multimodal nature, particular in large scale infrastructures during their lifecycle. The big data from these ubiquitous sensors are expected to generate a big impact for intelligent monitoring. A remarkable number of scientific papers demonstrate the possibility that objects can be remotely tracked and intelligently monitored for their physical/chemical/mechanical properties and environment conditions. Most of the work focuses on antenna design, and significant information has been generated to demonstrate feasibilities. Further information is needed to gain deep understanding of the passive RFID antenna sensor systems in order to make them reliable and practical. Nevertheless, this information is scattered over much literature. This paper is to comprehensively summarize and clearly highlight the challenges and state-of-the-art methods of passive RFID antenna sensors and systems in terms of sensing and communication from system point of view. Future trends are also discussed. The future research and development in UK are suggested as well.

An overview of systematic reviews on upper extremity outcome measures after stroke

BACKGROUND

Although use of standardized and scientifically sound outcome measures is highly encouraged in clinical practice and research, there is still no clear recommendation on which tools should be preferred for upper extremity assessment after stroke. As the aims, objectives and methodology of the existing reviews of the upper extremity outcome measures can vary, there is a need to bring together the evidence from existing multiple reviews. The purpose of this review was to provide an overview of evidence of the psychometric properties and clinical utility of upper extremity outcome measures for use in stroke, by systematically evaluating and summarizing findings from systematic reviews.

METHODS

A comprehensive systematic search was performed including systematic reviews from 2004 to February 2014. A methodological quality appraisal of the reviews was performed using the AMSTAR-tool.

RESULTS

From 13 included systematic reviews, 53 measures were identified of which 13 met the standardized criteria set for the psychometric properties. The strongest level of measurement quality and clinical utility was demonstrated for Fugl-Meyer Assessment, Action Research Arm Test, Box and Block Test, Chedoke Arm and Hand Activity Inventory, Wolf Motor Function Test and ABILHAND.

CONCLUSIONS

This overview of systematic reviews provides a comprehensive systematic synthesis of evidence on which outcome measures demonstrate a high level of measurement quality and clinical utility and which can be considered as most suitable for upper extremity assessment after stroke. This overview can provide a valuable resource to assist clinicians, researchers and policy makers in selection of appropriate outcome measures.

Assessment of safety and interference issues of radio frequency identification devices in 0.3 Tesla magnetic resonance imaging and computed tomography

The objective of this study was to evaluate two issues regarding magnetic resonance imaging (MRI) including device functionality and image artifacts for the presence of radio frequency identification devices (RFID) in association with 0.3 Tesla at 12.7 MHz MRI and computed tomography (CT) scanning. Fifteen samples of RFID tags with two different sizes (wristband and ID card types) were tested. The tags were exposed to several MR-imaging conditions during MRI examination and X-rays of CT scan. Throughout the test, the tags were oriented in three different directions (axial, coronal, and sagittal) relative to MRI system in order to cover all possible situations with respect to the patient undergoing MRI and CT scanning, wearing a RFID tag on wrist. We observed that the tags did not sustain physical damage with their functionality remaining unaffected even after MRI and CT scanning, and there was no alternation in previously stored data as well. In addition, no evidence of either signal loss or artifact was seen in the acquired MR and CT images. Therefore, we can conclude that the use of this passive RFID tag is safe for a patient undergoing MRI at 0.3 T/12.7 MHz and CT Scanning.