No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Medication intake adherence with real time activity recognition on IoT
... Pill bottle with multisensor system Boonnuddar & Wuttidittachotti, 2017;Chavez et al., 2020;Fathillah & Chellappan, 2022;Karagiannis & Nikita, 2020;Ma et al., 2018;Serdaroglu et al., 2015;So et al., 2021;Srinivas et al., 2018;Toh et al., 2016) A pill bottle with sensors to track medication adherence. ...
... In contrast to traditional methods such as self-reporting, proximity sensing systems provide more accurate data on medication adherence by actively detecting the opening of medication containers. Notable research utilising a timepiece, a mobile application, and a web server monitored medication adherence among geriatric patients (Serdaroglu et al., 2015). The smartwatch, a state-of-the-art wearable device, was ingeniously designed to detect the exact instant the medication container was opened, capturing vital information regarding medication usage. ...
The dynamic field of the Internet of Things (IoT) is constantly increasing, providing a plethora of potential integration across various sectors, most notably healthcare. The IoT represents a significant technological leap in healthcare management systems, coinciding with the rising preference for personalized, proactive, cost-effective treatment techniques. This review aimed to thoroughly assess the existing literature through a systematic review and bibliometric analysis, identifying untapped research routes and possible domains for further exploration. The overarching goal was to provide healthcare professionals with significant insights into the impact of IoT technology on Patient Medication Adherence (PMA) and related outcomes. An extensive review of 314 scientific articles on the deployment of IoT within pharmaceutical care services revealed a rising trend in publication volume, with a significant increase in recent years. Pertinently, from the 33 publications finally selected, substantial data support the potential of the IoT to improve PMA, particularly among senior patients with chronic conditions. This paper also comments on various regularly implemented IoT-based systems, noting their unique benefits and limitations. In conclusion, the critical relevance of PMA is highlighted, arguing for its emphasis in future discussions. Furthermore, the need for additional research endeavors is proposed to face and overcome existing constraints and establish the long-term effectiveness of IoT technologies in maximizing patient outcomes.
... Mo et al. [84] have used a smartphone to collect data from body worn sensors with energy harvesting capabilities, and to send it over to a distant server for processing and activity recognition. Serdaroglu et al. [85] have used a wrist worn watch with a built-in accelerometer to collect data in order to monitor patients' daily medication intake. The data is sent wirelessly from the watch to a gateway, connected to a computer via USB, before being sent over to a web server. ...
... This approach is generally used with environmental sensors, as using wires between every single sensor node and the gateway node would be impractical in a smart home. The gateway node is directly plugged into the central station, which could process the data locally or send it over to a cloud server [85]. In the last presented architecture, all of the sensor nodes are sending data wirelessly to a smartphone, acting as a gateway node, sending data to the local station. ...
In this paper, we review the evolution of the field of sensor-based activity recognition from early offline implementations to more recent, real-time distributed solutions. This review aims to give a wide overview of the state of the field, and it is aimed at anyone interested in exploring IoT oriented solutions for activity recognition, and wanting to know more about the various challenges, existing solutions, and axes of optimizations to carry out wireless sensor based real-time distributed activity recognition for healthcare applications.
... Furthermore, the combination of IoT with smartphones have an active influence on individuals, since its use (combined with social networks) influence the education, self-management and compliance of patients to a therapy and the use of substances (Serdaroglu et al., 2015). Additionally, the combination with cellular telephony allows a fast communication between health professionals and individuals, helping to deliver fast feedback about the subjective condition during a treatment or exposition to toxic substances. ...
... Other results obtained in simulation environment suggest that IoT solutions offer a feasible solution for the medication intake use case. (Serdaroglu et al., 2015). However, there are currently more than 5881 apps which have not been properly reviewed, and only about 20 of them have been developed with the aid of health professionals (Ahmed et al., 2018). ...
The ever-growing disruption of new technologies like machine learning, block chains, etc in the daily life not only concerns every consumer, but also any scientist, since such technologies explicitly or implicitly introduce novel methodologies challenging the conventional way to generate knowledge and perform science. This development also concerns toxicologists and pharmacologists. While in the past years there has been a tremendous advance in in silico models in some cases capable to replace in vivo models, we observe an ever-growing number of available bio-sensors and of computational capabilities able to track and predict the interaction between a substance and a patient in an unprecedent grade of precision. Because the fast development of these technologies precludes a comprehensive overview of the current state of the art, we think that is very important to make a critical review of the current and in our opinion most prominent application of these novel technologies in the field of toxicology and pharmacology. In this review we provide a critical analysis of the present and future of these technologies in the framework of the precision toxicology/pharmacology. Furthermore, in three theses we suggest how the current paradigm of precision and personalization will evolve to a novel cyclic paradigm where personalization is adjusted in short time cycles due to the availability of ubiquitous data.
... Hence, few of these approaches have already been used for diet and food intake activities monitoring as well [10]. [11] 2006 Smart pill box NA Lid opening [13] 2004 RFID NA Pill bottle removal [14] 2009 RFID NA Pill removal [15] 2010 RFID NA Pill bottle removal [16] 2012 RFID (NFC) NA Pill removal [19] 2015 Body sensors (smart necklace) Smart pill box Pill bottle pick up, pill swallowing [23] [24] 2015, 2016 Wearable sensors (smart watch) NA Pill box opening, medication removal, pill pouring into either hands, and water bottle handling [25] 2016 Wearable sensors (smart watch) NA Hand to mouth movement [26] 2014 Wearable sensors (smart watch) NA Hand movement gesture classification [27] 2014 Body sensors (inertial sensor) NA Cap twisting and hand to mouth movement [28] 2015 Wearable sensors (smart watch) NA Opening pill box, drinking water, putting pill in mouth, and putting glass back [29] 2017 Body sensors (inertial sensor) NA Hands movement [31] 2015 Visual NA Pill bottle weight change detection [33] 2014 RFID Body sensors Proximity sensing and hand movement gesture [34] 2013 RFID Video Pill bottle removal [35] 2011 Visual Ubiquitous sensors Pill removal detection and patient's behavior monitoring detects the lids opening of a 7-day reminder pillbox using plungers that were embedded in each compartment, where the plunger would release a switch inside the device that triggers the MCU about lid opening. Data were wirelessly transferred via a Bluetooth connection to a nearby computer. ...
... In [28], the authors used one smart watch placed on the right hand of the patient to collect the acceleration data for the actions associated with medication intake. The achieved accuracy for putting pill in mouth was 100%, but there was a lot of confusion associated with the processes of opening pill box and drinking activities. ...
... After identifying medicine intake, which type of medicine is taken will be find using RFID tags rested on the blister package. This system can be commercialized as a mobile application along with particular use of hardware [2]. It shows a device that is collected by various components that are direct by Arduino. ...
Medicines are synthesized to cure, cease, prevent diseases or help in the diagnosis of illnesses. Lots of aged people live unaccompanied; few of them are endure from disorder, making it difficult to take care by oneself. Delay of taking their tablets or even taking it at the incorrect interval may raise health consequences. The design of an IoT based medication system is established and it can be used by patients as well as caretakers in sequence to monitor and ensure that the correct amount of each medicine is being taken at the exact time. This provides audio communication to aware the user when a confirmed medicine is to be taken. Furthermore, a software application is used to send messages and email alerts to the patient and the caretaker.
... Serdaroglu et al [8] presented about the Medication adherence which have serious effects on health care services. Continuous medication observation is difficult for patients with forgetfulness which may lead to improper intake of medicines that results in disruption when undergoing a prescribed therapy. ...
The progress in IoT health care is considered to
be a massive contribution to the elderly people. The elderly
people and people who are suffering from chronic diseases
need to intake tablets regularly on timely basis. Care takers
with their busy daily routine may forget the instructions and
time about pills which are prescribed for patient. Also care
takers who are dealing increased number of patients may feel
hectic to sort the medicine list for corresponding patients at
proper time. Earlier many researches have been carried in this
area and different pill boxes have been proposed already. The
intelligent medication box proposed in this work have
specialized features including six sub boxes which helps to
organize six different pills, provides timely remainders for the
patient or caretaker in an android application like hand-held
devices like smartphone. This intelligent medication box
contains bio-sensor for monitoring of temperature and
heartbeat. Overdosage and improper intake of medicines may
lead to serious issues in health of elderly people to avoid mis
usage of medicines a simple authentication process either by
the care taker or the patient himself is performed. The
proposed medication is much safer as it clearly intimates about
time, dosage, stock of medicine and sorts out different pills in
correct sub boxes during the next fill by caretaker
... The study indicated that pharmacy assistants need to accompany nurses on rounds in order to reduce the number of times medication is missed. In [23], a smart medication adherence system is proposed that utilizes RFID for medication type, sensors, and notification to smart phones. The system will ensure that patients take the right medicine at the right time. ...
Internet of Things (IoT), cloud computing, fog computing, and other new technologies are expected to transform the healthcare industry among other. This chapter discusses the use of an automated system based on IoT, cloud and fog computing for constant monitoring of the patient’s health, dispensing medicinal dosage in a timely manner and other comprehensive function for the well-being of both sick and healthy individuals. The wearable embedded devices can capture the patient’s physiological signals including body temperature, blood pressure, electrocardiogram (ECG), oxygen saturation (SpO2), pulse rate, stress, sweating and send them to the cloud server for processing. The processors in individual devices can also communicate and make necessary decisions through fog computing. The medicine dispensing system can monitor the patient medicine details and timings. The real-time captured information can be processed and analyzed to check drug effectiveness and adverse effects on patients. Based on the analysis report, physicians can take decision to continue to use the same drug or change it. It can also help to reduce medication errors by the doctors, nurses, and pharmacists as all the drugs will be identified and recorded by the medicine dispensing system. The system can also improve the medication adherence and critical care based on the real-time medication and physiological signals notifications to the patients, doctors, and family members. The IoT-based system exhibits the ability to achieve objectives for continuous health monitoring through embedded devices with IoT capability and connected to cloud computing and fog computing.
... The potential for supporting people with respiratory conditions, such as asthma and COPD, have also been studied including; wireless body-worn sensors, measuring factors such as a heart rate, blood pressure, and oxygen saturation levels [13,57]; and environmental sensors measuring air humidity, and temperature [13] and air quality [44]. Finally, connected pill boxes have the potential to support medication adherence for a range of conditions, by providing relevant reminders that go beyond traditional preset alarms [65,83]. ...
... There are currently no standard methods for tracking medication adherence [16,17]; however, a variety of strategies are used including smart pill containers" [18][19][20] and wearable or ingestible sensors [21][22][23][24]. Each strategy has its strengths and limitations. ...
A common problem for healthcare providers is accurately tracking patients’ adherence to medication and providing real-time feedback on the management of their medication regimen. This is a particular problem for eye drop medications, as the current commercially available monitors focus on measuring adherence to pills, and not to eye drops. This work presents an intelligent bottle sleeve that slides onto a prescription eye drop medication bottle. The intelligent sleeve is capable of detecting eye drop use, measuring fluid level, and sending use information to a healthcare team to facilitate intervention. The electronics embedded into the sleeve measure fluid level, dropper orientation, the state of the dropper top (on/off), and rates of angular motion during an application. The sleeve was tested with ten patients (age ≥65) and successfully identified and timestamped 94% of use events. On-board processing enabled event detection and the measurement of fluid levels at a 0.4 mL resolution. These data were communicated to the healthcare team using Bluetooth and Wi-Fi in real-time, enabling rapid feedback to the subject. The healthcare team can therefore monitor a log of medication use behavior to make informed decisions on treatment or support for the patient.
... The study indicated that pharmacy assistants need to accompany nurses on rounds in order to reduce the number of times medication is missed. In [23], a smart medication adherence system is proposed that utilizes RFID for medication type, sensors, and notification to smart phones. The system will ensure that patients take the right medicine at the right time. ...
Internet of Things (IoT) is changing the way many sectors operate and special attention is paid to promoting healthy living by employing IoT based technologies. In this paper, a novel approach is developed with IoT prototype of Wireless Sensor Network and Cloud based system to provide continuous monitoring of a patient’s health status, ensuring timely scheduled and unscheduled medicinal dosage based on real-time patient vitals measurement, life-saving emergency prediction and communication. The designed integrated prototype consists of a wearable expandable health monitoring system, Smart Medicine Dispensing System, Cloud-based Big Data analytical diagnostic and Artificial Intelligence (AI) based reporting tool. A working prototype was developed and tested on few persons to ensure that it is working according to expected standards. Based on the initial experiments, the system fulfilled intended objectives including continuous health monitoring, scheduled timely medication, unscheduled emergency medication, life-saving emergency reporting, life-saving emergency prediction and early stage diagnosis. In addition, based on the analysis reports, physicians can diagnose/decide, view medication side effects, medication errors and prescribe medication accordingly. The proposed system exhibited the ability to achieve objectives it was designed using IoT to alleviate the pressure on hospitals due to crowdedness in hospital care and to reduce the healthcare service delays.
... Classification accuracies were 95% and 97.5% for cap twisting and hand-to-mouth actions, respectively. Another application of accelerometers embedded in smartwatches is presented in [71]. One smartwatch placed on the right hand of the user was used to collect the acceleration data for the actions associated with medication intake. ...
Medication non-adherence is a prevalent, complex problem. Failure to follow medication schedules may lead to major health complications, including death. Proper medication adherence is thus required in order to gain the greatest possible drug benefit during a patient's treatment. Interventions have been proven to improve medication adherence if deviations are detected. This review focuses on recent advances in the field of technology-based medication adherence approaches and pays particular attention to their technical monitoring aspects. The taxonomy space of this review spans multiple techniques including sensor systems, proximity sensing, vision systems, and combinations of these. As each technique has unique advantages and limitations, this work describes their trade-offs in accuracy, energy consumption, acceptability and user's comfort, and user authentication.
... Conceptually, it is possible to deduce activity by analyzing the patient movements. Serdaroglu, Uslu and Baydere [13] discuss the possibility of using an IoT device on the wrist such as a wrist watch on the patient to detect and track motion. It is argued that certain motions can be classified and the activities related to medicine adherence involves identifying the activities in an order of (1) pouring water into the glass,(2) taking medication from the blister pack, (3) taking pill to the mouth and (4) drinking water activities. ...
The impact of medication adherence is proven to affect hospitalization risk and healthcare cost. In Malaysia, failure in taking medication as prescribed is a complicated and common problem. There are many reasons people fail to take their medication as prescribed. Some of the few common issues are remembering to consume their medicine timely, doubts in medication effectiveness, concerns on side effects or having difficulty in taking the medication (especially with injections or inhalers) and rising cost of prescribed medications. Tracking patient medication intake leads to many advantages, one of which is lower medication due to adherence, health-care cost which effects insurance coverage premiums. In this paper, we develop an application to track medication using a smart phone and the use of QR code. Each medication transaction requires a proof of work to signify that the user have taken their medication at the correct date and time. A QR-Code reader will be used to capture medication taken printed on medication labels and each scan is considered as a proof of work. Our contribution is to create a mobile application that can help track medication intake, as well as remind, inform and warn users about the medication that they are taking.
With the advancement in wireless sensing technology, the notion of the Internet of Things (IoT) has become ubiquitous and widely adopted due to its extensive applications in smart living. In that regard, Human Activity Recognition (HAR) is an indispensable part of intelligent systems for continuous supervision of human behavior. The design of effective applications requires accurate and relevant information on people’s activities and behaviors. This chapter presents a comprehensive review of the existing IoT-based HAR systems in the literature that promote smart living. The contribution of this work is to drive interested researchers toward the concept of HAR through existing works. In addition, a hands-on experimental case study is also provided to demonstrate the practical application of the concept, for recognizing and predicting several activities, in real life. Accordingly, here we first study the general architecture of the HAR system along with its principal components in detail. Next, the conglomeration of HAR with IoT and its impact on each other has been intensely explored to provide extensive insight for further research. We discuss the research challenges associated with developing IoT-based HAR systems. Next, the state-of-the-art works in HAR based on wearable sensors are surveyed thoroughly. At the end, we present a case study on HAR, using UCIHAR, a standard benchmark dataset that exhibits the practical implementation of the framework on different Machine Learning as well as Deep Learning models.KeywordsWireless sensor networkIoTHARWearable sensorMachine learningDeep learning
Engineers play an important role in ensuring the trustworthiness of the clinical enterprise and in the ecosystem of clinical IoT development and implementation. This chapter considers (1) what trust is and why it matters; (2) ethical frameworks that can be used to evaluate one’s own work and professional development; (3) key use cases in trust, ethics, and engineering; and (4) leadership. We outline several frameworks for engineers to think about what trust means, how it applies to their work, and how ethics plays a role in the scope and impact of their work. We encourage engineers to consider how they can engage the wider ecosystem necessary for the effective development of trustworthy clinical IoT—including engagement with management, clinicians, and patients.KeywordsTrustEthicsProfessionalismEcosystemQualityEvidenceReliabilityPrinciplesSystemsLeadershipPolicyClinical IoT
In recent years, Internet of Things (IoT) technology has gained tremendous attention for its ability to relieve the burden on healthcare caused by an aging population and the increase in chronic disease. IoT technology facilitates the tracking of patients with different conditions and the processing of vast volumes of data, of which a substantial part of this data are physiological signals. Physiological signals are an invaluable source of data which helps to diagnose, rehabilitate, and treat diseases. The signals come from a Wireless Body Sensor Network or wearable devices placed on a patient's body. There are many difficulties in IoT-healthcare systems, such as data collection and processing especially that (1) wireless sensor nodes have limited energy, processing and memory resources, (2) the quantity of data collected periodically is enormous, (3) the quality of the data collected is not always satisfactory, and that such data are highly likely to include noisy or unreliable areas, and (4) the manual feature extraction process from the physiological signals requires significant human intervention and medical expertise.Firstly, an energy-efficient data compression technique is being proposed in this dissertation. The proposed scheme is based on an error-bound lossy compressor originally designed for high-performance computing applications and has been adapted for IoT devices that are resource constrained. The proposed solution is an easy-to-implement algorithm, which could reduce energy consumption by as much as 2.5 times. It also reduces the processing/transmission time by compressing large batches of data before their transfer from the IoT to the edge. Moreover, an empirical analysis was carried out to study the effect of lossy time series compression on the classification task. In addition to different variations of compression methods, various deep neural networks for time series classification were considered to identify the appropriate trade-off between the compression ratio and classification performance.Second, the problem of data distortion due to lossy compression was addressed. The reconstructed data may not always be satisfactory, given that the physiological signals collected are multivariate time series that are highly compressed with a lossy compressor prior to transmission. To solve this limitation, a convolutional autoencoder-based deep learning model was introduced. The proposed model was able to enhance the compressed data after reconstruction, thus fixing the shape of the physiological signal and allowing more accurate feature extraction.Then, the photoplethysmogram signals were given particular attention. Photoplethysmography (PPG) is used to measure the skin blood flow using infrared light. Photoplethysmography is a promising technique because of the capability of the new wrist-worn devices to provide the signal. The existence of motion artifacts and meaningless areas in the signal is a major challenge faced when working with this signal. A deep learning model for automatic motion artifacts detection based on a CNN-LSTM autoencoder architecture has been proposed to detect and discard irrelevant zones in photoplethysmogram signals to avoid analyzing and processing meaningless data.Finally, a deep learning model has been proposed for time series classification. Given that hand-designing features from physiological signals is a challenging task, the representation learning approach is a potential solution for automatically learning features from raw physiological data. A classification model based on the DenseNet architecture was proposed for the classification of multivariate time series. The results show that the proposed approach is capable of achieving and in several cases bypassing the performance of the state-of-the-art models on a benchmark dataset.
Background
Thanks to the increased interest towards health and lifestyle, a larger adoption in wearable devices for activity tracking is present among the general population. Wearable devices such as smart wristbands integrate inertial units, including accelerometers and gyroscopes, which can be utilised to perform automatic classification of hand gestures. This technology could also find an important application in automatic medication adherence monitoring. Accordingly, this study aims at comparing the performance of several Machine-Learning (ML) and Deep-Learning (DL) approaches for the automatic identification of hand gestures, with a specific focus on the drinking gesture, commonly associated to the action of oral intake of a pill-packed medication.
Methods
A method to automatically recognize hand gestures in daily living is proposed in this work. The method relies on a commercially available wristband sensor (MetaMotionR, MbientLab Inc.) integrating tri-axial accelerometer and gyroscope. Both ML and DL algorithms were evaluated for both multi-gesture (drinking, eating, pouring water, opening a bottle, typing, answering a phone, combing hair, and cutting) and binary gesture (drinking versus other gestures) classification from wristband sensor signals. Twenty-two participants were involved in the experimental analysis, performing a 10-minute acquisition in a laboratory setting. Leave-one-subject-out cross validation was performed for robust performance assessment.
Results
The highest performance was achieved using a convolutional neural network with long- short term memory (CNN-LSTM), with a median f1-score of 90.5 [first quartile: 84.5; third quartile: 92.5]% and 92.5 [81.5;98.0]% for multi-gesture and binary classification, respectively.
Conclusions
Experimental results showed that hand gesture classification with ML/DL from wrist accelerometers and gyroscopes signals can be performed with reasonable accuracy in laboratory settings, paving the way for a new generation of medical devices for monitoring medical adherence.
In recent years, IoT has been revolutionizing the technology landscape, leading to explosive growth in fields like automated manufacturing, asset management, and wearable consumer healthcare products. IoT’s presence can be seen in all domains. Ever since IoT gained its entry into the medical field, there has been a magnificent transformation in the healthcare domain. Personal healthcare is now made a reality with IoT offering solutions in several dimensions like remote healthcare; smart clothing such as smartwatches, smart bands, and smart pants; and telemedicine like smart pills and personal care robots. This chapter gives a comprehensive walkthrough of IoT in the healthcare ecosystem, addressing the different applications of IoT in healthcare, the architecture models, challenges faced by IoT in healthcare, security practices and issues, and the future of IoT in the domain. In the first section, IoT applications in healthcare are discussed, which include patient-centric applications like remote health monitoring and critical care monitoring; and hospital-centric IoT applications such as the deployment of the staff, reducing charting times, and real-time location of medical equipment, subsequently followed by a discussion on how the data collected from the patient-centric and hospital-centric applications contribute to the ease of other domains like health insurance; and then, IoT’s support toward the pharmaceutical industry to restrict counterfeit medicine is discussed. Secondly, the implementation designs of healthcare IoT are discussed. Apart from the traditional cloud services, new offerings like fog and edge computing have seen a spike in recent years. Fog and edge computing are considered intelligent and flexible architectures. The subsequent section deals with architecture designs and the advantages and challenges of the two computing models. The next section demonstrates the actual implementation methodologies of the two applications in the following domains in detail: (1) Heart disease prediction and (2) healthcare IoT-based affective state mining using deep convolutional neural networks. The following section discusses the challenges faced by IoT in the healthcare domain. In general, the challenges can be categorized into technological challenges; people-oriented challenges like the acceptance of IoT in the healthcare domain; and finally, security bottlenecks. The data generated and maintained by the IoT platforms serves as a gold mine for different healthcare professionals for future research and development in the medical field and the health insurance providers and pharmaceutical industries for their benefits. Hence, more emphasis is given to the security and privacy aspects of how the domain handles sensitive data of the patients. The next section provides insights into the security issues along with the cyber threats and attacks faced by healthcare IoT and the defensive mechanisms. Furthermore, this chapter deals with the IoT’s role in combatting the novel coronavirus that has caused an unprecedented global pandemic. Finally, the future of IoT is talked about. With the advent of 5G and an upsurge in artificial intelligence, the different dimensions of the IoT that are expected to see an outburst of growth are discussed.
The chapter at first focuses on IoT being used in healthcare in general and then it moves to Parkinson's disease, specifically, and how people with it can be benefitted by IoT.
The chapter at first focuses on IoT being used in healthcare in general and then it moves to Parkinson's disease, specifically, and how people with it can be benefitted by IoT.
Internet of things is a revolutionary domain, when we use it for the wellness of people in a smart way. As of now, the cost to implement IoT-enabled services is very high. So, this chapter introduces a cost effective and a reliable system to monitor patients at home and in hospitals with the help of IoT. The monitored details of a person can be drawn at any time with the help of an android app, which can produce output at real-time. The processed data are stored in the UBIDOTS cloud server, and the patients' needs can be met in time as well lives saved during critical cases with the help of the system proposed in this chapter.
Medication nonadherence is an important health consideration that affects the patient’s overall well-being and healthcare costs. This study conducts the literature review on medication adherence and presents the recent trends in measuring, predicting, and improving adherence for nonadherent patients using advanced analytical methods. A combination of advanced medication adherence metrics employing information technology capabilities and using analytical methods can help healthcare providers to discover future patterns, knowledge, and insights about the patient situation, at the same time enabling to shape a specific intervention to improve adherence to medication.
A method or apparatus for monitoring human activity using an inertial sensor is described. The method includes monitoring accelerations, and identifying a current user activity from a plurality of user activities based on the accelerations. In one embodiment, each of the plurality of user activities is associated with one of a plurality of types of periodic human motions that are detectable by the portable electronic device, and wherein the identification of the current user activity is made based on detecting two or more instances of the periodic human motion associated with the user activity. The method further includes counting periodic human motions appropriate to the current user activity.
The paper presents wireless sensor architecture based human fall detection system especially for elderly people. The human fall detection system is implemented using two wireless network architectures, namely cross bow wireless sensor node system and PIC18LF4620 wireless module. The fall detection mechanism is conceived with two sensors including tri-axis accelerometer and Passive Infrared (PIR) sensor which are interfaced to monitor the activity of elderly people. The resultant voltage changes because of human activities are the targeted interest which is detected by both sensors. The real time output is compared with stored templates and the computational analysis is done within microcontroller at site itself. Any abnormality from stored templates is informed to the caretaker through wireless architecture. (C) 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of ICCTSD 2011
Monitoring of posture allocations and activities en-ables accurate estimation of energy expenditure and may aid in obesity prevention and treatment. At present, accurate devices rely on multiple sensors distributed on the body and thus may be too obtrusive for everyday use. This paper presents a novel wearable sensor, which is capable of very accurate recognition of common postures and activities. The patterns of heel acceleration and plan-tar pressure uniquely characterize postures and typical activities while requiring minimal preprocessing and no feature extraction. The shoe sensor was tested in nine adults performing sitting and standing postures and while walking, running, stair ascent/descent and cycling. Support vector machines (SVMs) were used for clas-sification. A fourfold validation of a six-class subject-independent group model showed 95.2% average accuracy of posture/activity classification on full sensor set and over 98% on optimized sensor set. Using a combination of acceleration/pressure also enabled a pronounced reduction of the sampling frequency (25 to 1 Hz) with-out significant loss of accuracy (98% versus 93%). Subjects had shoe sizes (US) M9.5-11 and W7-9 and body mass index from 18.1 to 39.4 kg/m2 and thus suggesting that the device can be used by individuals with varying anthropometric characteristics.
In enclosed suits, such as those worn by explosive ordnance disposal (EOD) experts, evaporative cooling through perspiration is less effective and, particularly in hot environments, uncompensable heat stress (UHS) may occur. Although some suits have cooling systems, their effectiveness during missions is dependent on the operative's posture. In order to properly assess thermal state, temperature-based assessment systems need to take posture into account. This paper builds on previous work for instrumenting EOD suits with regard to temperature monitoring and proposes to also monitor operative posture with MEMS accelerometers. Posture is a key factor in predicting how body temperature will change and is therefore important in providing local or remote warning of the onset of UHS. In this work, the C4.5 decision tree algorithm is used to produce an on-line classifier that can differentiate between nine key postures from current acceleration readings. Additional features that summarize how acceleration is changing over time are used to improve average classification accuracy to around 97.2%. Without such temporal feature extraction, dynamic postures are difficult to classify accurately. Experimental results show that training over a variety of subjects, and in particular, mixing gender, improves results on unseen subjects. The main advantages of the on-line posture classification system described here are that it is accurate, does not require integration of acceleration over time, and is computationally lightweight, allowing it to be easily supported on wearable microprocessors.
Ambulatory monitoring of motor symptoms in Parkinsons disease (PD) can improve our therapeutic strategies, especially in patients with motor fluctuations. Previously published monitors usually assess only one or a few basic aspects of the cardinal motor symptoms in a laboratory setting. We developed a novel ambulatory monitoring system that provides a complete motor assessment by simultaneously analyzing current motor activity of the patient (e.g. sitting, walking) and the severity of many aspects related to tremor, bradykinesia, and hypokinesia. The monitor consists of a set of four inertial sensors. Validity of our monitor was established in seven healthy controls and six PD patients treated with deep brain stimulation (DBS) of the subthalamic nucleus. Patients were tested at three different levels of DBS treatment. Subjects were monitored while performing different tasks, including motor tests of the Unified Parkinsons Disease Rating Scale (UPDRS). Output of the monitor was compared to simultaneously recorded videos. The monitor proved very accurate in discriminating between several motor activities. Monitor output correlated well with blinded UPDRS ratings during different DBS levels. The combined analysis of motor activity and symptom severity by our PD monitor brings true ambulatory monitoring of a wide variety of motor symptoms one step closer..
Stroke is the leading cause of disability in the United States. It is estimated that 700,000 people in the United States will experience a stroke each year and that there are over 5 million Americans living with a stroke. In this paper we describe a novel methodology for automatic recognition of postures and activities in patients with stroke that may be used to provide behavioral enhancing feedback to patients with stroke as part of a rehabilitation program and potentially enhance rehabilitation outcomes. The recognition methodology is based on Support Vector classification of the sensor data provided by a wearable shoe-based device. The proposed methodology was validated in a case study involving an individual with a chronic stroke with impaired motor function of the affected lower extremity and impaired walking ability. The results suggest that recognition of postures and activities may be performed with very high accuracy.
Abstract Activity monitoring systems (AMS) detect actions performed by humans. For an AMS to be effectively deployed in daily life, it should partition sensor data streams in real time and determine what activity corresponds to each partition. In this work, a real time continuous activity monitoring system, named RAM, is proposed. RAM detects simple and composite activities, collecting the data with a single 3D accelerometer to produce a non-invasive solution. Classification module of RAM carries out non-predefined feature extraction and activity detection in a coalesced manner thanks to feature extractive training, whereas the state-of-art classifiers need to be fed with the output of a predefined feature extraction scheme. As being a Support Vector Machines (SVM) inspired solution, RAM fulfils multiclass classification with one-against-pseudo class strategy, without generating hyperplanes. The strength of the proposed model lies in that RAM achieves robustness in terms of inter-activity detection consistency and time efficiency with little overhead. Robustness property offers a potential to reduce the need for re-training an expert system, which faces the problem of growing set of activity classes in the real time activity recognition domain. We compared RAM for a set of hand oriented activities, against 8 different configurations, where SVM and K-Nearest Neighbour (KNN) classifiers are fed with different predefined features. We observed that RAM outperforms these configurations in overall accuracy as well as inter-activity detection consistency. We also presented the results of real tests as a proof-of-concept for transition detection in composite activities.
The future internetworking is envisioned to develop concepts that can autonomously interact with the physical world. Wireless Sensor Network (WSN) technology is considered as the defacto concept which is the bridge to the real world. So, this technology should be adapted to support interoperability with the commodity Internet. This is not a straightforward process because the technological background of these networks do not fit each other. Hence, seamless interconnection between WSN and the IP networks is a challenging research problem. In this experimental study, proxy-based and gateway-based approaches to the solution are analyzed and a hybrid approach which combines the advantages of them is defined. The proposed approach is used to build a web server for WSNs. The overall performance of the new server and its adaptation layer is tested on a real testbed with a series of experiments. The early results obtained in this study are very encouraging. Thus, further analysis is needed towards a seamless interconnection architecture for WSNs.
Live physiological monitoring of soccer players during sporting events can help maximise athlete performance while preventing injury, and enable new applications for referee-assist and enhanced television broadcast services. However, the harsh operating conditions in the soccer field pose several challenges: (a) body-mounted wireless sensor devices have limited radio range, (b) playing area is large, necessitating multi-hop transmission, (c) wireless connectivity is dynamic due to extreme mobility, and (d) data forwarding has to operate within tight delay/energy constraints. In this paper, we take a first step towards characterising wireless connectivity in the soccer field by undertaking experimental work with local soccer clubs, and assess the feasibility of real-time athlete monitoring. We make three specific contributions: (1) We develop an empirical profile of radio signal strength in an open soccer field taking into account distance and body orientation of the athlete. (2) Using data from several soccer games we profile key characteristics of wireless connectivity, highlighting aspects such as small power-law inter-encounters and link correlations. (3) We develop practical multi-hop routing algorithms that can be tuned to achieve the right balance between the competing objectives of resource consumption and data extraction delay. We believe our study is the first to characterise the wireless environment for mobile sensor networks in field sports, and paves the way towards realisation of real-time athlete monitoring systems.
Physical activity has a fundamental role in the prevention and treatment of chronic disease. The precise measurement of physical activity is key to many surveillance and epidemiological studies investigating trends and associations with disease. Public health initiatives aimed at increasing physical activity rely on the measurement of physical activity to monitor their effectiveness. Physical activity is multidimensional, and a complex behaviour to measure; its various domains are often misunderstood. Inappropriate or crude measures of physical activity have serious implications, and are likely to lead to misleading results and underestimate effect size. In this review, key definitions and theoretical aspects, which underpin the measurement of physical activity, are briefly discussed. Methodologies particularly suited for use in epidemiological research are reviewed, with particular reference to their validity, primary outcome measure and considerations when using each in the field. It is acknowledged that the choice of method may be a compromise between accuracy level and feasibility, but the ultimate choice of tool must suit the stated aim of the research. A framework is presented to guide researchers on the selection of the most suitable tool for use in a specific study.
The duration of a sit-to-stand (SiSt) transfer is a representative measure of a person's status of physical mobility. This paper measured the duration unobtrusively and automatically using a pressure sensor array under a bed mattress and a floor plate beside the bed. Pressure sequences were extracted from frames of sensor data measuring bed and floor pressure over time. The start time was determined by an algorithm based on the motion of the center of pressure (COP) on the mattress toward the front edge of the bed. The end time was determined by modeling the foot pressure exerted on the floor in the wavelet domain as the step response of a third-order transfer function. As expected, young and old healthy adults generated shorter SiSt durations of around 2.31 and 2.88 s, respectively, whereas post-hip fracture and post-stroke adults produced longer SiSt durations of around 3.32 and 5.00 s. The unobtrusive nature of pressure sensing techniques used in this paper provides valuable information that can be used for the ongoing monitoring of patients within extended-care facilities or within the smart home environment.
An imbalanced diet elevates health risks for many chronic diseases including obesity. Dietary monitoring could contribute vital information to lifestyle coaching and diet management, however, current monitoring solutions are not feasible for a long-term implementation. Towards automatic dietary monitoring, this work targets the continuous recognition of dietary activities using on-body sensors.
An on-body sensing approach was chosen, based on three core activities during intake: arm movements, chewing and swallowing. In three independent evaluation studies the continuous recognition of activity events was investigated and the precision-recall performance analysed. An event recognition procedure was deployed, that addresses multiple challenges of continuous activity recognition, including the dynamic adaptability for variable-length activities and flexible deployment by supporting one to many independent classes. The approach uses a sensitive activity event search followed by a selective refinement of the detection using different information fusion schemes. The method is simple and modular in design and implementation.
The recognition procedure was successfully adapted to the investigated dietary activities. Four intake gesture categories from arm movements and two food groups from chewing cycle sounds were detected and identified with a recall of 80-90% and a precision of 50- 64%. The detection of individual swallows resulted in 68% recall and 20% precision. Sample-accurate recognition rates were 79% for movements, 86% for chewing and 70% for swallowing.
Body movements and chewing sounds can be accurately identified using on-body sensors, demonstrating the feasibility of on-body dietary monitoring. Further investigations are needed to improve the swallowing spotting performance.
Activity recognition apparatus, method and program
- B Clarkson