1
|
Yang M, Peng L, Liu L, Wang Y, Zhang Z, Yuan Z, Zhou J. LCSED: A low complexity CNN based SED model for IoT devices. Neurocomputing 2022. [DOI: 10.1016/j.neucom.2021.02.104] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
|
2
|
Physiological and Behavior Monitoring Systems for Smart Healthcare Environments: A Review. SENSORS 2020; 20:s20082186. [PMID: 32290639 PMCID: PMC7218909 DOI: 10.3390/s20082186] [Citation(s) in RCA: 28] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 03/14/2020] [Revised: 04/01/2020] [Accepted: 04/10/2020] [Indexed: 02/04/2023]
Abstract
Healthcare optimization has become increasingly important in the current era, where numerous challenges are posed by population ageing phenomena and the demand for higher quality of the healthcare services. The implementation of Internet of Things (IoT) in the healthcare ecosystem has been one of the best solutions to address these challenges and therefore to prevent and diagnose possible health impairments in people. The remote monitoring of environmental parameters and how they can cause or mediate any disease, and the monitoring of human daily activities and physiological parameters are among the vast applications of IoT in healthcare, which has brought extensive attention of academia and industry. Assisted and smart tailored environments are possible with the implementation of such technologies that bring personal healthcare to any individual, while living in their preferred environments. In this paper we address several requirements for the development of such environments, namely the deployment of physiological signs monitoring systems, daily activity recognition techniques, as well as indoor air quality monitoring solutions. The machine learning methods that are most used in the literature for activity recognition and body motion analysis are also referred. Furthermore, the importance of physical and cognitive training of the elderly population through the implementation of exergames and immersive environments is also addressed.
Collapse
|
3
|
Rana R, Latif S, Gururajan R, Gray A, Mackenzie G, Humphris G, Dunn J. Automated screening for distress: A perspective for the future. Eur J Cancer Care (Engl) 2019; 28:e13033. [DOI: 10.1111/ecc.13033] [Citation(s) in RCA: 21] [Impact Index Per Article: 4.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/09/2019] [Revised: 02/05/2019] [Accepted: 02/18/2019] [Indexed: 01/13/2023]
Affiliation(s)
- Rajib Rana
- University of Southern Queensland Springfield Queensland Australia
| | - Siddique Latif
- University of Southern Queensland Springfield Queensland Australia
| | - Raj Gururajan
- University of Southern Queensland Springfield Queensland Australia
| | - Anthony Gray
- University of Southern Queensland Springfield Queensland Australia
| | | | | | - Jeff Dunn
- University of Southern Queensland Springfield Queensland Australia
- Griffith University Brisbane Queensland Australia
- University of Technology Sydney Sydney New South Wales Australia
| |
Collapse
|
4
|
|
5
|
Uddin MZ, Khaksar W, Torresen J. Ambient Sensors for Elderly Care and Independent Living: A Survey. SENSORS (BASEL, SWITZERLAND) 2018; 18:E2027. [PMID: 29941804 PMCID: PMC6068532 DOI: 10.3390/s18072027] [Citation(s) in RCA: 85] [Impact Index Per Article: 14.2] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 05/09/2018] [Revised: 06/14/2018] [Accepted: 06/18/2018] [Indexed: 11/17/2022]
Abstract
Elderly care at home is a matter of great concern if the elderly live alone, since unforeseen circumstances might occur that affect their well-being. Technologies that assist the elderly in independent living are essential for enhancing care in a cost-effective and reliable manner. Elderly care applications often demand real-time observation of the environment and the resident’s activities using an event-driven system. As an emerging area of research and development, it is necessary to explore the approaches of the elderly care system in the literature to identify current practices for future research directions. Therefore, this work is aimed at a comprehensive survey of non-wearable (i.e., ambient) sensors for various elderly care systems. This research work is an effort to obtain insight into different types of ambient-sensor-based elderly monitoring technologies in the home. With the aim of adopting these technologies, research works, and their outcomes are reported. Publications have been included in this survey if they reported mostly ambient sensor-based monitoring technologies that detect elderly events (e.g., activities of daily living and falls) with the aim of facilitating independent living. Mostly, different types of non-contact sensor technologies were identified, such as motion, pressure, video, object contact, and sound sensors. Besides, multicomponent technologies (i.e., combinations of ambient sensors with wearable sensors) and smart technologies were identified. In addition to room-mounted ambient sensors, sensors in robot-based elderly care works are also reported. Research that is related to the use of elderly behavior monitoring technologies is widespread, but it is still in its infancy and consists mostly of limited-scale studies. Elderly behavior monitoring technology is a promising field, especially for long-term elderly care. However, monitoring technologies should be taken to the next level with more detailed studies that evaluate and demonstrate their potential to contribute to prolonging the independent living of elderly people.
Collapse
Affiliation(s)
- Md Zia Uddin
- Department of Informatics, University of Oslo, 0316 Oslo, Norway.
| | - Weria Khaksar
- Department of Informatics, University of Oslo, 0316 Oslo, Norway.
| | - Jim Torresen
- Department of Informatics, University of Oslo, 0316 Oslo, Norway.
| |
Collapse
|
6
|
Han W, Coutinho E, Ruan H, Li H, Schuller B, Yu X, Zhu X. Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments. PLoS One 2016; 11:e0162075. [PMID: 27627768 PMCID: PMC5023122 DOI: 10.1371/journal.pone.0162075] [Citation(s) in RCA: 41] [Impact Index Per Article: 5.1] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/17/2015] [Accepted: 08/17/2016] [Indexed: 11/19/2022] Open
Abstract
Coping with scarcity of labeled data is a common problem in sound classification tasks. Approaches for classifying sounds are commonly based on supervised learning algorithms, which require labeled data which is often scarce and leads to models that do not generalize well. In this paper, we make an efficient combination of confidence-based Active Learning and Self-Training with the aim of minimizing the need for human annotation for sound classification model training. The proposed method pre-processes the instances that are ready for labeling by calculating their classifier confidence scores, and then delivers the candidates with lower scores to human annotators, and those with high scores are automatically labeled by the machine. We demonstrate the feasibility and efficacy of this method in two practical scenarios: pool-based and stream-based processing. Extensive experimental results indicate that our approach requires significantly less labeled instances to reach the same performance in both scenarios compared to Passive Learning, Active Learning and Self-Training. A reduction of 52.2% in human labeled instances is achieved in both of the pool-based and stream-based scenarios on a sound classification task considering 16,930 sound instances.
Collapse
Affiliation(s)
- Wenjing Han
- Language Computing Lab, Samsung R&D Institute of China - Beijing (SRC-B), Beijing, China
| | - Eduardo Coutinho
- Department of Music, University of Liverpool, Liverpool, United Kingdom
- Department of Computing, Imperial College London, London, United Kingdom
| | - Huabin Ruan
- Department of Computer Science and Technology, Tsinghua University, Beijing, China
- * E-mail:
| | - Haifeng Li
- School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China
| | - Björn Schuller
- Department of Computing, Imperial College London, London, United Kingdom
- School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China
- Complex Systems Engineering, University of Passau, Passau, Germany
| | - Xiaojie Yu
- Language Computing Lab, Samsung R&D Institute of China - Beijing (SRC-B), Beijing, China
| | - Xuan Zhu
- Language Computing Lab, Samsung R&D Institute of China - Beijing (SRC-B), Beijing, China
| |
Collapse
|
7
|
Do HM, Sheng W, Liu M. Human-assisted sound event recognition for home service robots. ROBOTICS AND BIOMIMETICS 2016; 3:7. [PMID: 27330932 PMCID: PMC4889638 DOI: 10.1186/s40638-016-0042-2] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 03/03/2016] [Accepted: 05/17/2016] [Indexed: 11/10/2022]
Abstract
This paper proposes and implements an open framework of active auditory learning for a home service robot to serve the elderly living alone at home. The framework was developed to realize the various auditory perception capabilities while enabling a remote human operator to involve in the sound event recognition process for elderly care. The home service robot is able to estimate the sound source position and collaborate with the human operator in sound event recognition while protecting the privacy of the elderly. Our experimental results validated the proposed framework and evaluated auditory perception capabilities and human-robot collaboration in sound event recognition.
Collapse
Affiliation(s)
- Ha Manh Do
- School of Electrical and Computer Engineering, Oklahoma State University, Stillwater, OK 74078 USA ; Beijing Advanced Innovation Center for Imaging Technology, Capital Normal University, Beijing, 100048 China
| | - Weihua Sheng
- School of Electrical and Computer Engineering, Oklahoma State University, Stillwater, OK 74078 USA ; Beijing Advanced Innovation Center for Imaging Technology, Capital Normal University, Beijing, 100048 China
| | - Meiqin Liu
- College of Electrical Engineering, Zhejiang University, Hangzhou, 310027 China
| |
Collapse
|
8
|
Chen KY, Janz KF, Zhu W, Brychta RJ. Redefining the roles of sensors in objective physical activity monitoring. Med Sci Sports Exerc 2012; 44:S13-23. [PMID: 22157770 DOI: 10.1249/mss.0b013e3182399bc8] [Citation(s) in RCA: 91] [Impact Index Per Article: 7.6] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/21/2022]
Abstract
BACKGROUND Because physical activity researchers are increasingly using objective portable devices, this review describes the current state of the technology to assess physical activity, with a focus on specific sensors and sensor properties currently used in monitors and their strengths and weaknesses. Additional sensors and sensor properties desirable for activity measurement and best practices for users and developers also are discussed. BEST PRACTICES We grouped current sensors into three broad categories for objectively measuring physical activity: associated body movement, physiology, and context. Desirable sensor properties for measuring physical activity and the importance of these properties in relationship to specific applications are addressed, and the specific roles of transducers and data acquisition systems within the monitoring devices are defined. Technical advancements in sensors, microcomputer processors, memory storage, batteries, wireless communication, and digital filters have made monitors more usable for subjects (smaller, more stable, and longer running time) and for researchers (less costly, higher time resolution and memory storage, shorter download time, and user-defined data features). FUTURE DIRECTIONS Users and developers of physical activity monitors should learn about the basic properties of their sensors, such as range, accuracy, and precision, while considering the data acquisition/filtering steps that may be critical to data quality and may influence the desirable measurement outcome(s).
Collapse
Affiliation(s)
- Kong Y Chen
- Diabetes, Endocrinology, and Obesity Branch, Intramural Research Program, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892, USA.
| | | | | | | |
Collapse
|
9
|
Li KF. Smart home technology for telemedicine and emergency management. JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING 2012; 4:535-546. [PMID: 32218875 PMCID: PMC7090692 DOI: 10.1007/s12652-012-0129-8] [Citation(s) in RCA: 11] [Impact Index Per Article: 0.9] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 02/06/2012] [Accepted: 04/17/2012] [Indexed: 06/10/2023]
Abstract
With the ageing population, mobility is an important issue and it deters the elderlies to visit health clinics on a regular basis. Individuals with disabilities also face the same obstacles for their out-of-home medical visits. In addition, people living in remote areas often do not get the needed health care attention unless they are willing to spend the time, effort and cost to travel. Advances in information and telecommunication technologies have made telemedicine possible. Using the latest sensor technologies, a person's vital data can be collected in a smart home environment. The bio-information can then be transferred wirelessly or via the Internet to medical databases and the healthcare professionals. Using the appropriate sensing apparatus at a smart home setting, patients, elderlies and people with disabilities can have their health signals and information examined on a real-time and archival basis. Recovery process can be charted on a regular basis. Remote emergency alerts can be intercepted and responded quickly. Health deterioration can be monitored closely enabling corrective actions. Medical practitioners can therefore provide the necessary health-related services to more people. This paper surveys and compiles the state-of-the-art smart home technologies and telemedicine systems.
Collapse
Affiliation(s)
- Kin Fun Li
- Department of Electrical and Computer Engineering, University of Victoria, Victoria, BC V8W 3P6 Canada
| |
Collapse
|
10
|
Souli S, Lachiri Z. Environmental Sounds Classification Based on Visual Features. PROGRESS IN PATTERN RECOGNITION, IMAGE ANALYSIS, COMPUTER VISION, AND APPLICATIONS 2011. [DOI: 10.1007/978-3-642-25085-9_54] [Citation(s) in RCA: 9] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/11/2022]
|
11
|
Fleury A, Vacher M, Noury N. SVM-based multimodal classification of activities of daily living in Health Smart Homes: sensors, algorithms, and first experimental results. ACTA ACUST UNITED AC 2009; 14:274-83. [PMID: 20007037 DOI: 10.1109/titb.2009.2037317] [Citation(s) in RCA: 334] [Impact Index Per Article: 22.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/10/2022]
Abstract
By 2050, about one third of the French population will be over 65. Our laboratory's current research focuses on the monitoring of elderly people at home, to detect a loss of autonomy as early as possible. Our aim is to quantify criteria such as the international activities of daily living (ADL) or the French Autonomie Gerontologie Groupes Iso-Ressources (AGGIR) scales, by automatically classifying the different ADL performed by the subject during the day. A Health Smart Home is used for this. Our Health Smart Home includes, in a real flat, infrared presence sensors (location), door contacts (to control the use of some facilities), temperature and hygrometry sensor in the bathroom, and microphones (sound classification and speech recognition). A wearable kinematic sensor also informs postural transitions (using pattern recognition) and walk periods (frequency analysis). This data collected from the various sensors are then used to classify each temporal frame into one of the ADL that was previously acquired (seven activities: hygiene, toilet use, eating, resting, sleeping, communication, and dressing/undressing). This is done using support vector machines. We performed a 1-h experimentation with 13 young and healthy subjects to determine the models of the different activities, and then we tested the classification algorithm (cross validation) with real data.
Collapse
Affiliation(s)
- Anthony Fleury
- AFIRM Team, Techniques de l'Ingénierie Médicale et de la Complexité-Informatique, Mathématique et Applications, Grenoble laboratory, Unité Mixte de Recherche 5525, Centre National de la Recherche Scientifique/Université Joseph Fourier, Faculté de Médecine de Grenoble, F-38706 La Tronche Cedex, France.
| | | | | |
Collapse
|
12
|
Popescu M, Mahnot A. Acoustic fall detection using one-class classifiers. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2009; 2009:3505-3508. [PMID: 19964801 DOI: 10.1109/iembs.2009.5334521] [Citation(s) in RCA: 16] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/28/2023]
Abstract
Falling represents a major health concern for the elderly. To address this concern we proposed in a previous paper an acoustic fall detection system, FADE, composed of a microphone array and a motion detector. FADE may help the elderly living alone by alerting a caregiver as soon as a fall is detected. A crucial component of FADE is the classification software that labels an event as a fall or part of the daily routine based on its sound signature. A major challenge in the design of the classifier is that it is almost impossible to obtain realistic fall sound signatures for training purposes. To address this problem we investigate a type of classifier, one-class classifier, that requires only examples from one class (i.e., non-fall sounds) for training. In our experiments we used three one-class (OC) classifiers: nearest neighbor (OCNN), SVM (OCSVM) and Gaussian mixture (OCGM). We compared the results of OC to the regular (two-class) classifiers on two datasets.
Collapse
Affiliation(s)
- Mihail Popescu
- Health Management and Informatics Department, University of Missouri, Columbia, MO 65211, USA.
| | | |
Collapse
|