The present study introduces a method for improving the classification performance of imbalanced ... more The present study introduces a method for improving the classification performance of imbalanced multiclass data streams from wireless body worn sensors. Data imbalance is an inherent problem in activity recognition caused by the irregular time distribution of activities, which are sequential and dependent on previous movements. We use conditional random fields (CRF), a graphical model for structured classification, to take advantage of dependencies between activities in a sequence. However, CRFs do not consider the negative effects of class imbalance during training. We propose a class-wise dynamically weighted CRF (dWCRF) where weights are automatically determined during training by maximizing the expected overall F-score. Our results based on three case studies from a healthcare application using a batteryless body worn sensor, demonstrate that our method, in general, improves overall and minority class F-score when compared to other CRF based classifiers and achieves similar or ...
Visvanathan R ,, Ranasinghe DC3, Mahajan N2, Shinmoto-Torrest R3, Stewart L1,4, Khoo A1,2, Gentil... more Visvanathan R ,, Ranasinghe DC3, Mahajan N2, Shinmoto-Torrest R3, Stewart L1,4, Khoo A1,2, Gentilcore D4, Hoskins S1, Tremaine M1, Hill K5 Aged & Extended Care Services, Division of Medicine & The Health Observatory, the Queen Elizabeth Hospital, SA, AUSTRALIA Department of Medicine, University of Adelaide, SA, AUSTRALIA Auto-ID Lab, Faculty of Engineering, Computer and Mathematical Sciences, University of Adelaide, SA, AUSTRALIA Department of Nuclear Medicine, University of South Australia, SA, AUSTRALIA 5 School of Physiotherapy, Curtin University, WA, AUSTRALIA
Falls in hospitals are common, therefore strategies to minimize the impact of these events in old... more Falls in hospitals are common, therefore strategies to minimize the impact of these events in older patients and needs to be examined. In this pilot study, we investigate a movement monitoring sensor system for identifying bed and chair exits using a wireless wearable sensor worn by hospitalized older patients. We developed a movement monitoring sensor system that recognizes bed and chair exits. The system consists of a machine learning based activity classifier and a bed and chair exit recognition process based on an activity score function. Twenty-six patients, aged 71 to 93 years old, hospitalized in the Geriatric Evaluation and Management Unit participated in the supervised trials. They wore over their attire a battery-less, lightweight and wireless sensor and performed scripted activities such as getting off the bed and chair. We investigated the system performance in recognizing bed and chair exits in hospital rooms where RFID antennas and readers were in place. The system'...
Aging populations are increasing worldwide and strategies to minimize the impact of falls on olde... more Aging populations are increasing worldwide and strategies to minimize the impact of falls on older people need to be examined. Falls in hospitals are common and current hospital technological implementations use localized sensors on beds and chairs to alert caregivers of unsupervised patient ambulations; however, such systems have high false alarm rates. We investigate the recognition of bed and chair exits in real-time using a wireless wearable sensor worn by healthy older volunteers. Fourteen healthy older participants joined in supervised trials. They wore a batteryless, lightweight and wireless sensor over their attire and performed a set of broadly scripted activities. We developed a movement monitoring approach for the recognition of bed and chair exits based on a machine learning activity predictor. We investigated the effectiveness of our approach in generating bed and chair exit alerts in two possible clinical deployments (Room 1 and Room 2). The system obtained recall resu...
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2012
Falls related injuries among elderly patients in hospitals or residents in residential care facil... more Falls related injuries among elderly patients in hospitals or residents in residential care facilities is a significant problem that causes emotional and physical trauma to those involved while presenting a rising healthcare expense in countries such as Australia where the population is ageing. Novel approaches using low cost and privacy preserving sensor enabled Radio Frequency Identification (RFID) technology may have the potential to provide a low cost and effective technological intervention to prevent falls in hospitals. We outline the details of a wearable sensor enabled RFID tag that is battery free, low cost, lightweight, maintenance free and can be worn continuously for automatic and unsupervised remote monitoring of activities of frail patients at acute hospitals or residents in residential care. The technological developments outlined in the paper forms part of an overall technological intervention developed to reduce falls at acute hospitals or in residential care facili...
The present study introduces a method for improving the classification performance of imbalanced ... more The present study introduces a method for improving the classification performance of imbalanced multiclass data streams from wireless body worn sensors. Data imbalance is an inherent problem in activity recognition caused by the irregular time distribution of activities, which are sequential and dependent on previous movements. We use conditional random fields (CRF), a graphical model for structured classification, to take advantage of dependencies between activities in a sequence. However, CRFs do not consider the negative effects of class imbalance during training. We propose a class-wise dynamically weighted CRF (dWCRF) where weights are automatically determined during training by maximizing the expected overall F-score. Our results based on three case studies from a healthcare application using a batteryless body worn sensor, demonstrate that our method, in general, improves overall and minority class F-score when compared to other CRF based classifiers and achieves similar or ...
Visvanathan R ,, Ranasinghe DC3, Mahajan N2, Shinmoto-Torrest R3, Stewart L1,4, Khoo A1,2, Gentil... more Visvanathan R ,, Ranasinghe DC3, Mahajan N2, Shinmoto-Torrest R3, Stewart L1,4, Khoo A1,2, Gentilcore D4, Hoskins S1, Tremaine M1, Hill K5 Aged & Extended Care Services, Division of Medicine & The Health Observatory, the Queen Elizabeth Hospital, SA, AUSTRALIA Department of Medicine, University of Adelaide, SA, AUSTRALIA Auto-ID Lab, Faculty of Engineering, Computer and Mathematical Sciences, University of Adelaide, SA, AUSTRALIA Department of Nuclear Medicine, University of South Australia, SA, AUSTRALIA 5 School of Physiotherapy, Curtin University, WA, AUSTRALIA
Falls in hospitals are common, therefore strategies to minimize the impact of these events in old... more Falls in hospitals are common, therefore strategies to minimize the impact of these events in older patients and needs to be examined. In this pilot study, we investigate a movement monitoring sensor system for identifying bed and chair exits using a wireless wearable sensor worn by hospitalized older patients. We developed a movement monitoring sensor system that recognizes bed and chair exits. The system consists of a machine learning based activity classifier and a bed and chair exit recognition process based on an activity score function. Twenty-six patients, aged 71 to 93 years old, hospitalized in the Geriatric Evaluation and Management Unit participated in the supervised trials. They wore over their attire a battery-less, lightweight and wireless sensor and performed scripted activities such as getting off the bed and chair. We investigated the system performance in recognizing bed and chair exits in hospital rooms where RFID antennas and readers were in place. The system'...
Aging populations are increasing worldwide and strategies to minimize the impact of falls on olde... more Aging populations are increasing worldwide and strategies to minimize the impact of falls on older people need to be examined. Falls in hospitals are common and current hospital technological implementations use localized sensors on beds and chairs to alert caregivers of unsupervised patient ambulations; however, such systems have high false alarm rates. We investigate the recognition of bed and chair exits in real-time using a wireless wearable sensor worn by healthy older volunteers. Fourteen healthy older participants joined in supervised trials. They wore a batteryless, lightweight and wireless sensor over their attire and performed a set of broadly scripted activities. We developed a movement monitoring approach for the recognition of bed and chair exits based on a machine learning activity predictor. We investigated the effectiveness of our approach in generating bed and chair exit alerts in two possible clinical deployments (Room 1 and Room 2). The system obtained recall resu...
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2012
Falls related injuries among elderly patients in hospitals or residents in residential care facil... more Falls related injuries among elderly patients in hospitals or residents in residential care facilities is a significant problem that causes emotional and physical trauma to those involved while presenting a rising healthcare expense in countries such as Australia where the population is ageing. Novel approaches using low cost and privacy preserving sensor enabled Radio Frequency Identification (RFID) technology may have the potential to provide a low cost and effective technological intervention to prevent falls in hospitals. We outline the details of a wearable sensor enabled RFID tag that is battery free, low cost, lightweight, maintenance free and can be worn continuously for automatic and unsupervised remote monitoring of activities of frail patients at acute hospitals or residents in residential care. The technological developments outlined in the paper forms part of an overall technological intervention developed to reduce falls at acute hospitals or in residential care facili...
Uploads