Can we make a carpet smart enough to detect falls?

In this paper, we have enhanced smart carpet, which is a floor based personnel detector system, to detect falls using a faster but low cost processor. Our hardware front end reads 128 sensors, with sensors output a voltage due to a person walking or falling on the carpet. The processor is Jetson TK1, which provides more computing power than before. We generated a dataset with volunteers who walked and fell to test our algorithms. Data obtained allowed examining data frames (a frame is a single scan of the carpet sensors) read from the data acquisition system. We used different algorithms and techniques, and varied the windows size of number of frames (WS ≥ 1) and threshold (TH) to build our data set, which later used machine learning to help decide a fall or no fall. We then used the dataset obtained from applying a set of fall detection algorithms and the video recorded for the fall pattern experiments to train a set of classifiers using multiple test options using the Weka framework. We measured the sensitivity and specificity of the system and other metrics for intelligent detection of falls. Results showed that Computational Intelligence techniques detect falls with 96.2% accuracy and 81% sensitivity and 97.8% specificity. In addition to fall detection, we developed a database system and web applications to retain these data for years. We can display this data in realtime and for all activities in the carpet for extensive data analysis any time in the future.

Faux-floor development system for personnel detection using signal scavenging sensors.

Motivated by the need to detect motion in elderly people, resulting in falls, we have developed a low cost sensor system using aluminum foil as the sensor of static electricity and electromagnetic energy. But to make this a system we need to amplify the data and use it by displaying the motion or activity of a person. We constructed a faux floor development board to provide an initial pilot test of the idea of using stray electric energy, or as we call it signal scavenging. The foils are placed on the faux floor (in this case a 1 m X 1 m wooden surface) allowing foil excitation from the motion of a test subject. The faux floor is a useful tool allowing testing of different foils, analog and digital electronics circuits and different carpeting. Importantly, even though the system supported a small number of foil sensors its performance characteristics clearly show the excellent detection capability of the system. Testing the timing characteristics resulted in reading the 4 sensors in 3.11 msec, indicating that for even a large system of a few hundred sensors we can poll the foils in sufficient time to detect the motion of people. Our data show true positive rates of 98% and false positive and false negative rates of 2%, a high detection rate. Using the development board has provided much helpful information on the use of signal scavenging for personnel detection.

Inkjet printed arrays of pressure sensors based on all-organic field effect transistors.

In this paper we propose totally flexible organic field effect transistors (OFETs) assembled on plastic films as sensors for mechanical variables. First mechanical sensors for pressure and bending detection are presented. A sharp and reversible sensitivity of the output current of the device to an elastic deformation induced by means of a mechanical stimulus on the device channel has been observed and suggested the idea of employing arrays of such sensors for detecting the deformation applied onto a planar surface. Second the possibility of using similar devices for bio- and chemo-detection is described. By exploiting the properties of the basic structure, the device can be combined with any kind of substrate to detect for instance the pressure applied by people walking or standing on a functionalized carpet. This emerging technology seems to be promising for applications in the field of remote and non invasive monitoring of elderly and disabled people.

Smart Carpet: Developing a sensor system to detect falls and summon assistance.

Cognitive deficits experienced by older adults with dementia limit use of wearable devices (necklaces or bracelets) that summon assistance after the older adult falls. To use these wearable devices, older adults must choose to wear them, remember how to use them, and be conscious after falling. Devices such as the Smart Carpet substitute pre-programmed or automatic functions for functions requiring deliberation and decision. After development of a Smart Carpet prototype, 11 volunteers participated in tests to measure sensitivity of sensors embedded in the Smart Carpet. The embedded sensors were not perceptible to the volunteers as they walked across the Smart Carpet and successfully detected gait characteristics. Findings confirmed the feasibility of fall detection. Measurements obtained of gait characteristics will be used in development of more advanced versions of the Smart Carpet.

Signal scavenging for passive monitoring in eldercare technology.

Signal scavenging is analogous to energy scavenging: seemingly ubiquitous energy in the environs provides the signal for usage as a personnel sensor. Such energy can be used to detect motion, and most importantly falls. Stray signals can be detected in aluminum foil as voltage differences between touched foil (say by hand) compared to that untouched. Spectrum analysis shows the stray electromagnetic noise signal consists substantially of 60 Hz and its harmonics. Also the signal intensity for both touched and untouched monotonically increases with foil area. While personnel monitors find utility in many areas including security, personnel control and activity detection, we believe these putative sensors to be useful in inobtrusive monitoring of elders to provide them with increased independence at a critical time in their lives.

Arrays of pressure sensors based on organic field effect: a new perspective for non invasive monitoring.

In this paper we propose totally flexible organic field effect transistors (OFETs) assembled on plastic films as sensors for mechanical variables. In the first part, mechanical sensors for pressure and bending detection are presented. A sharp and reversible sensitivity of the output current of the device to an elastic deformation induced by means of a mechanical stimulus on the device channel has been observed and suggested the idea of employing arrays of such sensors for detecting the deformation applied onto a planar surface. In the second part, the possibility of using similar devices for bio- and chemo-detection is described. By exploiting the properties of the basic structure, the device can be combined with any kind of substrate to detect for instance the pressure applied by people walking or standing on a functionalized carpet. This emerging technology seems to be promising for applications in the field of remote and non invasive monitoring of elderly and disabled people.

Early detection of health changes in older adults.

We have placed a network of sensors in a residential home for the elderly who are aging in place. Restlessness data is displayed as graph of event counts detected by sensors over some time interval, typically a day. This data is related to the actual activities as recorded by the resident. We show two cases of elderly individuals. In both cases the individuals underwent surgery. The restlessness indicators showed changes in patterns that were related to those events. Analyzing the data even at this level we gain increased confidence that technology will be a welcome addition as the population ages and require increasing care.

Facilitating interdisciplinary design specification of "smart" homes for aging in place.

"Smart homes" are defined as residences equipped with sensors and other advanced technology applications that enhance residents' independence and can be used for aging in place. The objective of this study is to determine design specifications for smart residences as defined by professional groups involved both in care delivery to senior citizens and development of devices and technologies to support aging. We assessed the importance of specific devices and sensors and their advantages and disadvantages as perceived by the interdisciplinary expert team. This work lays the ground for the implementation of smart home residencies and confirms that only an interdisciplinary design approach can address all the technical, clinical and human factors related challenges associated with home-based technologies that support aging. Our findings indicate that the use of adaptive technology that can be installed in the home environment has the potential to not only support but also empower individual senior users.

Technology for successful aging.

Americans are living longer, and research shows that seniors are embracing independence, and will benefit from living in the same place. These are the rationale for 'Aging in Place' and the development of Tiger Place, an 'Aging in Place' Environment in Columbia Missouri. Our goal is to minimize intrusion, allow the resident complete control over privacy and treatment (if any), and to provide substantive improvement in quality of life. Nevertheless there continues to be significant risks to the elderly which results in reduced functional and cognitive activity. While there has been much technology developed to ameliorate these factors, there is no comprehensive evaluation of the benefit of these devices nor a comprehensive strategy to improve the quality of life of seniors as determined by functional ability and possibly later cognitive ability. With our partners at the University of Virginia we are developing a system of sensors, to monitor the activity of seniors in their residences. We measure motion, footfalls, sleep and restlessness, we have stove sensors and sensing mats, all connected wirelessly to a computer which performs an initial evaluation and data transfer to a secure server for further study. Based upon the monitor data we will implement an intervention to ameliorate functional decline. Focus group studies determine the attitudes, concerns and impressions of the residents and staff. We find that senior's attitude to technology is healthy and they will try helpful approaches. In addition to the statistical comparisons, we model the data using hidden Markov models, integrate or fuse the monitor data with video images, and reason about behavior using fuzzy logic. The results of this work will additionally reduce the workload on caregivers, foster communication between residents and family, and give these seniors independence. We have requested and received IRB approval for this study.

A technology and nursing collaboration to help older adults age in place.

This is an account of an active collaboration between Computer Engineering, Health Informatics, and Nursing within an academic health science center to improve the quality of life of older adults as they "age in place." The Sinclair School of Nursing at the University of Missouri-Columbia has developed a licensed home health agency, Senior Care, to provide the care needed by residents of TigerPlace, a specially designed independent living center near the University. Technology has the potential to help address common problems encountered by older adults related to functional decline. Collaboration between Nursing, Computer Engineering, and Health Informatics is likely on a path to improve the quality of life of seniors.

Home-based assistive technologies for elderly: attitudes and perceptions.

This study aim is to explore the perceptions of seniors in regard to "smart home" technology aiming to improve their quality of life and/or monitor their health status. A total of 15 older adults participated in three focus groups. Participants had a positive attitude towards these technologies and identified application areas such as emergency help, detection of falls, monitoring of physiological parameters. Concerns were expressed about privacy and the need for tailored training.