Sensewear software

For example, a technology approved with a decision for 7. Anything - 7. However, a 7. Anything , but would not cover any version of 7. For example, a technology approved with a decision for Section of the Rehabilitation Act Amendments of is a federal law that sets the guidelines for technology accessibility.

A VA Section assessment of this technology has not been completed at the time of publication. Therefore, as of April 23, only users of this technology who have deployed the technology to the production environment, or have project design and implementation plans approved, may continue to operate this technology. In the case of a project that has implemented, or been approved for a specific site or number of users, and that project needs to expand operations to other sites or to an increased user base, it may do so as long as the project stays on the existing version of the technology that was approved or implemented as of April 22, Use of this technology in all other cases is prohibited.

References The following reference s are associated with this entry: Type Name Source Description There are no references identified for this entry. Technology Components Note: This list may not be complete. No component, listed or unlisted, may be used outside of the technology in which it is released.

The usage decision for a component is found in the Decision and Decision Constraints. Name Description No components have been identified for this entry. Runtime Dependencies: No runtime dependencies have been identified. General Analysis Adoption Benefits There are no known security vulnerabilities associated with this technology at the time of writing.

SenseWear Software. Go to site. SenseWear Software provides the ability to collect and store physiological and lifestyle data blood pressure, blood glucose, and weight captured by a SenseWear Armband and transfer that data to a workstation.

Section does not apply. The Vendor Release table provides the known releases for the TRM Technology, obtained from the vendor or from the release source.

Decision Constraints. The requirements on the FOVEA system were determined using our scenario-based requirements elicitation methodology [ 4 ], an extension of the work of Benyon and Macaulay [ 5 ]. The full list of requirements is not reproduced here. The requirements relating to monitoring of dietary and exercise behaviour of individual users include.

The SenseWear sensor was tested for suitability for the project with respect to requirements iii — vi above. Requirement iii is of interest here mainly for its role in the estimation of EE requirement iv , Requirement v implies that all the raw data from the relevant sensors should be exportable.

The format is immaterial, so long as it is known. Requirement vi is prerequisite for provision of real-time user feedback assuming analysis and interpretation algorithms can be run in near real time. With respect to requirement iv , at the time the SenseWear BMS sensor was the only commercially available sensor we could find that provided a good estimate of daily EE compared to the gold standard of the doubly labelled water method DLW.

In one evaluation [ 6 ], the SenseWear sensor was compared with the DLW method in 45 subjects over a day period. The armband may therefore be useful to estimate daily EE. In a previous study [ 8 ], researchers at Wageningen University had good experience of the SenseWear system used in combination with a heart rate monitor to study the effects of ambient aromas on a number of physiological and behavioural factors. Results demonstrated that different ambient aromas, even at barely detectable concentrations, are associated with differential and selective effects relating to EE and food choice and even to autonomic physiological function heart rate.

Furthermore, as well as including analysis software capable of outputting a range of graphical views and reports, data from the sensor could be uploaded for further processing partly addressing requirement v above. The evaluation described here was conducted in an early phase of FOVEA; later phases were concerned with architecture and high-level design and following derivation of technical requirements and detailed design prototyping of the FOVEA system, including both mobile and fixed parts.

The Restaurant of the Future provides an instrumented environment which is used in this and other projects as a testbed for interactive research in a real-life setting. The RoF infrastructure includes steerable video cameras and stereo video cameras for behavioural observation, weight scales at the checkouts, and automatic registration at point of sale terminal of individual food and drink consumptions as well as offering possibilities for altering the ambient environment in order to investigate effects of subtle changes in environmental factors on physiology and behaviour.

The FOVEA system, currently under development, integrates components from the RoF infrastructure with the food database from the canteen supplier.

In addition, specific new applications are being developed and integrated, including a tracking system for position determination and for analysis and verification of spatiotemporal behaviour patterns of users in the restaurant and a mobile application for real-time monitoring and personalized feedback to users.

The University of Twente, with experience in developing mobile monitoring and feedback systems based on Body Area Network technology, is responsible for developing the mobile part of the FOVEA system.

The following section gives some background on the research at Twente into remote monitoring and feedback using mobile and wireless technologies. A multidisciplinary team of computer scientists, clinicians, and biomedical engineers at the University of Twente in the Netherlands has been researching mobile monitoring and feedback systems based on Body Area Networks BANs since The Twente BAN system and various healthcare applications are reported, for example, in [ 9 — 16 ].

Two potential health and wellbeing applications involving monitoring in extreme environments are described in [ 17 ]. Our definition of a health BAN is a network of communicating devices e. For telemonitoring applications, the patient wears a BAN equipped with biosensors and possibly other devices e. The BAN data may be processed by humans, automatically, or a combination of the two, depending on the requirements of the specific application.

For example, a remote healthcare professional can view a multimedia display including graphical and numerical representation of multiple biosignals and other measurements of the patient and their environment, or selectable subsets of the same kind either in real time or stored.

By including a feedback loop and actuation as well as sensing, monitoring services can be augmented with feedback and control enabling teletreatment services. Such services, especially when automated or semiautomated, require accurate and reliable processing, transmission, and interpretation of the output of multiple biosignal sources in combination with context sources which may include visual, auditory, text, and other types of information.

In various BAN applications, we have combined output from multiple sensors and context sources and delivered feedback and treatment to the patient via multiple modalities including tactile, text, auditory signals, and images. During MobiHealth, an m-health service platform and a number of variants of the health BAN, equipped with different sensor sets, were trialled in four European countries with various biosignals monitored and transmitted to remote healthcare centres over GPRS and UMTS.

The nine trials in MobiHealth included telemonitoring for cardiology and COPD respiratory insufficiency patients, for pregnant women, for casualties in trauma care, and a professional ABN for ambulance paramedics.

Awareness focussed on neurology applications epilepsy, spasticity, and chronic pain and addressed the issue of adding context awareness to BAN applications.

In Awareness teletreatment, services were introduced alongside telemonitoring services. Over the course of these projects, we gained experience of signal processing and interpreting the output from different combinations of sensors and other devices. Sensors which have been integrated into the health BAN to date include electrodes for measuring ECG and EMG, pulse oxymeter, various motion sensors step counter, 3D accelerometer , temperature, and respiration sensors.

Other devices which have been incorporated into the BAN include positioning devices and a multimodal biofeedback device which measures surface EMG and gives feedback in the form of vibration and auditory signals.

In this case, the biofeedback device which was incorporated into the BAN could also operate as a standalone device and was already available in this capacity as a commercial product.

With FOVEA, the work in the health and wellbeing domain was extended to providing ambulatory monitoring and personalised user feedback in the weight management application. The Fovea mobile application currently under development is designed to register food and drink selections and physical activity and provide feedback and advice tailored to the individual's specific weight management goals.

The application will enable the user to choose, amongst the available lunch items in a restaurant, the ones that are in line with his dietary plan, as elaborated with a nutritionist, and his daily energy budget.

The user's physical activity is monitored throughout the day, and energy expenditure is estimated in Kcal. This information is then used to calculate the energy budget, taking also into account the energy intake, which the user can spend on food items during lunch.

The system also incorporates a Bluetooth beacon discovery process which enables the user to locate his office, restaurants, buffets, and weight scales within a workplace. Once a restaurant is discovered and selected, its floor plan is displayed to the user. This floor plan highlights those buffets which contain items that are compliant with the individual's preselected healthy lunch compositions.

Once a given buffet is selected, all the lunch items it contains are displayed on the smart phone screen, with those which are not compliant with his healthy lunch composition highlighted, but leaving the decision ultimately to the user. When a food item is selected, its detailed information can be visualized on the smart phone, so that an informed decision can be made.

If a lunch item is then confirmed, feedback is given to the user according to its impact in the available energy budget, and the lunch item information is stored in the user's food diary.

It is intended for ambulatory use by patients, in consultation with their physicians, for monitoring and assessing activity levels and sleep patterns. According to [ 22 ], the sensors are 2-axis accelerometer, a heat flux sensor, a galvanic skin response sensor, a skin temperature sensor, and a near-body ambient temperature sensor.

Figure 1 shows the SenseWear device, and Figure 2 shows the rear view showing the arrangement of sensors. Following configuration, the system starts up automatically when the user puts it on and only when the sensors have made a secure contact with the skin. The user can press the button to register a time stamp at any time.

According to the manufacturer's website, the most important derived parameters calculated or estimated from data gathered by the sensors are total energy expenditure in calories, active energy expenditure, physical activity duration and levels measured in metabolic equivalents METs , and sleep duration and efficiency.

After wearing the system and registering sensor data, the data can be uploaded to a PC. Optionally all data can be wiped from the device after upload. Hence, the device can be used subsequently by a different patient and possibly a different clinician. The device gives warning of battery failure and of memory full. Analysis software on a dongle license key enforcing single PC use at any one time per unit and a USB cable are included with the system. The proprietary analysis software that accompanies the device can be run on a PC to perform various data analyses and produce on screen visualizations and reports.

The reports can be saved to pdf files. There is one software application for the user the SenseWear software , which enables the user to configure the system and to save and retrieve physiological and lifestyle data collected by the device. Another software application for use by the clinician the SenseWear Professional software additionally organizes the data and generates reports and visualizations in the form of graphs and permits export of data for further analysis in the form of Excel spreadsheets.

The SenseWear device was tested by volunteers from the University of Twente acting as test users. The device was worn during sleep and removed only when taking a shower. Subject 3 kept a log of activities during the trial. Of greatest interest in the context of the FOVEA project is EE expressed in calories and metabolic equivalents total and average METs and the duration and intensity of physical activity. Activity levels can be displayed graphically as well as numerically.

Following our trial, we uploaded the data, performed various explorations using the professional software, and captured screenshots of the visualizations as described in the following sections.

Figure 3 , shows a display of selected parameters over a period of one week. Any arbitrary period and any combination of parameters can be selected for display. At the right hand side of Figure 3 all the sensors, and the parameters that can be derived from them, can be seen.

In the on-screen visualizations, the selected parameters are superimposed and displayed on a timeline. Physical activity is measured by a 2-axis accelerometer. Total EE includes a correction for off-body time. Figures 4 a and 4 b , respectively, reproduce pages 1 and 2 of the pdf report corresponding to the visualization shown in Figure 3. The parameters selected on the visualization also appear separately and graphically in the pdf report.

The nine graphs in Figure 4 b correspond to the selection of nine parameters made on that occasion see Figure 3. Figure 5 shows the pattern over a single day the first 24 hours of use for subject 3. Pattern over one day note activity levels around Note activity levels around Subject 3 is a chronic insomniac. Sleep patterns shown in Figure 5 correspond closely to the periods of sleep and wakefulness reported in the subject log.

The user log records a short low-intensity exercise session around We return to the driving episode in the discussion section below.

Figures 6 a and 6 b , respectively, reproduce pages 1 and 2 of the report covering a period of one day and correspond to Figure 5. The first page of the report see Figure 6 a gives summary information on clinician and hospital, patient data, timing and duration of usage on body, and data on EE and sleep over the selected period the first 24 hours of use.

Because the user test for subject 3 started at Over the hour period split over two partial days , total EE was calories. Total step count over the 24 hours was The second page of the report reproduced in Figure 6 b gives more detailed graphical representations of the selected parameters cf Figure 5. Figure 6 b also shows that spikes in EE are sometimes echoed in the GSR trace, which would be expected due to increased transpiration.

Average skin temperature rises during the afternoon and evening and peaks during the night. Skin temperature and heat flux sometimes show an apparently inverse relationship in the detail between 12 midnight and 12 noon. Figure 7 shows the pattern over the whole test period of eight days for subject 3, allowing diurnal patterns to be compared visually over a period of days.

The armband was worn The red trace shows average skin temperature; the white trace shows average galvanic skin response; the blue trace shows energy expenditure. Figure 8 shows the pattern over another single day. The red trace shows average skin temperature; the white trace shows galvanic skin response average; the blue trace shows energy expenditure; the green trace shows heat flux average. The activity pattern around Four kinds of data are included.

Table 1 shows an extract of the timestamped pre-processed sensor data for subject 3. The data is recorded once per minute. The table is split into two in order to fit on the page. The following information can be entered when the user configures the device:. Table 2 shows the entire summary sheet for all the data for subject 3 8 days. The first day was a short day with the experiment starting on day 1 at Similarly, the last day was also a short day as the experiment ended at It can be seen that the subject was able to wear the device up to Hours off body on the 5th of September corresponds to the entry on the log where the subject removed the device when attending a party so as not to attract attention.

The software compensates for time off body by estimating EE during that time. Obviously, the four kinds of data from the spreadsheets can easily be input to other applications, however, only retrospectively in the case of the system tested.

Moreover, the raw data is not available. These two facts limit the potential utility of the device if the ambition is to perform real-time analysis and provide real time feedback to users, as is the intention in the FOVEA project.

The simple and intuitive interface, using a single button, vibratory and auditory feedback two different patterns of beeps , and two lights, was considered to be easy to use and well designed by the standard expected for quality consumer electronics products. However, after two days' continuous wear apart from in the shower , subject 3 found that friction from the edge of the Velcro armband caused skin damage serious enough to stop the experiment.

In the event, the user found a makeshift solution by cutting a strip from a compression bandage and inserting it between skin and Velcro armband, carefully avoiding interfering with the sensor-skin contact. The result was effective if not aesthetically pleasing! Exercise levels and sleep patterns e. User experience as reported was positive with the exception of the skin problem caused by chafing of the Velcro armband. As described above, a work around was easily found, so that the experiment could continue.

The proprietary algorithms used by the SenseWear software are not in the public domain; hence, it is difficult to tell exactly how energy expenditure and other derived parameters are calculated or estimated. A number of evaluation studies have been conducted on different models in the products range.

One issue raised is the point that the accuracy of energy expenditure estimations appears to be affected by many different parameters. One study examined the validity of the SenseWear Pro 2 armband to assess energy expenditure during various modes of physical activity in twenty-four healthy female and male adolescents.

Energy expenditure measured by the armband during treadmill at different speeds and gradients and cycle ergometer exercise was compared with respiratory metabolic system RMS. Other studies, for example, [ 27 , 28 ] arrived at favourable conclusions. SenseWear subsequently developed age-specific algorithms.

A comparison of three products was made in [ 29 ], concluding that all three needed further development but that the SenseWear device might be more feasible for use under free-living conditions, although it was less accurate than another product the IDEEA device in assessing energy cost. In an earlier study in [ 30 ], the SenseWear armband's EE estimates were compared with IC indirect calorimetry in adults during rest and exercise.

In , a comparison of the energy expenditure estimates of the SenseWear Pro 2 against the IC method on adults during rest and during three exercise sessions was conducted [ 23 ].

In this study, the experimental group consisted of obese adults; the controls were lean and overweight adults. The REE resting energy expenditure estimated by the SenseWear device showed high correlation and very good agreement with measured IC in lean and overweight adults but showed poor accuracy in obese adults, especially those with high REE, in rest and in exercise.

This is not so serious for the FOVEA project since overweight subjects will be included, but clinically obese subjects will be excluded. According to the Instructions For Use [ 31 ], the device should be worn on the back of the upper right arm.

We presume that lower activity levels are expected in the nondominant arm. There is an option to record handedness in the subject information although this is not enforced. It would seem then that activity levels will be underestimated for left-handed users, and interindividual comparisons would need to correct somehow for handedness.

We cannot see any way to interface the model tested to a PDA or smartphone; hence, it appears that it would be difficult or impossible to incorporate it into a body area network. It could be used in this way, for example, to show a subject's recent history of energy expenditure and compare that with current energy intake if the subject uploads data regularly, for example, once per day. The model tested by us, however, has no real-time data transmission for upload of data, so the user must perform periodic uploads to a PC via a USB connection.

Another drawback in the model tested is that although the data can be uploaded to a PC, the raw data is not accessible and only pre-processed data is uploaded see Table 1.