Overview
The Microlife WatchBP is a vital signs (VS) device that allows for up to 6 measurements with a given acquisition. The ClinSpark Agent implementation allows for it to be leveraged through the browser, or via patient home by way of a locally stored token.
Within ClinSpark this device is represented in Devices > Manage as microlife BP; model WatchBP.
Capabilities / Customization
The devices can be programmed with a protocol driven form, and the settings can be established by way of setting default item values for a given item group item reference
Measurement count - the number of vitals signs tests to execute; defaults to 1
Rest time - the time to wait until a vitals test is run; defaults to 5 seconds
Blinding - should the measurements be hidden from the user / subject; defaults to no
Interval time - the time to wait in seconds between vitals tests; is only relevant if more than 1 measurements are to be acquired; defaults to 15 seconds
Exclude first measurement in average - if more than 2 measurements are to be taken, should the first measurement be ignored?
Device Parameters
Because the device allows for up to six measurements, the interface must ensure that all six measurements can be returned. In the case where more than one measurement is to be returned, the non-counter parameters will be the average.
Date - the date returned by the firmware for a given test. Only supports minute-level resolution, and should not be used when obtaining multiple measurements.
Systolic - the Systolic value when only a single test is being done, or the average Systolic when multiple
Diastolic - the Diastolic value when only a single test is being done, or the average Diastolic when multiple
MAP - the MAP value when only a single test is being done, or the average MAP when multiple
Rate - the Rate value when only a single test is being done, or the average Rate when multiple
Systolic1-6, Diastolic1-6, MAP1-6, Rate1-6 - the relevant parameters if multiple tests are to be performed
DEVICEID - unique identifier stored in the device; should be used with enrollment and as the monitoring source when configuring for remote patient uploads
VSCNTDOWN - time in seconds to wait before the initial test is to be performed after invoking the device; this should be configured as a static item reference value with one of the following: 15, 30, 60, 120, 180, 240, or 300
VSRCRDNUM - the number of tests to perform; if more than 1 it should be configured as a static item reference with a maximum value of 6
VSRCRDINTVL - the amount of time to wait in between tests; it should be configured as a static item reference with one of the following: 15, 30, 60, 120, 180, 240, or 300
VSBLIND - should the results be blinded from the user?
VSEXCLUDEFRST - if performing more than 2 tests, should the first test be excluded when an average is calculated
Settings
All settings are optional and are not required to be configured. They are only described here to over-ride the sensible defaults.
Note: when doing an enrollment, the settings are retained in the token and read accordingly when doing remote patient connectivity.
LeftImageEncoded - base64 encoded image for the remote workflow customization of the user interface
RightImageEncoded - base64 encoded image for the remote workflow customization of the user interface
ConnectionTimeoutSeconds - the maximum number of seconds to to wait before giving up on connecting to the device. Defaults to 60.
PollerDelaySeconds - when successfully performing an enrollment, the number of seconds to wait until the first poll cycle occurs. The idea with this setting is that a site user will perform the enrollment on a patient computer and that a sensible delay should occur until next looking for data. Defaults to 300 seconds.
ReadDeviceIdAttempts - the maximum number of attempts to try and read the device ID upon acquisition. When not found, the interface will sleep 200 ms before trying again. Defaults to 5.
RequireDeviceId - if after attempting to obtain the device ID, should the interface block data collection if the device ID is not found? Defaults to false.
DeviceCommunicationTimeoutSeconds - during an acquisition, the device constantly communicates back to the ClinSpark Agent with somewhat of a ping. This communication does not occur during the inflate / wait process. This setting defines maximum number of seconds to wait after a programmatic inflate of the cuff occurs and no asynchronous communication has occurred from the device. This is a very low-level configuration option. Defaults to 180 seconds.
CancelMinutes - when a patient clicks ‘Cancel’ through the remote upload user interface, this is the (approximate) number of minutes to wait until the next poll cycle runs. The purpose of this delay is to potentially allow a home health nurse to use the same patient computer to perform an ‘office’ acquisition fear of a connectivity collision (ie connecting via browser and concurrently with the remote upload). Defaults to 10.
EraseWhenHide - the device supports a blinding mechanism, but by default the results are no longer blinded after the last test is performed. By default, the interface will initialize an ‘erase’ command immediately to force blinding. If this value is set to ‘No’, the erase command will not take place.
Execution
The interface is designed to return the full set of vital sign test data as defined by the protocol form ‘measurement count’. If any of the measurement counts are not returned successfully, communication with the device will end and the entire test set is disregarded.
For example, if the measurement count is 4, but only 2/4 tests succeed, then none of the measurement counts are captured. It’s important to ensure that the device is properly used for every test for the full set of measurements to successfully complete.
Browser Workflow
Upon clicking ‘Invoke Device’ - user is given brief message to observe the device during the test execution
User clicks ‘Ok’
DLLs are loaded and initial communication with the device takes place
If the device is not available, the user is given a relevant message
The device ID is read from the firmware
Device is erased to ensure no prior data will be returned
Using the time of the data collection computer, the time is stored in the device
Based on the static parameters defined prior, the ‘settings’ / ‘protocol' are written to the device
An API call is then issued to initiate the protocol with the configured settings
The Agent then waits until all of the measurements are obtained, or the timeout occurs
For each measurement, the API returns a ‘Condition’ setting - only measurements with a 0 or 1 are leveraged. Definitions from the API are as follows:
0:Original BP only
1:Original BP and CBP both
2:Empty battery
3:Error code for original BP
4:Stop measurement by manual
5:Pill memo
6:Device switch off
7:Re-start during a 24h profile
Enrollment
Once the device ID has been obtained, the token and settings are stored in the ‘AppData/Roaming’ directory associated with the user’s account in their operating system settings
The poller is then scheduled to start execution within 300 seconds
Token Management
If the PC has been enrolled into a study for remote upload workflows, the MicrolifeBpDevice token will persist on users local machines through Agent uninstall/reinstall. This is by design, to support Agent upgrade efforts without losing the ability to maintain remove upload workflows for that enrolled subject.
Persistent tokens can be removed from users local PC through these scenarios:
Through form interactions in ClinSpark, the Agent re-enrolls the PC into a site-based upload workflow
A local PC user manually intervenes and removes the token file directly from the AppData/Roaming directory
Dependencies
Please note these Agent and Runtime requirements https://foundryhealth.atlassian.net/wiki/spaces/DOCS/pages/3700490299/ClinSpark+Agent#Device-Dependencies-on-Visual-C++-Runtime-2013.