To identify discriminatory features of smart phone and smart watch technology data to assess disease activity and if successful to evaluate the chosen technologies in a larger patient group with different levels of clinical disease activity in…
ID
Source
Brief title
Condition
- Joint disorders
Synonym
Research involving
Sponsors and support
Intervention
Outcome measures
Primary outcome
As described above we aim to develop digital biomarkers of disease activity
which could include data from the accelerometers of the telephone, metadata of
the keystroke path possibly combined with camera data. This will be then
compared to clinical disease activity and self-reported symptoms.
Clinical disease activity is measured by validated measures of disease
activity: Minimal disease Activity * MDA - [20], Psoriatic Arthritis Disease
Activity Score (PASDAS)[22], and Disease Activity Psoriatic Arthritis) (DAPSA)
Secondary outcome
not applicable
Background summary
The level of disease activity in Psoriatic Arthritis determines the actions the
rheumatologist takes to optimise treatment outcome among patients with this
disease. Currently disease activity is measured by a combination of clinical
measures and patients self-reported symptoms and functional ability. This
requires the patients to visit the outpatient clinic on regular intervals.,
which during Covid-19 pandemic was not always possible. The use of
questionnaires to collect Patients Reported Outcome (PRO*s) is a feasible
option, the questionnaires fatigue is however a known limiting factor from a
long term perspective. Currently there is no valid alternative for remote
unobtrusive disease activity monitoring.
The wide spread use of smart devices by the general population, such as
smartphone or smartwatch, provides opportunities to develop and study
possibilities for Unobtrusive Remote Disease activity monitoring (URD) using
behavioural data captured by the sensors embedded within the
smartphones/smartwatches. We hypothesize that high level of disease activity in
Psoriatic Arthritis (PsA) will lead to lower degree of physical activity as
registered by patients* smartphone as compared to a low disease activity state.
Additionally, digital biomarkers are likely to provide information on other
disease characteristics such as tiredness and sleep problems. Adding these will
enhance discriminative ability of our approach. We also hypothesize that the
information acquired by digital biomarkers will be comparable to the
information received through usage of clinical measures and PROs. Last but not
least, we hypothesize that patient will see the use of smartphone data as safe
privacy-wise and a fair deal in return for lower amount of follow-up
appointments at the out-patient clinic.
Overarching aim: To dynamically monitor the current status of disease activity
in Psoriatic Arthritis (PsA) patients using novel personalized digital
biomarkers to allow for early treatment adjustments, better disease control and
lessen the burden of follow up visits.
Study objective
To identify discriminatory features of smart phone and smart watch technology
data to assess disease activity and if successful to evaluate the chosen
technologies in a larger patient group with different levels of clinical
disease activity in Psoriatic Arthritis patients
Study design
An explorative longitudinal study will be set up using Design Thinking
principals involving patients from set up onwards. The study consists of 2
stages. Stage 1 aims to find relevant data sources in smart devices and stage 2
aims to test the most promising features of stage 1 in a larger, but still
modest, patient sample
Stage 1
Define
In the Define phase, the problem will be rephrased and a framework will be
provided for functional and non-functional user requirements (mandatory,
desired, optional) for possible solutions. This will be done by the research
team consisting of patients (n=5), doctors (n=3), specialized rheumatology
nurses(n=3) and researchers/technicians (n=8).
Digital disease activity measure - Prototype development
We plan to use technics build on the extensive experience and previous work by
the AUTH* and FMH# in using artificial intelligence (AI)-based digital
biomarkers in the monitoring of disease activity and progression in Parkinson*s
Disease (www.i-prognosis.eu). By creating a behavioral feature vector, derived
from the advanced processing of each PsA patient*s interaction with her or his
smart device, novel personalized digital biomarkers can be formed that capture
subtle changes towards the evolution of the PsA symptoms over time. The related
data sources in smart devices may include: a) 3-axes accelerometer/gyro (to
measure body's specific force and movement; b) keystroke dynamics (virtual
keyboard on smart devices) that may relate with upper extremity activity; c)
smartphone camera, to acquire the nails, skin and eye status, for signs of
inflammation as part of systemic involvement in PsA. It also might be part of
the solution to use a smartwatch for acquiring data on heartrate/sleep changes.
Another example may be intelligent gamified tests that includes gamified
exercises to be followed by the PsA patient in a periodic manner. Patients*
interaction within the game is captured by smartphone camera and provide data
for digital marker to quantify patients* range of motion, balance and
coordination status over time
The identified solutions will be graded according to pre-set requirements from
the Define phase. In the Proof-of-concept phase, one or more working prototypes
will be constructed and tested in 10 patients in several *test and adjust*
cycles. This will provide us with a feasible solution to be tested among a
larger sample of patients in stage 2
Stage 2
In the early exploratory test phase, the digital biomarker(s) will be tested in
daily clinical practice over a 3-month period. To do so, we propose to ask 80
individuals to participate: 40 patients with low disease activity and 40 newly
diagnosed patients with a high level of disease activity.
Clinical disease activity will be captured by the treating physicians as part
of usual care at start and finish of the 3-month study interval. As patients
are already participating in DEPAR (DEPAR MEC-2012-549) we aim to use DEPAR
data for the purpose of this study. In brief, in the first year of DEPAR each
patient is evaluated at 3-month intervals by both the treating physician and a
research nurse and fills out the standard DEPAR PROs. In the second year this
is at 6-month intervals followed by yearly intervals. As we aim to capture
digital disease activity over a 3-month period we will schedule additional
visits for those patients that have no clinical or study visit close by.
Digital disease activity will be captured by the method developed in stage 1,
for which we currently assume it will be an app to be installed on the mobile
phone. This will be accompanied by questions on pain, fatigue, skin irritation
and stiffness (Likert Scale) that will be generate in a random sequence 1 to 3
times over a 24-hour period, taking a few seconds to be answered. The latter
will provide us with information to assess the disease symptoms over time
outside the window of the clinical assessment of disease activity.
Study burden and risks
There are no health risks associated with participation in this study
Burden
Patients will be requested to install an app on their phone that will collect
the metadata of the keystrokes and the accelerometer data of the phone. In the
app they have full control of the data they want to share. This means that they
could stop data sharing at all times without asking our permission.
To monitor the levels of pain, fatigue and stiffness during the day the app
will also send requests to complete questions on these symptoms. This will be
very short questions that are answerable within a few seconds. These questions
will appear between 1 to 3 times a day.
Clinical disease activity will be monitored each 3 months as described above.
For most patients this will be a regular visit to the physician. If they only
visit their physician at 6 months or at longer intervals, they are asked to
have an additional 3 month appoint for clinical disease activity assessment
As participants already participate in DEPAR we will use their DEPAR
self-reported measures. If they are diagnosed less than 12 month ago no
additional work is required. If they participate longer than 12 months they may
receive additional questions if we could not combine the current data
collection with their regular DEPAR visit. This will take about 10 minutes
extra per visit.
Doctor Molenwaterplein 40
Rotterdam 3015 GD
NL
Doctor Molenwaterplein 40
Rotterdam 3015 GD
NL
Listed location countries
Age
Inclusion criteria
active disease defined as not in MDA - 30 patients
inactive disease defined as in MDA - 30 patients
healthy controls - 20 subjects
Exclusion criteria
other disease that linfluence movement such CVA, prostethic limb etc in
patients and sport trauma in healthy controls
Design
Recruitment
Followed up by the following (possibly more current) registration
No registrations found.
Other (possibly less up-to-date) registrations in this register
No registrations found.
In other registers
Register | ID |
---|---|
CCMO | NL78550.078.21 |