The study can be divided in three parts1. A laboratory based experiment to addresses the accuracy of the Neural Network based method. A secondary objective is the minimization of activities needed to generate training data for the Neural Network.2.…
ID
Source
Brief title
Condition
- Other condition
Synonym
Health condition
Onderzoek naar ambulante meet techniek, en procedures
Research involving
Sponsors and support
Intervention
Outcome measures
Primary outcome
Part 1: Accuracy of the NN method compared to output of a musculoskeletal
model, over subjects.
Part 2: Feasibility of the longterm measurement of upper extremity kinematics
using IMMS and surface EMG under daily conditions, the application of the NN
method on large datasets, descriptives of a first impression of ambulatory
obtained shoulder joint load profile of healthy subjects.
Part 3: Feasibility of the NN-method on subjects with a shoulder endoprothesis
under daily conditions, descriptives of the first impression of ambulatory
obtained shoulder joint load profile of subjects with a shoulder endoprothesis.
Secondary outcome
Part 1: The minimization of activities needed to generate `sufficient` training
data for the Neural Network. `Sufficient` means here without degrading the
initial performance of the Neural Network performance. This minimized set of
activities will be used in part 2 and 3 of the experiment to train the NN under
daily conditions.
Background summary
To facilitate the development of enhanced shoulder endo protheses, a long term
loading profile of the shoulder joint under daily conditions is needed. This
joint load, in terms of estimated muscle forces, joint reaction forces (JRF),
forces on ligaments etc. can be estimated by musculoskeletal models, using 3D
kinematics and external forces as input. For long term ambulatory measurements,
3D kinematics can be measured by means of Inertial Magnetic Measurement Systems
(IMMS). However, recording of external forces under daily conditions is not
practicable.
This study addresses the feasibility and accuracy of a Neural Network (NN)
approach in predicting shoulder joint load. The input is based upon arm
kinematics and shoulder muscle surface EMG, as an ambulatory obtainable measure
of exerted force. Output of the NN are variables that represent joint load
directly (JRF), and variables that can be used as input for a musculoskeletal
model to calculate mechanical loading of the shoulder joint in full detail.
Study objective
The study can be divided in three parts
1. A laboratory based experiment to addresses the accuracy of the Neural
Network based method. A secondary objective is the minimization of activities
needed to generate training data for the Neural Network.
2. Technical feasibility of the NN-method outside the laboratory, in *True
Life*, for healthy subjects. A method in handling large datasets will be
evaluated. Given the accuracy obtained in part 1, a first impression of joint
loading of the shoulder under daily conditions will be described.
3. Feasibility of the NN method under daily conditions for subjects with a
shoulder endo-prothesis. A first impression of a shoulder endo-prothesis* load
profile in Real Life will be described.
Study design
Feasibility study
Study burden and risks
The risks for the participants are limited to a minimum. The tasks to be
performed for calibration of measurement equipment, and generating training
data for the Neural Network are functional movements with low external loading,
within the normal daily range of movements. In part two and three the
additional measurements comprise the performing of the normal daily routine of
the subject, observation with non-invasive measurements.
The measurements comprise motion analysis and recording of surface EMG with
portable, battery operated equipment, and pose no risk or inconvenience for the
subjects. Part 1 of the study requires a one day-part (4 hours) visit to a
motion laboratory, part two and three require initial preparation time of about
45 minutes before actual measurement of the daily routine can start.
Due to the explorative nature of the study, there is no direct personal benefit
for the subjects, but their support in this explorative pilot will help in
determining the accuracy and feasibility of the method, and in accessing
exemplary Real Life data, which is needed to further develop the described
method.
Roessinghsbleekweg 33B
7522AH, Enschede
NL
Roessinghsbleekweg 33B
7522AH, Enschede
NL
Listed location countries
Age
Inclusion criteria
For part 1 and 2 of the study:
Healthy male
right handed
over 50 years of age
no history of shoulder complaints
active in ADL
ability to understand and follow instructions
ability to complete measurement session;For part 3 of the study:
Healthy male
right handed
over 50 years of age
with a shoulder endoprotheses (right shoulder)
at least 6 month after completion of the rehabilitation process (concerning implant)
active in ADL
ability to understand and follow instructions
ability to complete measurement session
Exclusion criteria
For part 3:
Co-morbidity of disorders affecting use of the upper extremity, like rheuma.
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 | NL32115.044.10 |