Test 3 Flashcards
Gait analysis can be used to
-Provide a quatitative assessment of function or mobility (Frailty and fall risk in old adults)
-Support treatment options (Surgical options OA and CP, orthoses for CP)
-Examine disease state or progression (ie PD, MS, OA)
Gait cyle: stace subphases
1-Loading response
2-mid-stance
3-terminal stance
4-pre-swing
Gait cyle: stance subphases- loading response
inital contact to opposite toe off (double support)
Gait cyle: stance subphases- mid-stance
opposite toe off to heel rise (single support)
foot is flat (data on foot is flat)
Gait cyle: stance subphases- terminal stance
heel rise to opposite initial contact (single support)
Gait cyle: stance subphases- pre-swing
opposide inital contact to tow odd (double support)
Gait cyle: swing
1-inital swing
2-mid-swing
3-terminal swing
Gait cyle: swing subphases- initial swing
tow off to feet adjacent (aligned)
Gait cyle: swing- mid-swing
feet adjacent to tibia vertical
Gait cyle: swing- terminal swing
tibia vertical to initial contact
common spatiotemporal parameters
-temporal parameters (seconds)
-Spatial parameters (meters)
-Gait speed (meters/sec)
temporal parameters
-step and stride time
-stance time and swing time
-single support time
-double support time
spatial parameters
-step/stide length
-base width
-foot angle
Ground reaction forces- kinetics: force exerted by the ground on the body
-offer insights into actions at each subphase of gait
-gait is just falling with style (propell and catch each gait cycle)
joint angles- kinetics: The movment patterns without considering forces
-sensitive to changes with age, clinical considerations, and injuries
-stuied for years with “marker-based”optical methods
-newer technologies make this easier
joint angles- kinematics: can be quite complicated
-occurs in three dimensions
-ankle, kmee, and hip all influence each other
-often focus is often on fewer joints/planes
what should we measure: it depends on…
-the question you are trying to answer
-the information important/useful to your end user (what do we care about)
-the patient population or conditon
The literature is the best place to start when trying to understand what you should be measuring
General findings from the literature with respect to aging and gait
-decreased gair speed, spatial parameters
-increased temporal parameters, variability
However, healthy older adults may have little to no change
Larger changes can occur with advanced age, additonof clinical conditons, reduced from executive function,ect
Where should we measure it (in vs out of lab)
trade off between more controlled vs more realistic
normal vs perturbed (eg dual task)
-optimal gait vs stressed system
Assesed during functional tests?
-timed up and go, 6 minute walk test, self-paced walk test
How do we measure it (with sensors)
-Temporal parameters (seconds): timing of heel strike and toe off
-Spatial parameters (m or m/s): requires integration of acceleration and displacment (more complex)
-joint angles (degrees): sensor on each segment or “rigid body”, requires integration of angular velocity to angular displacment
How do we get displacment
accleration–> veolcity–> displacment
use numerical integration to get between
getting displacment is complicated by…
-any offset in data will be greatly amplified (low frequency- eg gravity/calibration issue)
-Error in signal will accumulate (high frequency noise)
-Unknown inital conditons; just measure accleration (only able to determine changes from inital stae)
getting from accleration to velocity (on way to displacement)
1-Remove high-frequency noise (low pass filter)
2-remove offset/gravity (detrend/high plass filter)
3-numerical intergration (trapezoidal integration)
getting from velcity to displacment
1- remove offset (detrend/high pass filter)
2- numerical integration (trapezoidal integration)
Joint angles- sagittal plane
-Relative angles between segements- can measure 3D at each joints, but focus is often sagittal
sagittal knee joint angle; gait cycle
- At heel strike, the knee is typically flexed (slightly), and continues to flex (absorbs load) as knee extensors work eccentrically
- Knee then extends (knee extensors working concentrically) as body moves forward over stance limb)
- Knee flexors as toe off approaches and plantarflexion occurs/heel begins to lift
4.Flexion continues to mid swing peak
5.Knee begins to extend in preperation for heel strike again
Saggital knee joint angle: changes in knee flexion during gair can be related to:
\
-pain, mucle function, flexibility, ect
-Aging, injuries (eg ACL), OA, ect
two common abnormaliteis in knee joint angle
-stiff knee gait
-flexure contracture
Stiff knee gait
-Reduced knee flexion at heel stike, midstance, and ir swing
-common in OA: reduced ROM, bening= painful, Quad avoidance strategy
-Knee arthroplasty to improve pain and function (less stiff knee gait)
Flexion contracture
-inability to fully straighten knee
-can occur w/ OA or other conditons: congenital deformities, rheumatoid arthritis, CP
Can influence or be influenced by joints above or below
Clinical case study- CP
CP results from damage to one or more areas of the developing brain
-can occur pre or post natally
-Numerous causes (eg premature birth, hypoxia)
-various difficulties with gait
Clinal case study: CP- 3 main differences
-anterior pelvic tilit
-equinus- restricted dorsifelxion (toe walking)
-reduved knee flexion (also see decreased extension at contact)
Clinal case study: CP- cause and effect?
PLantarfleor (gastric/soleus) spasticity
-results in equinus
-results in reduced knee flexion at heel strike (and stance-locked)
-Reduced in excessive ant pevlic tilt to progress forward- try to get over footl leaning to move forward
clinical case study- Cp treatment
AFOS- equinus is prevented/lmited
secondary anterior pelvic tilit and extended knee are resolved
How do we measure joint angles with IMUs?
- Need to get orientation estimates from IMUs
-Angular velcity to angular displacement
2.compare orientations from segments on either side of the joint
-angle of shank vs angle of thigh
getting to angluar displacment (gyro)
similar accumulation of erros…
-Bias instabilties- low freqeuncy offset changes
-angluar random walk- stochastic noise (random noise in the data that accumulates when integrated)
-unknown initial conditions- only able to determine changes from inital state
getting to angluar displacement from angular velocity
-remove high frequency noise (low pass filter)
-remove offset (remove bias)
-numerical integration (trapezoidal integration)
*** need to consider drift!!
Zero velocity update (ZUPT)
used on foot mounted sensors
-Foot is fixed to the ground (ie has zero veolicty) during the stance phase
-can use this to “anchor” to output of the signal
-Bring it back to zero and minimize any dift
Can minimize error within a single sensor but teaming up with other sensors is most effective
sensor fusion
-Combining data from different types of sensors to obtain estimates of displacment and orientation that have less uncertaininty
sensor fusion- the elephant parable
-the accelerometer, gyroscope, and magnetometer are all measuring the same motion, but seeing it in different ways
-Each sees different partial truths (all asses one part of whole, alone they are all wrong, together get better picture)
-sensor fusion aims to exploit their strengths (while understnading their limitations) so that we can obtain a measure of the true motion
important considerations: sensor fusion- kalman filters
-previous state (i-1) + current inputs (i) to estimate current state (i)
-more complex weighting of sensors inputs
-significanwork needed to create robust algorithms for different movments
important considerations: machine learning- input raw IMU data and outputs joint angles
-Black box model
-requires lots of trianing data
-Can work well in ideal conditons, but may transfer poorly to others
Other gait related considerations- physical activry assesment: estimates metabolic equivalents (METs) from accelerometr counts
-working metabolic rate in comparison to your resting metabolic rate (1 met= energy used while resting)
-count is a measure of the number of peaks over a sufficeint threashold in a signal (soft of…)
Other gait related considerations- physical activry assesment: detrmine activity level based on counts per minute
calculate based on shaking of sensor
sed: 0-99cpm
light: 100-1951cpm
mod: 1952-5724 cpm
vigorous: 5725-9498 cpm
very vig: >9499cpm
Other gait related considerations- activity classification
-suing inertial signals to classify time spent in various activites
-can range in complexity and form: stationary vs dynamic- simply threashold of inertial data
seperating all different types of activites- utilizes artificial intelligence (eg machine learning)
