def test_functions(self):
filename = pyCGM2.TEST_DATA_PATH + "LowLevel\\IO\\Hånnibøl_c3d\\gait1.c3d"
acq = btkTools.smartReader(filename, translators=None)
btkTools.GetMarkerNames(acq)
btkTools.findNearestMarker(acq, 0, "LASI")
btkTools.GetAnalogNames(acq)
btkTools.isGap(acq, "LASI")
btkTools.findMarkerGap(acq)
btkTools.isPointExist(acq, "LASI")
btkTools.isPointsExist(acq, ["LASI", "RASI"])
btkTools.clearPoints(acq, ["LASI", "RASI"])
btkTools.checkFirstAndLastFrame(acq, "LASI")
btkTools.isGap_inAcq(acq, ["LASI", "RASI"])
btkTools.findValidFrames(acq, ["LASI", "RASI"])
btkTools.checkMultipleSubject(acq)
btkTools.checkMarkers(acq, ["LASI", "RASI"])
btkTools.clearEvents(acq, ["Foot Strike"])
btkTools.modifyEventSubject(acq, "Hän")
btkTools.modifySubject(acq, "Han")
btkTools.getVisibleMarkersAtFrame(acq, ["LASI", "RASI"], 0)
btkTools.isAnalogExist(acq, "emg-Hän")
btkTools.createZeros(acq, ["LASI", "RASI"])
btkTools.constructEmptyMarker(acq, "zéros", desc="Hän")
btkTools.getStartEndEvents(acq, "Left")
btkTools.changeSubjectName(acq, "Hän")
btkTools.smartGetMetadata(acq, "SUBJECTS", "USED")
btkTools.smartSetMetadata(acq, "SUBJECTS", "USED", 0, "Hän")
def detect(self, acq):
"""
"""
ff = acq.GetFirstFrame()
if btkTools.isPointsExist(
acq, ["LPSI", "RPSI", "LHEE", "LTOE", "RHEE", "RTOE"]):
pfp = progressionFrame.PelvisProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acq, pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
longAxisIndex = 0 if progressionAxis == "X" else 1
sacrum = (acq.GetPoint("LPSI").GetValues() +
acq.GetPoint("RPSI").GetValues()) / 2.0
#Left
heel_left = acq.GetPoint("LHEE").GetValues()
toe_left = acq.GetPoint("LTOE").GetValues()
diffHeel_left = heel_left - sacrum
diffToe_left = toe_left - sacrum
#Right
heel_right = acq.GetPoint("RHEE").GetValues()
toe_right = acq.GetPoint("RTOE").GetValues()
diffHeel_right = heel_right - sacrum
diffToe_right = toe_right - sacrum
if forwardProgression:
indexes_fs_left = detect_peaks.detect_peaks(
diffHeel_left[:, longAxisIndex]) + ff
indexes_fo_left = detect_peaks.detect_peaks(
-diffToe_left[:, longAxisIndex]) + ff
else:
indexes_fs_left = detect_peaks.detect_peaks(
-diffHeel_left[:, longAxisIndex]) + ff
indexes_fo_left = detect_peaks.detect_peaks(
diffToe_left[:, longAxisIndex]) + ff
if forwardProgression:
indexes_fs_right = detect_peaks.detect_peaks(
diffHeel_right[:, longAxisIndex]) + ff
indexes_fo_right = detect_peaks.detect_peaks(
-diffToe_right[:, longAxisIndex]) + ff
else:
indexes_fs_right = detect_peaks.detect_peaks(
-diffHeel_right[:, longAxisIndex]) + ff
indexes_fo_right = detect_peaks.detect_peaks(
diffToe_right[:, longAxisIndex]) + ff
return indexes_fs_left + self.footStrikeOffset, indexes_fo_left + self.footOffOffset, indexes_fs_right + self.footStrikeOffset, indexes_fo_right + self.footOffOffset
else:
logging.error(
"[pyCGM2]: Zeni event detector impossible to run. Pelvic LPSI-RPSI or foot markers(HEE or TOE) are missing "
)
return 0
def fitting(model,DATA_PATH, reconstructFilenameLabelled,
translators,
markerDiameter,
pointSuffix,
mfpa,
momentProjection,**kwargs):
"""
Fitting of the CGM2.1
:param model [str]: pyCGM2 model previously calibrated
:param DATA_PATH [str]: path to your data
:param reconstructFilenameLabelled [string list]: c3d files
:param translators [dict]: translators to apply
:param mfpa [str]: manual force plate assignement
:param markerDiameter [double]: marker diameter (mm)
:param pointSuffix [str]: suffix to add to model outputs
:param momentProjection [str]: Coordinate system in which joint moment is expressed
"""
detectAnomaly = False
if "anomalyException" in kwargs.keys():
anomalyException = kwargs["anomalyException"]
else:
anomalyException=False
# --------------------------ACQUISITION ------------------------------------
# --- btk acquisition ----
if "forceBtkAcq" in kwargs.keys():
acqGait = kwargs["forceBtkAcq"]
else:
acqGait = btkTools.smartReader((DATA_PATH + reconstructFilenameLabelled))
btkTools.checkMultipleSubject(acqGait)
if btkTools.isPointExist(acqGait,"SACR"):
translators["LPSI"] = "SACR"
translators["RPSI"] = "SACR"
LOGGER.logger.info("[pyCGM2] Sacrum marker detected")
acqGait = btkTools.applyTranslators(acqGait,translators)
trackingMarkers = cgm.CGM1.LOWERLIMB_TRACKING_MARKERS + cgm.CGM1.THORAX_TRACKING_MARKERS+ cgm.CGM1.UPPERLIMB_TRACKING_MARKERS
actual_trackingMarkers,phatoms_trackingMarkers = btkTools.createPhantoms(acqGait, trackingMarkers)
vff,vlf = btkTools.getFrameBoundaries(acqGait,actual_trackingMarkers)
if "frameInit" in kwargs.keys() and kwargs["frameInit"] is not None:
vff = kwargs["frameInit"]
LOGGER.logger.info("[pyCGM2] first frame forced to frame [%s]"%(vff))
if "frameEnd" in kwargs.keys() and kwargs["frameEnd"] is not None:
vlf = kwargs["frameEnd"]
LOGGER.logger.info("[pyCGM2] end frame forced to frame [%s]"%(vlf))
flag = btkTools.getValidFrames(acqGait,actual_trackingMarkers,frameBounds=[vff,vlf])
LOGGER.logger.info("[pyCGM2] Computation from frame [%s] to frame [%s]"%(vff,vlf))
# --------------------ANOMALY------------------------------
for marker in actual_trackingMarkers:
if marker not in model.getStaticTrackingMarkers():
LOGGER.logger.warning("[pyCGM2-Anomaly] marker [%s] - not used during static calibration - wrong kinematic for the segment attached to this marker. "%(marker))
# --marker presence
markersets = [cgm.CGM1.LOWERLIMB_TRACKING_MARKERS, cgm.CGM1.THORAX_TRACKING_MARKERS, cgm.CGM1.UPPERLIMB_TRACKING_MARKERS]
for markerset in markersets:
ipdp = InspectorProcedure.MarkerPresenceDetectionProcedure( markerset)
idf = InspectorFilter.InspectorFilter(acqGait,reconstructFilenameLabelled,ipdp)
inspector = idf.run()
# --marker outliers
if inspector["In"] !=[]:
madp = AnomalyDetectionProcedure.MarkerAnomalyDetectionRollingProcedure( inspector["In"], plot=False, window=5,threshold = 3)
adf = AnomalyFilter.AnomalyDetectionFilter(acqGait,reconstructFilenameLabelled,madp, frameRange=[vff,vlf])
anomaly = adf.run()
anomalyIndexes = anomaly["Output"]
if anomaly["ErrorState"]: detectAnomaly = True
if btkTools.checkForcePlateExist(acqGait):
afpp = AnomalyDetectionProcedure.ForcePlateAnomalyProcedure()
adf = AnomalyFilter.AnomalyDetectionFilter(acqGait,reconstructFilenameLabelled,afpp, frameRange=[vff,vlf])
anomaly = adf.run()
if anomaly["ErrorState"]: detectAnomaly = True
if detectAnomaly and anomalyException:
raise Exception ("Anomalies has been detected - Check Warning message of the log file")
# --------------------MODELLING------------------------------
# filtering
# -----------------------
if "fc_lowPass_marker" in kwargs.keys() and kwargs["fc_lowPass_marker"]!=0 :
fc = kwargs["fc_lowPass_marker"]
order = 4
if "order_lowPass_marker" in kwargs.keys():
order = kwargs["order_lowPass_marker"]
signal_processing.markerFiltering(acqGait,trackingMarkers,order=order, fc =fc)
if "fc_lowPass_forcePlate" in kwargs.keys() and kwargs["fc_lowPass_forcePlate"]!=0 :
fc = kwargs["fc_lowPass_forcePlate"]
order = 4
if "order_lowPass_forcePlate" in kwargs.keys():
order = kwargs["order_lowPass_forcePlate"]
signal_processing.forcePlateFiltering(acqGait,order=order, fc =fc)
scp=modelFilters.StaticCalibrationProcedure(model)
# ---Motion filter----
modMotion=modelFilters.ModelMotionFilter(scp,acqGait,model,enums.motionMethod.Determinist,
markerDiameter=markerDiameter)
modMotion.compute()
progressionFlag = False
if btkTools.isPointExist(acqGait, 'LHEE',ignorePhantom=False) or btkTools.isPointExist(acqGait, 'RHEE',ignorePhantom=False):
pfp = progressionFrame.PointProgressionFrameProcedure(marker="LHEE") \
if btkTools.isPointExist(acqGait, 'LHEE',ignorePhantom=False) \
else progressionFrame.PointProgressionFrameProcedure(marker="RHEE")
pff = progressionFrame.ProgressionFrameFilter(acqGait,pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
progressionFlag = True
elif btkTools.isPointsExist(acqGait, ['LASI', 'RASI', 'RPSI', 'LPSI'],ignorePhantom=False) and not progressionFlag:
LOGGER.logger.info("[pyCGM2] - progression axis detected from Pelvic markers ")
pfp = progressionFrame.PelvisProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acqGait,pfp)
pff.compute()
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
progressionFlag = True
elif btkTools.isPointsExist(acqGait, ['C7', 'T10', 'CLAV', 'STRN'],ignorePhantom=False) and not progressionFlag:
LOGGER.logger.info("[pyCGM2] - progression axis detected from Thoracic markers ")
pfp = progressionFrame.ThoraxProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acqGait,pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
else:
globalFrame = "XYZ"
progressionAxis = "X"
forwardProgression = True
LOGGER.logger.error("[pyCGM2] - impossible to detect progression axis - neither pelvic nor thoracic markers are present. Progression set to +X by default ")
if "displayCoordinateSystem" in kwargs.keys() and kwargs["displayCoordinateSystem"]:
csp = modelFilters.ModelCoordinateSystemProcedure(model)
csdf = modelFilters.CoordinateSystemDisplayFilter(csp,model,acqGait)
csdf.setStatic(False)
csdf.display()
#---- Joint kinematics----
# relative angles
modelFilters.ModelJCSFilter(model,acqGait).compute(description="vectoriel", pointLabelSuffix=pointSuffix)
modelFilters.ModelAbsoluteAnglesFilter(model,acqGait,
segmentLabels=["Left Foot","Right Foot","Pelvis","Thorax","Head"],
angleLabels=["LFootProgress", "RFootProgress","Pelvis","Thorax", "Head"],
eulerSequences=["TOR","TOR", "ROT","YXZ","TOR"],
globalFrameOrientation = globalFrame,
forwardProgression = forwardProgression).compute(pointLabelSuffix=pointSuffix)
#---- Body segment parameters----
bspModel = bodySegmentParameters.Bsp(model)
bspModel.compute()
modelFilters.CentreOfMassFilter(model,acqGait).compute(pointLabelSuffix=pointSuffix)
# Inverse dynamics
if btkTools.checkForcePlateExist(acqGait):
if model.m_bodypart != enums.BodyPart.UpperLimb:
# --- force plate handling----
# find foot in contact
mappedForcePlate = forceplates.matchingFootSideOnForceplate(acqGait,mfpa=mfpa)
forceplates.addForcePlateGeneralEvents(acqGait,mappedForcePlate)
LOGGER.logger.info("Manual Force plate assignment : %s" %mappedForcePlate)
# assembly foot and force plate
modelFilters.ForcePlateAssemblyFilter(model,acqGait,mappedForcePlate,
leftSegmentLabel="Left Foot",
rightSegmentLabel="Right Foot").compute(pointLabelSuffix=pointSuffix)
#---- Joint kinetics----
idp = modelFilters.CGMLowerlimbInverseDynamicProcedure()
modelFilters.InverseDynamicFilter(model,
acqGait,
procedure = idp,
projection = momentProjection,
globalFrameOrientation = globalFrame,
forwardProgression = forwardProgression
).compute(pointLabelSuffix=pointSuffix)
#---- Joint energetics----
modelFilters.JointPowerFilter(model,acqGait).compute(pointLabelSuffix=pointSuffix)
btkTools.cleanAcq(acqGait)
btkTools.applyOnValidFrames(acqGait,flag)
if detectAnomaly and not anomalyException:
LOGGER.logger.error("Anomalies has been detected - Check Warning messages of the log file")
return acqGait,detectAnomaly
def calibrate(DATA_PATH,calibrateFilenameLabelled,translators,
required_mp,optional_mp,
leftFlatFoot,rightFlatFoot,headFlat,
markerDiameter,hjcMethod,
pointSuffix,**kwargs):
"""
Calibration of the CGM2.1
:param DATA_PATH [str]: path to your data
:param calibrateFilenameLabelled [str]: c3d file
:param translators [dict]: translators to apply
:param required_mp [dict]: required anthropometric data
:param optional_mp [dict]: optional anthropometric data (ex: LThighOffset,...)
:param leftFlatFoot [bool]: enable of the flat foot option for the left foot
:param rightFlatFoot [bool]: enable of the flat foot option for the right foot
:param headFlat [bool]: enable of the head flat option
:param markerDiameter [double]: marker diameter (mm)
:param hjcMethod [str or list of 3 float]: method for locating the hip joint centre
:param pointSuffix [str]: suffix to add to model outputs
"""
detectAnomaly = False
if "anomalyException" in kwargs.keys():
anomalyException = kwargs["anomalyException"]
else:
anomalyException=False
# --------------------------ACQUISITION ------------------------------------
# ---btk acquisition---
if "forceBtkAcq" in kwargs.keys():
acqStatic = kwargs["forceBtkAcq"]
else:
acqStatic = btkTools.smartReader((DATA_PATH+calibrateFilenameLabelled))
btkTools.checkMultipleSubject(acqStatic)
if btkTools.isPointExist(acqStatic,"SACR"):
translators["LPSI"] = "SACR"
translators["RPSI"] = "SACR"
LOGGER.logger.info("[pyCGM2] Sacrum marker detected")
acqStatic = btkTools.applyTranslators(acqStatic,translators)
trackingMarkers = cgm.CGM1.LOWERLIMB_TRACKING_MARKERS + cgm.CGM1.THORAX_TRACKING_MARKERS+ cgm.CGM1.UPPERLIMB_TRACKING_MARKERS
actual_trackingMarkers,phatoms_trackingMarkers = btkTools.createPhantoms(acqStatic, trackingMarkers)
vff = acqStatic.GetFirstFrame()
vlf = acqStatic.GetLastFrame()
# vff,vlf = btkTools.getFrameBoundaries(acqStatic,actual_trackingMarkers)
flag = btkTools.getValidFrames(acqStatic,actual_trackingMarkers,frameBounds=[vff,vlf])
gapFlag = btkTools.checkGap(acqStatic,actual_trackingMarkers,frameBounds=[vff,vlf])
if gapFlag:
raise Exception("[pyCGM2] Calibration aborted. Gap find during interval [%i-%i]. Crop your c3d " %(vff,vlf))
# --------------------ANOMALY------------------------------
# --Check MP
adap = AnomalyDetectionProcedure.AnthropoDataAnomalyProcedure( required_mp)
adf = AnomalyFilter.AnomalyDetectionFilter(None,None,adap)
mp_anomaly = adf.run()
if mp_anomaly["ErrorState"]: detectAnomaly = True
# --marker presence
markersets = [cgm.CGM1.LOWERLIMB_TRACKING_MARKERS, cgm.CGM1.THORAX_TRACKING_MARKERS, cgm.CGM1.UPPERLIMB_TRACKING_MARKERS]
for markerset in markersets:
ipdp = InspectorProcedure.MarkerPresenceDetectionProcedure( markerset)
idf = InspectorFilter.InspectorFilter(acqStatic,calibrateFilenameLabelled,ipdp)
inspector = idf.run()
# # --marker outliers
if inspector["In"] !=[]:
madp = AnomalyDetectionProcedure.MarkerAnomalyDetectionRollingProcedure(inspector["In"], plot=False, window=4,threshold = 3)
adf = AnomalyFilter.AnomalyDetectionFilter(acqStatic,calibrateFilenameLabelled,madp)
anomaly = adf.run()
anomalyIndexes = anomaly["Output"]
if anomaly["ErrorState"]: detectAnomaly = True
if detectAnomaly and anomalyException:
raise Exception ("Anomalies has been detected - Check Warning message of the log file")
# --------------------MODELLING------------------------------
# ---check marker set used----
dcm = cgm.CGM.detectCalibrationMethods(acqStatic)
# ---definition---
model=cgm2.CGM2_1()
model.configure(detectedCalibrationMethods=dcm)
model.addAnthropoInputParameters(required_mp,optional=optional_mp)
if dcm["Left Knee"] == enums.JointCalibrationMethod.KAD: actual_trackingMarkers.append("LKNE")
if dcm["Right Knee"] == enums.JointCalibrationMethod.KAD: actual_trackingMarkers.append("RKNE")
model.setStaticTrackingMarkers(actual_trackingMarkers)
# --store calibration parameters--
model.setStaticFilename(calibrateFilenameLabelled)
model.setCalibrationProperty("leftFlatFoot",leftFlatFoot)
model.setCalibrationProperty("rightFlatFoot",rightFlatFoot)
model.setCalibrationProperty("headFlat",headFlat)
model.setCalibrationProperty("markerDiameter",markerDiameter)
# --------------------------STATIC CALBRATION--------------------------
scp=modelFilters.StaticCalibrationProcedure(model) # load calibration procedure
# ---initial calibration filter----
# use if all optional mp are zero
modelFilters.ModelCalibrationFilter(scp,acqStatic,model,
leftFlatFoot = leftFlatFoot, rightFlatFoot = rightFlatFoot,
headFlat= headFlat,
markerDiameter=markerDiameter,
).compute()
# ---- Decorators -----
decorators.applyBasicDecorators(dcm, model,acqStatic,optional_mp,markerDiameter)
decorators.applyHJCDecorators(model,hjcMethod)
# ----Final Calibration filter if model previously decorated -----
if model.decoratedModel:
# initial static filter
modelFilters.ModelCalibrationFilter(scp,acqStatic,model,
leftFlatFoot = leftFlatFoot,
rightFlatFoot = rightFlatFoot,
markerDiameter=markerDiameter,
headFlat= headFlat,
).compute()
modMotion=modelFilters.ModelMotionFilter(scp,acqStatic,model,enums.motionMethod.Determinist,
markerDiameter=markerDiameter)
modMotion.compute()
# ----progression Frame----
progressionFlag = False
if btkTools.isPointsExist(acqStatic, ['LASI', 'RASI', 'RPSI', 'LPSI'],ignorePhantom=False):
LOGGER.logger.info("[pyCGM2] - progression axis detected from Pelvic markers ")
pfp = progressionFrame.PelvisProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acqStatic,pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
progressionFlag = True
elif btkTools.isPointsExist(acqStatic, ['C7', 'T10', 'CLAV', 'STRN'],ignorePhantom=False) and not progressionFlag:
LOGGER.logger.info("[pyCGM2] - progression axis detected from Thoracic markers ")
pfp = progressionFrame.ThoraxProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acqStatic,pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
else:
globalFrame = "XYZ"
progressionAxis = "X"
forwardProgression = True
LOGGER.logger.error("[pyCGM2] - impossible to detect progression axis - neither pelvic nor thoracic markers are present. Progression set to +X by default ")
if "displayCoordinateSystem" in kwargs.keys() and kwargs["displayCoordinateSystem"]:
csp = modelFilters.ModelCoordinateSystemProcedure(model)
csdf = modelFilters.CoordinateSystemDisplayFilter(csp,model,acqStatic)
csdf.setStatic(False)
csdf.display()
# ----------------------CGM MODELLING----------------------------------
# ----motion filter----
# notice : viconCGM1compatible option duplicate error on Construction of the foot coordinate system
if "noKinematicsCalculation" in kwargs.keys() and kwargs["noKinematicsCalculation"]:
LOGGER.logger.warning("[pyCGM2] No Kinematic calculation done for the static file")
return model, acqStatic
else:
#---- Joint kinematics----
# relative angles
modelFilters.ModelJCSFilter(model,acqStatic).compute(description="vectoriel", pointLabelSuffix=pointSuffix)
modelFilters.ModelAbsoluteAnglesFilter(model,acqStatic,
segmentLabels=["Left Foot","Right Foot","Pelvis","Thorax","Head"],
angleLabels=["LFootProgress", "RFootProgress","Pelvis","Thorax", "Head"],
eulerSequences=["TOR","TOR", "ROT","YXZ","TOR"],
globalFrameOrientation = globalFrame,
forwardProgression = forwardProgression).compute(pointLabelSuffix=pointSuffix)
# BSP model
bspModel = bodySegmentParameters.Bsp(model)
bspModel.compute()
modelFilters.CentreOfMassFilter(model,acqStatic).compute(pointLabelSuffix=pointSuffix)
btkTools.cleanAcq(acqStatic)
if detectAnomaly and not anomalyException:
LOGGER.logger.error("Anomalies has been detected - Check Warning messages of the log file")
return model, acqStatic,detectAnomaly
def fitting(model, DATA_PATH, reconstructFilenameLabelled, translators,
weights, ik_flag, markerDiameter, pointSuffix, mfpa,
momentProjection, **kwargs):
"""
Fitting of the CGM2.5
:param model [str]: pyCGM2 model previously calibrated
:param DATA_PATH [str]: path to your data
:param reconstructFilenameLabelled [string list]: c3d files
:param translators [dict]: translators to apply
:param ik_flag [bool]: enable the inverse kinematic solver
:param mfpa [str]: manual force plate assignement
:param markerDiameter [double]: marker diameter (mm)
:param pointSuffix [str]: suffix to add to model outputs
:param momentProjection [str]: Coordinate system in which joint moment is expressed
"""
detectAnomaly = False
if "anomalyException" in kwargs.keys():
anomalyException = kwargs["anomalyException"]
else:
anomalyException = False
if "forceFoot6DoF" in kwargs.keys() and kwargs["forceFoot6DoF"]:
forceFoot6DoF_flag = True
else:
forceFoot6DoF_flag = False
if "Fitting" in weights.keys():
weights = weights["Fitting"]["Weight"]
# --- btk acquisition ----
if "forceBtkAcq" in kwargs.keys():
acqGait = kwargs["forceBtkAcq"]
else:
acqGait = btkTools.smartReader(
(DATA_PATH + reconstructFilenameLabelled))
btkTools.checkMultipleSubject(acqGait)
if btkTools.isPointExist(acqGait, "SACR"):
translators["LPSI"] = "SACR"
translators["RPSI"] = "SACR"
LOGGER.logger.info("[pyCGM2] Sacrum marker detected")
acqGait = btkTools.applyTranslators(acqGait, translators)
trackingMarkers = cgm2.CGM2_5.LOWERLIMB_TRACKING_MARKERS + cgm2.CGM2_5.THORAX_TRACKING_MARKERS + cgm2.CGM2_5.UPPERLIMB_TRACKING_MARKERS
actual_trackingMarkers, phatoms_trackingMarkers = btkTools.createPhantoms(
acqGait, trackingMarkers)
vff, vlf = btkTools.getFrameBoundaries(acqGait, actual_trackingMarkers)
if "frameInit" in kwargs.keys() and kwargs["frameInit"] is not None:
vff = kwargs["frameInit"]
LOGGER.logger.info("[pyCGM2] first frame forced to frame [%s]" %
(vff))
if "frameEnd" in kwargs.keys() and kwargs["frameEnd"] is not None:
vlf = kwargs["frameEnd"]
LOGGER.logger.info("[pyCGM2] end frame forced to frame [%s]" % (vlf))
flag = btkTools.getValidFrames(acqGait,
actual_trackingMarkers,
frameBounds=[vff, vlf])
LOGGER.logger.info("[pyCGM2] Computation from frame [%s] to frame [%s]" %
(vff, vlf))
# --------------------ANOMALY------------------------------
for marker in actual_trackingMarkers:
if marker not in model.getStaticTrackingMarkers():
LOGGER.logger.warning(
"[pyCGM2-Anomaly] marker [%s] - not used during static calibration - wrong kinematic for the segment attached to this marker. "
% (marker))
# --marker presence
markersets = [
cgm2.CGM2_5.LOWERLIMB_TRACKING_MARKERS,
cgm2.CGM2_5.THORAX_TRACKING_MARKERS,
cgm2.CGM2_5.UPPERLIMB_TRACKING_MARKERS
]
for markerset in markersets:
ipdp = InspectorProcedure.MarkerPresenceDetectionProcedure(markerset)
idf = InspectorFilter.InspectorFilter(acqGait,
reconstructFilenameLabelled,
ipdp)
inspector = idf.run()
# --marker outliers
if inspector["In"] != []:
madp = AnomalyDetectionProcedure.MarkerAnomalyDetectionRollingProcedure(
inspector["In"], plot=False, window=5, threshold=3)
adf = AnomalyFilter.AnomalyDetectionFilter(
acqGait,
reconstructFilenameLabelled,
madp,
frameRange=[vff, vlf])
anomaly = adf.run()
anomalyIndexes = anomaly["Output"]
if anomaly["ErrorState"]: detectAnomaly = True
if btkTools.checkForcePlateExist(acqGait):
afpp = AnomalyDetectionProcedure.ForcePlateAnomalyProcedure()
adf = AnomalyFilter.AnomalyDetectionFilter(acqGait,
reconstructFilenameLabelled,
afpp,
frameRange=[vff, vlf])
anomaly = adf.run()
if anomaly["ErrorState"]: detectAnomaly = True
if detectAnomaly and anomalyException:
raise Exception(
"Anomalies has been detected - Check Warning message of the log file"
)
# --------------------MODELLING------------------------------
# filtering
# -----------------------
if "fc_lowPass_marker" in kwargs.keys(
) and kwargs["fc_lowPass_marker"] != 0:
fc = kwargs["fc_lowPass_marker"]
order = 4
if "order_lowPass_marker" in kwargs.keys():
order = kwargs["order_lowPass_marker"]
signal_processing.markerFiltering(acqGait,
trackingMarkers,
order=order,
fc=fc)
if "fc_lowPass_forcePlate" in kwargs.keys(
) and kwargs["fc_lowPass_forcePlate"] != 0:
fc = kwargs["fc_lowPass_forcePlate"]
order = 4
if "order_lowPass_forcePlate" in kwargs.keys():
order = kwargs["order_lowPass_forcePlate"]
signal_processing.forcePlateFiltering(acqGait, order=order, fc=fc)
# --- initial motion Filter ---
scp = modelFilters.StaticCalibrationProcedure(model)
modMotion = modelFilters.ModelMotionFilter(scp, acqGait, model,
enums.motionMethod.Sodervisk)
modMotion.compute()
progressionFlag = False
if btkTools.isPointExist(acqGait, 'LHEE',
ignorePhantom=False) or btkTools.isPointExist(
acqGait, 'RHEE', ignorePhantom=False):
pfp = progressionFrame.PointProgressionFrameProcedure(marker="LHEE") \
if btkTools.isPointExist(acqGait, 'LHEE',ignorePhantom=False) \
else progressionFrame.PointProgressionFrameProcedure(marker="RHEE")
pff = progressionFrame.ProgressionFrameFilter(acqGait, pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
progressionFlag = True
elif btkTools.isPointsExist(acqGait, ['LASI', 'RASI', 'RPSI', 'LPSI'],
ignorePhantom=False) and not progressionFlag:
LOGGER.logger.info(
"[pyCGM2] - progression axis detected from Pelvic markers ")
pfp = progressionFrame.PelvisProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acqGait, pfp)
pff.compute()
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
progressionFlag = True
elif btkTools.isPointsExist(acqGait, ['C7', 'T10', 'CLAV', 'STRN'],
ignorePhantom=False) and not progressionFlag:
LOGGER.logger.info(
"[pyCGM2] - progression axis detected from Thoracic markers ")
pfp = progressionFrame.ThoraxProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acqGait, pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
else:
globalFrame = "XYZ"
progressionAxis = "X"
forwardProgression = True
LOGGER.logger.error(
"[pyCGM2] - impossible to detect progression axis - neither pelvic nor thoracic markers are present. Progression set to +X by default "
)
for target in weights.keys():
if target not in actual_trackingMarkers or target not in model.getStaticIkTargets(
):
weights[target] = 0
LOGGER.logger.warning(
"[pyCGM2] - the IK targeted marker [%s] is not labelled in the acquisition [%s]"
% (target, reconstructFilenameLabelled))
if ik_flag:
# ---OPENSIM IK---
# --- opensim calibration Filter ---
osimfile = pyCGM2.OPENSIM_PREBUILD_MODEL_PATH + "models\\osim\\lowerLimb_ballsJoints.osim" # osimfile
markersetFile = pyCGM2.OPENSIM_PREBUILD_MODEL_PATH + "models\\settings\\cgm2_4\\cgm2_4-markerset.xml" # markerset
cgmCalibrationprocedure = opensimFilters.CgmOpensimCalibrationProcedures(
model) # procedure
oscf = opensimFilters.opensimCalibrationFilter(
osimfile, model, cgmCalibrationprocedure, DATA_PATH)
oscf.addMarkerSet(markersetFile)
scalingOsim = oscf.build()
# --- opensim Fitting Filter ---
iksetupFile = pyCGM2.OPENSIM_PREBUILD_MODEL_PATH + "models\\settings\\cgm2_4\\cgm2_4-ikSetUp_template.xml" # ik tl file
cgmFittingProcedure = opensimFilters.CgmOpensimFittingProcedure(
model) # procedure
cgmFittingProcedure.updateMarkerWeight("LASI", weights["LASI"])
cgmFittingProcedure.updateMarkerWeight("RASI", weights["RASI"])
cgmFittingProcedure.updateMarkerWeight("LPSI", weights["LPSI"])
cgmFittingProcedure.updateMarkerWeight("RPSI", weights["RPSI"])
cgmFittingProcedure.updateMarkerWeight("RTHI", weights["RTHI"])
cgmFittingProcedure.updateMarkerWeight("RKNE", weights["RKNE"])
cgmFittingProcedure.updateMarkerWeight("RTIB", weights["RTIB"])
cgmFittingProcedure.updateMarkerWeight("RANK", weights["RANK"])
cgmFittingProcedure.updateMarkerWeight("RHEE", weights["RHEE"])
cgmFittingProcedure.updateMarkerWeight("RTOE", weights["RTOE"])
cgmFittingProcedure.updateMarkerWeight("LTHI", weights["LTHI"])
cgmFittingProcedure.updateMarkerWeight("LKNE", weights["LKNE"])
cgmFittingProcedure.updateMarkerWeight("LTIB", weights["LTIB"])
cgmFittingProcedure.updateMarkerWeight("LANK", weights["LANK"])
cgmFittingProcedure.updateMarkerWeight("LHEE", weights["LHEE"])
cgmFittingProcedure.updateMarkerWeight("LTOE", weights["LTOE"])
cgmFittingProcedure.updateMarkerWeight("LTHAP", weights["LTHAP"])
cgmFittingProcedure.updateMarkerWeight("LTHAD", weights["LTHAD"])
cgmFittingProcedure.updateMarkerWeight("LTIAP", weights["LTIAP"])
cgmFittingProcedure.updateMarkerWeight("LTIAD", weights["LTIAD"])
cgmFittingProcedure.updateMarkerWeight("RTHAP", weights["RTHAP"])
cgmFittingProcedure.updateMarkerWeight("RTHAD", weights["RTHAD"])
cgmFittingProcedure.updateMarkerWeight("RTIAP", weights["RTIAP"])
cgmFittingProcedure.updateMarkerWeight("RTIAD", weights["RTIAD"])
cgmFittingProcedure.updateMarkerWeight("LSMH", weights["LSMH"])
cgmFittingProcedure.updateMarkerWeight("LFMH", weights["LFMH"])
cgmFittingProcedure.updateMarkerWeight("LVMH", weights["LVMH"])
cgmFittingProcedure.updateMarkerWeight("RSMH", weights["RSMH"])
cgmFittingProcedure.updateMarkerWeight("RFMH", weights["RFMH"])
cgmFittingProcedure.updateMarkerWeight("RVMH", weights["RVMH"])
# cgmFittingProcedure.updateMarkerWeight("LTHL",weights["LTHL"])
# cgmFittingProcedure.updateMarkerWeight("LTHLD",weights["LTHLD"])
# cgmFittingProcedure.updateMarkerWeight("LPAT",weights["LPAT"])
# cgmFittingProcedure.updateMarkerWeight("LTIBL",weights["LTIBL"])
# cgmFittingProcedure.updateMarkerWeight("RTHL",weights["RTHL"])
# cgmFittingProcedure.updateMarkerWeight("RTHLD",weights["RTHLD"])
# cgmFittingProcedure.updateMarkerWeight("RPAT",weights["RPAT"])
# cgmFittingProcedure.updateMarkerWeight("RTIBL",weights["RTIBL"])
osrf = opensimFilters.opensimFittingFilter(iksetupFile, scalingOsim,
cgmFittingProcedure,
DATA_PATH, acqGait)
osrf.setTimeRange(acqGait, beginFrame=vff, lastFrame=vlf)
if "ikAccuracy" in kwargs.keys():
osrf.setAccuracy(kwargs["ikAccuracy"])
LOGGER.logger.info("-------INVERSE KINEMATICS IN PROGRESS----------")
try:
acqIK = osrf.run(DATA_PATH + reconstructFilenameLabelled,
progressionAxis=progressionAxis,
forwardProgression=forwardProgression)
LOGGER.logger.info("[pyCGM2] - IK solver complete")
except:
LOGGER.logger.error("[pyCGM2] - IK solver fails")
acqIK = acqGait
detectAnomaly = True
LOGGER.logger.info(
"---------------------------------------------------")
# eventual gait acquisition to consider for joint kinematics
finalAcqGait = acqIK if ik_flag else acqGait
if "displayCoordinateSystem" in kwargs.keys(
) and kwargs["displayCoordinateSystem"]:
csp = modelFilters.ModelCoordinateSystemProcedure(model)
csdf = modelFilters.CoordinateSystemDisplayFilter(
csp, model, finalAcqGait)
csdf.setStatic(False)
csdf.display()
# --- final pyCGM2 model motion Filter ---
# use fitted markers
modMotionFitted = modelFilters.ModelMotionFilter(
scp,
finalAcqGait,
model,
enums.motionMethod.Sodervisk,
markerDiameter=markerDiameter,
forceFoot6DoF=forceFoot6DoF_flag)
modMotionFitted.compute()
#---- Joint kinematics----
# relative angles
modelFilters.ModelJCSFilter(model, finalAcqGait).compute(
description="vectoriel", pointLabelSuffix=pointSuffix)
modelFilters.ModelAbsoluteAnglesFilter(
model,
finalAcqGait,
segmentLabels=["Left Foot", "Right Foot", "Pelvis", "Thorax", "Head"],
angleLabels=[
"LFootProgress", "RFootProgress", "Pelvis", "Thorax", "Head"
],
eulerSequences=["TOR", "TOR", "ROT", "YXZ", "TOR"],
globalFrameOrientation=globalFrame,
forwardProgression=forwardProgression).compute(
pointLabelSuffix=pointSuffix)
#---- Body segment parameters----
bspModel = bodySegmentParameters.Bsp(model)
bspModel.compute()
modelFilters.CentreOfMassFilter(
model, finalAcqGait).compute(pointLabelSuffix=pointSuffix)
# Inverse dynamics
if btkTools.checkForcePlateExist(acqGait):
# --- force plate handling----
# find foot in contact
mappedForcePlate = forceplates.matchingFootSideOnForceplate(
finalAcqGait, mfpa=mfpa)
forceplates.addForcePlateGeneralEvents(finalAcqGait, mappedForcePlate)
LOGGER.logger.warning("Manual Force plate assignment : %s" %
mappedForcePlate)
# assembly foot and force plate
modelFilters.ForcePlateAssemblyFilter(
model,
finalAcqGait,
mappedForcePlate,
leftSegmentLabel="Left Foot",
rightSegmentLabel="Right Foot").compute(
pointLabelSuffix=pointSuffix)
#---- Joint kinetics----
idp = modelFilters.CGMLowerlimbInverseDynamicProcedure()
modelFilters.InverseDynamicFilter(
model,
finalAcqGait,
procedure=idp,
projection=momentProjection,
globalFrameOrientation=globalFrame,
forwardProgression=forwardProgression).compute(
pointLabelSuffix=pointSuffix)
#---- Joint energetics----
modelFilters.JointPowerFilter(
model, finalAcqGait).compute(pointLabelSuffix=pointSuffix)
#---- zero unvalid frames ---
btkTools.cleanAcq(finalAcqGait)
btkTools.applyOnValidFrames(finalAcqGait, flag)
if detectAnomaly and not anomalyException:
LOGGER.logger.error(
"Anomalies has been detected - Check Warning messages of the log file"
)
return finalAcqGait, detectAnomaly
def calibrate(DATA_PATH, calibrateFilenameLabelled, translators, weights,
required_mp, optional_mp, ik_flag, leftFlatFoot, rightFlatFoot,
headFlat, markerDiameter, hjcMethod, pointSuffix, **kwargs):
"""
Calibration of the CGM2.5
:param DATA_PATH [str]: path to your data
:param calibrateFilenameLabelled [str]: c3d file
:param translators [dict]: translators to apply
:param required_mp [dict]: required anthropometric data
:param optional_mp [dict]: optional anthropometric data (ex: LThighOffset,...)
:param ik_flag [bool]: enable the inverse kinematic solver
:param leftFlatFoot [bool]: enable of the flat foot option for the left foot
:param rightFlatFoot [bool]: enable of the flat foot option for the right foot
:param headFlat [bool]: enable of the head flat option
:param markerDiameter [double]: marker diameter (mm)
:param hjcMethod [str or list of 3 float]: method for locating the hip joint centre
:param pointSuffix [str]: suffix to add to model outputs
"""
detectAnomaly = False
if "anomalyException" in kwargs.keys():
anomalyException = kwargs["anomalyException"]
else:
anomalyException = False
if "Fitting" in weights.keys():
weights = weights["Fitting"]["Weight"]
# ---btk acquisition---
if "forceBtkAcq" in kwargs.keys():
acqStatic = kwargs["forceBtkAcq"]
else:
acqStatic = btkTools.smartReader(
(DATA_PATH + calibrateFilenameLabelled))
btkTools.checkMultipleSubject(acqStatic)
if btkTools.isPointExist(acqStatic, "SACR"):
translators["LPSI"] = "SACR"
translators["RPSI"] = "SACR"
LOGGER.logger.info("[pyCGM2] Sacrum marker detected")
acqStatic = btkTools.applyTranslators(acqStatic, translators)
trackingMarkers = cgm2.CGM2_5.LOWERLIMB_TRACKING_MARKERS + cgm2.CGM2_5.THORAX_TRACKING_MARKERS + cgm2.CGM2_5.UPPERLIMB_TRACKING_MARKERS
actual_trackingMarkers, phatoms_trackingMarkers = btkTools.createPhantoms(
acqStatic, trackingMarkers)
vff = acqStatic.GetFirstFrame()
vlf = acqStatic.GetLastFrame()
# vff,vlf = btkTools.getFrameBoundaries(acqStatic,actual_trackingMarkers)
flag = btkTools.getValidFrames(acqStatic,
actual_trackingMarkers,
frameBounds=[vff, vlf])
gapFlag = btkTools.checkGap(acqStatic,
actual_trackingMarkers,
frameBounds=[vff, vlf])
if gapFlag:
raise Exception(
"[pyCGM2] Calibration aborted. Gap find during interval [%i-%i]. Crop your c3d "
% (vff, vlf))
# --------------------ANOMALY------------------------------
# --Check MP
adap = AnomalyDetectionProcedure.AnthropoDataAnomalyProcedure(required_mp)
adf = AnomalyFilter.AnomalyDetectionFilter(None, None, adap)
mp_anomaly = adf.run()
if mp_anomaly["ErrorState"]: detectAnomaly = True
# --marker presence
markersets = [
cgm2.CGM2_5.LOWERLIMB_TRACKING_MARKERS,
cgm2.CGM2_5.THORAX_TRACKING_MARKERS,
cgm2.CGM2_5.UPPERLIMB_TRACKING_MARKERS
]
for markerset in markersets:
ipdp = InspectorProcedure.MarkerPresenceDetectionProcedure(markerset)
idf = InspectorFilter.InspectorFilter(acqStatic,
calibrateFilenameLabelled, ipdp)
inspector = idf.run()
# # --marker outliers
if inspector["In"] != []:
madp = AnomalyDetectionProcedure.MarkerAnomalyDetectionRollingProcedure(
inspector["In"], plot=False, window=4, threshold=3)
adf = AnomalyFilter.AnomalyDetectionFilter(
acqStatic, calibrateFilenameLabelled, madp)
anomaly = adf.run()
anomalyIndexes = anomaly["Output"]
if anomaly["ErrorState"]: detectAnomaly = True
if detectAnomaly and anomalyException:
raise Exception(
"Anomalies has been detected - Check Warning message of the log file"
)
# --------------------MODELLING------------------------------
# ---check marker set used----
dcm = cgm.CGM.detectCalibrationMethods(acqStatic)
# --------------------------MODEL--------------------------------------
# ---definition---
model = cgm2.CGM2_5()
model.configure(detectedCalibrationMethods=dcm)
model.addAnthropoInputParameters(required_mp, optional=optional_mp)
if dcm["Left Knee"] == enums.JointCalibrationMethod.KAD:
actual_trackingMarkers.append("LKNE")
if dcm["Right Knee"] == enums.JointCalibrationMethod.KAD:
actual_trackingMarkers.append("RKNE")
model.setStaticTrackingMarkers(actual_trackingMarkers)
# --store calibration parameters--
model.setStaticFilename(calibrateFilenameLabelled)
model.setCalibrationProperty("leftFlatFoot", leftFlatFoot)
model.setCalibrationProperty("rightFlatFoot", rightFlatFoot)
model.setCalibrationProperty("headFlat", headFlat)
model.setCalibrationProperty("markerDiameter", markerDiameter)
# --------------------------STATIC CALBRATION--------------------------
scp = modelFilters.StaticCalibrationProcedure(
model) # load calibration procedure
# ---initial calibration filter----
# use if all optional mp are zero
modelFilters.ModelCalibrationFilter(
scp,
acqStatic,
model,
leftFlatFoot=leftFlatFoot,
rightFlatFoot=rightFlatFoot,
headFlat=headFlat,
markerDiameter=markerDiameter,
).compute()
# ---- Decorators -----
decorators.applyBasicDecorators(dcm, model, acqStatic, optional_mp,
markerDiameter)
decorators.applyHJCDecorators(model, hjcMethod)
# ----Final Calibration filter if model previously decorated -----
if model.decoratedModel:
# initial static filter
modelFilters.ModelCalibrationFilter(
scp,
acqStatic,
model,
leftFlatFoot=leftFlatFoot,
rightFlatFoot=rightFlatFoot,
headFlat=headFlat,
markerDiameter=markerDiameter).compute()
# ----------------------CGM MODELLING----------------------------------
# ----motion filter----
modMotion = modelFilters.ModelMotionFilter(scp,
acqStatic,
model,
enums.motionMethod.Sodervisk,
markerDiameter=markerDiameter)
modMotion.compute()
# ----progression Frame----
progressionFlag = False
if btkTools.isPointsExist(acqStatic, ['LASI', 'RASI', 'RPSI', 'LPSI'],
ignorePhantom=False):
LOGGER.logger.info(
"[pyCGM2] - progression axis detected from Pelvic markers ")
pfp = progressionFrame.PelvisProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acqStatic, pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
progressionFlag = True
elif btkTools.isPointsExist(acqStatic, ['C7', 'T10', 'CLAV', 'STRN'],
ignorePhantom=False) and not progressionFlag:
LOGGER.logger.info(
"[pyCGM2] - progression axis detected from Thoracic markers ")
pfp = progressionFrame.ThoraxProgressionFrameProcedure()
pff = progressionFrame.ProgressionFrameFilter(acqStatic, pfp)
pff.compute()
progressionAxis = pff.outputs["progressionAxis"]
globalFrame = pff.outputs["globalFrame"]
forwardProgression = pff.outputs["forwardProgression"]
else:
globalFrame = "XYZ"
progressionAxis = "X"
forwardProgression = True
LOGGER.logger.error(
"[pyCGM2] - impossible to detect progression axis - neither pelvic nor thoracic markers are present. Progression set to +X by default "
)
# ----manage IK Targets----
ikTargets = list()
for target in weights.keys():
if target not in actual_trackingMarkers:
weights[target] = 0
LOGGER.logger.warning(
"[pyCGM2] - the IK targeted marker [%s] is not labelled in the acquisition [%s]"
% (target, calibrateFilenameLabelled))
else:
ikTargets.append(target)
model.setStaticIkTargets(ikTargets)
if "noKinematicsCalculation" in kwargs.keys(
) and kwargs["noKinematicsCalculation"]:
LOGGER.logger.warning(
"[pyCGM2] No Kinematic calculation done for the static file")
return model, acqStatic, detectAnomaly
else:
if ik_flag:
# ---OPENSIM IK---
# --- opensim calibration Filter ---
osimfile = pyCGM2.OPENSIM_PREBUILD_MODEL_PATH + "models\\osim\\lowerLimb_ballsJoints.osim" # osimfile
markersetFile = pyCGM2.OPENSIM_PREBUILD_MODEL_PATH + "models\\settings\\cgm2_4\\cgm2_4-markerset.xml" # markerset
cgmCalibrationprocedure = opensimFilters.CgmOpensimCalibrationProcedures(
model) # procedure
oscf = opensimFilters.opensimCalibrationFilter(
osimfile, model, cgmCalibrationprocedure, DATA_PATH)
oscf.addMarkerSet(markersetFile)
scalingOsim = oscf.build()
# --- opensim Fitting Filter ---
iksetupFile = pyCGM2.OPENSIM_PREBUILD_MODEL_PATH + "models\\settings\\cgm2_4\\cgm2_4-ikSetUp_template.xml" # ik tool file
cgmFittingProcedure = opensimFilters.CgmOpensimFittingProcedure(
model) # procedure
cgmFittingProcedure.updateMarkerWeight("LASI", weights["LASI"])
cgmFittingProcedure.updateMarkerWeight("RASI", weights["RASI"])
cgmFittingProcedure.updateMarkerWeight("LPSI", weights["LPSI"])
cgmFittingProcedure.updateMarkerWeight("RPSI", weights["RPSI"])
cgmFittingProcedure.updateMarkerWeight("RTHI", weights["RTHI"])
cgmFittingProcedure.updateMarkerWeight("RKNE", weights["RKNE"])
cgmFittingProcedure.updateMarkerWeight("RTIB", weights["RTIB"])
cgmFittingProcedure.updateMarkerWeight("RANK", weights["RANK"])
cgmFittingProcedure.updateMarkerWeight("RHEE", weights["RHEE"])
cgmFittingProcedure.updateMarkerWeight("RTOE", weights["RTOE"])
cgmFittingProcedure.updateMarkerWeight("LTHI", weights["LTHI"])
cgmFittingProcedure.updateMarkerWeight("LKNE", weights["LKNE"])
cgmFittingProcedure.updateMarkerWeight("LTIB", weights["LTIB"])
cgmFittingProcedure.updateMarkerWeight("LANK", weights["LANK"])
cgmFittingProcedure.updateMarkerWeight("LHEE", weights["LHEE"])
cgmFittingProcedure.updateMarkerWeight("LTOE", weights["LTOE"])
cgmFittingProcedure.updateMarkerWeight("LTHAP", weights["LTHAP"])
cgmFittingProcedure.updateMarkerWeight("LTHAD", weights["LTHAD"])
cgmFittingProcedure.updateMarkerWeight("LTIAP", weights["LTIAP"])
cgmFittingProcedure.updateMarkerWeight("LTIAD", weights["LTIAD"])
cgmFittingProcedure.updateMarkerWeight("RTHAP", weights["RTHAP"])
cgmFittingProcedure.updateMarkerWeight("RTHAD", weights["RTHAD"])
cgmFittingProcedure.updateMarkerWeight("RTIAP", weights["RTIAP"])
cgmFittingProcedure.updateMarkerWeight("RTIAD", weights["RTIAD"])
cgmFittingProcedure.updateMarkerWeight("LSMH", weights["LSMH"])
cgmFittingProcedure.updateMarkerWeight("LFMH", weights["LFMH"])
cgmFittingProcedure.updateMarkerWeight("LVMH", weights["LVMH"])
cgmFittingProcedure.updateMarkerWeight("RSMH", weights["RSMH"])
cgmFittingProcedure.updateMarkerWeight("RFMH", weights["RFMH"])
cgmFittingProcedure.updateMarkerWeight("RVMH", weights["RVMH"])
# cgmFittingProcedure.updateMarkerWeight("LTHL",weights["LTHL"])
# cgmFittingProcedure.updateMarkerWeight("LTHLD",weights["LTHLD"])
# cgmFittingProcedure.updateMarkerWeight("LPAT",weights["LPAT"])
# cgmFittingProcedure.updateMarkerWeight("LTIBL",weights["LTIBL"])
# cgmFittingProcedure.updateMarkerWeight("RTHL",weights["RTHL"])
# cgmFittingProcedure.updateMarkerWeight("RTHLD",weights["RTHLD"])
# cgmFittingProcedure.updateMarkerWeight("RPAT",weights["RPAT"])
# cgmFittingProcedure.updateMarkerWeight("RTIBL",weights["RTIBL"])
osrf = opensimFilters.opensimFittingFilter(iksetupFile,
scalingOsim,
cgmFittingProcedure,
DATA_PATH,
acqStatic,
accuracy=1e-5)
LOGGER.logger.info(
"-------INVERSE KINEMATICS IN PROGRESS----------")
try:
acqStaticIK = osrf.run(DATA_PATH + calibrateFilenameLabelled,
progressionAxis=progressionAxis,
forwardProgression=forwardProgression)
LOGGER.logger.info("[pyCGM2] - IK solver complete")
except:
LOGGER.logger.error("[pyCGM2] - IK solver fails")
acqStaticIK = acqStatic
detectAnomaly = True
LOGGER.logger.info(
"-----------------------------------------------")
# eventual static acquisition to consider for joint kinematics
finalAcqStatic = acqStaticIK if ik_flag else acqStatic
# --- final pyCGM2 model motion Filter ---
# use fitted markers
modMotionFitted = modelFilters.ModelMotionFilter(
scp, finalAcqStatic, model, enums.motionMethod.Sodervisk)
modMotionFitted.compute()
if "displayCoordinateSystem" in kwargs.keys(
) and kwargs["displayCoordinateSystem"]:
csp = modelFilters.ModelCoordinateSystemProcedure(model)
csdf = modelFilters.CoordinateSystemDisplayFilter(
csp, model, finalAcqStatic)
csdf.setStatic(False)
csdf.display()
#---- Joint kinematics----
# relative angles
modelFilters.ModelJCSFilter(model, finalAcqStatic).compute(
description="vectoriel", pointLabelSuffix=pointSuffix)
modelFilters.ModelAbsoluteAnglesFilter(
model,
finalAcqStatic,
segmentLabels=[
"Left Foot", "Right Foot", "Pelvis", "Thorax", "Head"
],
angleLabels=[
"LFootProgress", "RFootProgress", "Pelvis", "Thorax", "Head"
],
eulerSequences=["TOR", "TOR", "ROT", "YXZ", "TOR"],
globalFrameOrientation=globalFrame,
forwardProgression=forwardProgression).compute(
pointLabelSuffix=pointSuffix)
# BSP model
bspModel = bodySegmentParameters.Bsp(model)
bspModel.compute()
modelFilters.CentreOfMassFilter(
model, finalAcqStatic).compute(pointLabelSuffix=pointSuffix)
btkTools.cleanAcq(finalAcqStatic)
if detectAnomaly and not anomalyException:
LOGGER.logger.error(
"Anomalies has been detected - Check Warning messages of the log file"
)
return model, finalAcqStatic, detectAnomaly