Gait-Library-HARLab

使用说明

环境搭建

下载数据集，下载Opensim，注意Opensim使用默认路径，不要自定义安装

解压fcgm3chfff-1，解压scripts，解压AB06等数据集，文件路径如下

fcgm3chfff-1
-AB06
    --10_09_18
        ---levelground
        ---ramp
        ---stair
        ---treadmill
-AB07
    --10_14_18
        ---levelground
        ---ramp
        ---stair
        ---treadmill
......

-scripts
    --EpicToolbox
    --lib
    --MoCapTools
    --*.m
    
SubjectInfo.mat

运行EpicToobox文件夹下的install.m文件

Installing EPICTOOLBOX...
Scripts added to path...
Path saved

运行MoCapTools文件架下的install.m文件

Scripts added to path...
Path saved...
If you are new to using this repo, please see +Osim/examples/example.m for information on how to use OpenSim
via MATLAB, or +Vicon/examples/example.m for information on automated iterative gap-filling via MATLAB.
Please restart MATLAB.

Matlab脚本基本使用方法

将仓库中My_Scripts的文件放在fcgm3chfff-1/scripts中运行

利用EpicToolBox进行基本数据读取

init();def=defaults;
SUBJECT = 'AB06';%选择受试者
AMBULATION = 'treadmill';%选择地形
TRIAL = {'treadmill_02_01.mat','treadmill_01_01.mat'};%选择实验对应文件名称
SENSORS = {'markers','gcLeft','gcRight','conditions','ik','id','emg','imu','gon','jp'};%选择需要的种类
%获取所有同名实验不同类型数据的路径
allfiles = f.fileList("Subject",SUBJECT,"Mode",AMBULATION,"Trial",TRIAL,"Sensor",SENSORS);
%每个实验对应的数据被封装在一个trial结构体内部
trials = f.EpicToolbox(allfiles);
%选择第一次实验，treadmill_02_01的数据
trial_1 = trails{1};

可以通过trial_1读取对应类型的数据

通过SubjectInfo.mat获取患者信息

在fcgm**文件夹解压后有SubjectInfo.mat文件，内部包含用户身高体重等数据

init();def = defaults;
subjectLists = load(def.SUBJECT_INFO_FILE);

通过gcLeft/gcRight划分步态

gcTopic定义了HeelStrike和Toeoff

% 通过右脚HeelStrike划分步态
allstrides = [];
for i =1:numel(trials)
    trial = trials{i};
    strides = segment_gc(trial,'GCtopic','gcRight','GCchannel','HeelStrike');
    fprintf("\t %d gaits in treadmill_0%d \n",size(strides,1),i);
    allstrides = [allstrides;strides];
end
fprintf('\t Totally %d gaits\n',size(allstrides,1));
% Header归一化
%% 归一化步态对应的时间为0-1，对应0-100%
allstrides = Topics.normalize(allstrides,Topics.topics(allstrides),'Header');
%% interpolate将根据Header的0-1对数据进行插值
allstrides = Topics.interpolate(allstrides,0:0.01:1);
%% 对于ID和JP的数据往往需要根据用户体重进行归一化
allstrides = Topics.transform(@(x)(x/subjectWeight),allstrides,{'id','jp'});

获得均匀速度的数据

跑步机速度变换稳定时的数据标记为treadmill

alllabels = cell(size(allstrides));
for i=1:numel(allstrides)
    stride = allstrides{i};
    stride_condition = stride.conditions;
    if strcmp(ambulationModes,'treadmill')
        if numel(unique(stride_condition.speed.Speed))>2
            alllabels{i}='discard';%速度不稳定，无法分类
        else
            alllabels{i}='treadmill';%速度
        end
        continue;
    end
end

关于数据集

数据集包含四种地形，每种地形下采集如下信息，且对应同一组实验的不同种数据名相同，存储在不同文件夹中。

conditions

Experiment description, locomotion mode labels,terrain condition (walking speed, ramp incline, or stair height).Sampled at 1000Hz.

emg

Electromyography from 11 muscles. Sampled at 1000Hz, bandpass filtered (20-400Hz).

fp

Ground reaction force data of all force plates used in a trial.Sampled at 1000Hz.

gcLeft/gcRight

Gait cycle segmented by heel strike or toe off of left/right foot. Sampled at 200Hz

id

Inverse dynamics calculated by OpenSim using ground reaction force and motion capture kinematic data. Sampled at 200Hz.

ik

Inverse kinematics calculated by OpenSim using motion capture data. Sampled at 200Hz.

imu

Inertial Measurement Unit data from trunk, thigh, shank, and foot segments. Sampled at 200Hz.

jp

Estimation of instantaneous joint power using joint moment and angular velocity. Sampled at 200Hz.

markers

Marker trajectories from motion capture. Sampled at 200Hz.

Seas00n/Gait-Library-HARLab