Human motion function evaluation method and human motion function evaluation system

Abstract

A method and system for assessing human motor function. At least one image sequence is captured from a subject undergoing a clinical motor function test using at least one image capturing device. This clinical motor function test includes multiple movement stages. Skeletal analysis is performed on the at least one image sequence to generate skeletal motion data. This skeletal motion data includes human skeletal information from multiple pose frames. Based on the human skeletal information from the multiple pose frames, the skeletal motion data is divided into multiple skeletal motion data segments, each associated with a different movement stage. Based on multiple pose frames of at least one of the multiple skeletal motion data segments, multiple motor ability assessment indicators for the subject are determined. A clinical assessment report including multiple motor ability assessment indicators is output.

Claims

1. A method for assessing human motor function, suitable for execution by a processor, the method comprising: The method involves: capturing at least one image sequence from a subject performing a clinical motor function test using at least one image capturing device, wherein the clinical motor function test includes multiple movement stages; performing skeletal analysis on the at least one image sequence to generate skeletal motion data, wherein the skeletal motion data includes human skeletal information from multiple pose frames; dividing the skeletal motion data into multiple skeletal motion data segments respectively associated with the multiple movement stages based on the human skeletal information from the multiple pose frames of at least one of the multiple skeletal motion data segments; determining multiple motor ability assessment indicators for the subject based on the multiple pose frames of at least one of the multiple skeletal motion data segments; and outputting a clinical assessment report including the multiple motor ability assessment indicators.

2. The human motor function assessment method as described in claim 1, wherein the clinical motor function test includes the Timed Up and Go (TUG) test.

3. The human motor function assessment method as described in claim 2, wherein the plurality of skeletal motion data segments include a first skeletal motion data segment associated with a walking phase among the plurality of motion phases, and the step of determining the plurality of motor ability assessment indicators of the subject based on the plurality of posture frames of at least one of the plurality of skeletal motion data segments includes: By analyzing the human skeleton information of the multiple pose frames of the first skeleton motion data segment, the gait parameters of the test subject are determined, wherein the multiple motion ability assessment indicators include the gait parameters.

4. The human motor function assessment method as described in claim 3, wherein the walking phase includes an outward walking phase and a return walking phase, and the gait parameters include a first gait parameter corresponding to the outward walking phase and a second gait parameter corresponding to the return walking phase.

5. The human motion function assessment method as described in claim 3, wherein the step of determining the gait parameters of the subject by analyzing the human skeleton information of the multiple pose frames of the first skeleton motion data segment includes: Based on the coordinates of a foot key point in each of the multiple posture frames of the first skeleton motion data segment, the number of steps, step length, step frequency, and walking speed of the walking phase are determined.

6. The human motion function assessment method as described in claim 3, wherein the step of determining the gait parameters of the subject by analyzing the human skeleton information of the multiple pose frames of the first skeleton motion data segment includes: Based on the human skeleton information of the multiple posture frames of the first skeleton motion data segment, multiple left foot landing periods and multiple right foot landing periods are determined; based on the intersection between the multiple left foot landing periods and the multiple right foot landing periods, multiple simultaneous foot landing periods are determined; and the duration of the multiple simultaneous foot landing periods is accumulated to obtain the foot support time associated with the walking phase in the gait parameters.

7. The human motion function assessment method as described in claim 3, wherein the step of determining the gait parameters of the subject by analyzing the human skeleton information of the multiple pose frames of the first skeleton motion data segment includes: Based on the human skeleton information of the multiple posture frames of the first skeleton motion data segment, determine the joint angle of a knee joint in each of the multiple posture frames of the first skeleton motion data segment; based on the joint angle of the knee joint in each of the multiple posture frames of the first skeleton motion data segment, obtain the trend of multiple joint angle changes corresponding to multiple walking steps; and based on the comparison result of the multiple joint angle change trends, determine a knee stability index in the gait parameters.

8. The human motion function assessment method as described in claim 7, wherein the step of determining the gait parameters of the subject by analyzing the human skeleton information of the multiple pose frames of the first skeleton motion data segment further includes: A body balance index among the gait parameters is determined based on the knee stability index corresponding to the knee joint and another knee stability index corresponding to the other knee joint.

9. The human motor function assessment method as described in claim 2, wherein the plurality of skeletal motion data segments include a second skeletal motion data segment associated with a rotation phase among the plurality of motion phases, and the step of determining the plurality of motor ability assessment indicators of the subject based on the plurality of pose frames of at least one of the plurality of skeletal motion data segments includes: The motion time of the rotation phase is calculated based on the number of frames in the multiple pose frames of the second skeleton motion data segment. And by analyzing the human skeleton information of the multiple pose frames of the second skeleton motion data segment, the rotation motion parameters of the rotation phase are determined, wherein the multiple motion capability evaluation indicators include the motion time of the rotation phase and the rotation motion parameters.

10. The human motor function assessment method as claimed in claim 2, wherein the plurality of skeletal motion data segments include a third skeletal motion data segment associated with a sitting phase among the plurality of motion phases, and the step of determining the plurality of motor ability assessment indicators of the subject based on the plurality of posture frames of at least one of the plurality of skeletal motion data segments includes: The time consumed by the sitting phase is calculated based on the number of frames in the multiple posture frames in the third skeleton motion data segment. And by analyzing the human skeleton information of the multiple pose frames of the third skeleton motion data segment, the sitting motion parameters of the sitting phase are determined, wherein the multiple motion capability evaluation indicators include the motion time of the sitting phase and the sitting motion parameters.

11. The human motor function assessment method as described in claim 2, wherein the plurality of skeletal motion data segments include a fourth skeletal motion data segment associated with a standing phase among the plurality of motion phases, and the step of determining the plurality of motor ability assessment indicators of the subject based on the plurality of pose frames of at least one of the plurality of skeletal motion data segments includes: The time consumed by the standing-up phase is calculated based on the number of frames in the multiple posture frames in the fourth skeleton motion data segment. And by analyzing the human skeleton information of the multiple posture frames of the fourth skeleton motion data segment, the standing motion parameters of the standing phase are determined, wherein the multiple motion capability evaluation indicators include the motion time of the standing phase and the standing motion parameters.

12. The human motor function assessment method as described in claim 2, wherein the step of determining the multiple motor ability assessment indicators of the subject based on the multiple pose frames of at least one of the multiple skeletal motion data segments includes: Based on the human skeleton information of at least one of the plurality of skeleton motion data segments in the plurality of pose frames, determine the position of the body center of gravity in each of the plurality of pose frames of the plurality of skeleton motion data segments; based on the position of the body center of gravity in each of the plurality of pose frames of the plurality of skeleton motion data segments, obtain a center of gravity movement trajectory; and based on the degree of change of the center of gravity movement trajectory relative to the horizontal plane, determine a stability index.

13. A human motor function assessment system, comprising: At least one image capturing device; a storage device; A processor, coupled to the at least one image capturing device and the storage device, is configured to perform: capturing at least one image sequence from a subject performing a clinical motor function test via the at least one image capturing device, wherein the clinical motor function test includes multiple movement stages; performing skeletal analysis on the at least one image sequence to generate skeletal motion data, wherein the skeletal motion data includes human skeletal information from multiple pose frames; dividing the skeletal motion data into multiple skeletal motion data segments respectively associated with the multiple movement stages based on the human skeletal information from the multiple pose frames; determining multiple motor ability assessment indicators for the subject based on the multiple pose frames of at least one of the multiple skeletal motion data segments; and outputting a clinical assessment report including the multiple motor ability assessment indicators.

14. The human motor function assessment system as described in claim 13, wherein the clinical motor function test includes the Timed Up and Go (TUG) test.

15. The human motion function assessment system of claim 14, wherein the plurality of skeletal motion data segments include a first skeletal motion data segment associated with a walking phase among the plurality of motion phases, and the processor is configured to perform: determining a gait parameter of the subject by analyzing the human skeleton information of the plurality of pose frames of the first skeletal motion data segment, wherein the plurality of motion ability assessment indicators include the gait parameter.

16. The human motion function assessment system of claim 15, wherein the processor is configured to perform: determining a number of steps, a step length, a step frequency, and a walking speed of the walking phase based on the coordinates of a foot key point in each of the plurality of posture frames of the first skeletal motion data segment; determining a plurality of left foot landing periods and a plurality of right foot landing periods based on the human skeletal information of the plurality of posture frames of the first skeletal motion data segment; determining a plurality of simultaneous foot landing periods based on the intersection between the plurality of left foot landing periods and the plurality of right foot landing periods; and accumulating the duration of the plurality of simultaneous foot landing periods to obtain a bipedal support time associated with the walking phase in the gait parameters.

17. The human motion function assessment system of claim 15, wherein the processor is configured to perform: determining a joint angle of a knee joint in each of the plurality of posture frames of the first skeletal motion data segment based on human skeletal information from the plurality of posture frames of the first skeletal motion data segment; acquiring multiple joint angle change trends corresponding to multiple walking steps based on the joint angle of the knee joint in each of the plurality of posture frames of the first skeletal motion data segment; determining a knee stability index in the gait parameters based on a comparison result of the multiple joint angle change trends; and determining a body balance index in the gait parameters based on the knee stability index corresponding to the knee joint and another knee stability index corresponding to another knee joint.

18. The human motion function assessment system of claim 14, wherein the plurality of skeletal motion data segments include a second skeletal motion data segment associated with a rotation phase among the plurality of motion phases, the processor being configured to perform: calculating the motion time of the rotation phase based on the number of frames in the plurality of pose frames of the second skeletal motion data segment; and determining rotation motion parameters of the rotation phase by analyzing human skeletal information in the plurality of pose frames of the second skeletal motion data segment, wherein the plurality of motion capability assessment metrics include the motion time of the rotation phase and the rotation motion parameters.

19. The human motion function assessment system of claim 14, wherein the plurality of skeletal motion data segments include a third skeletal motion data segment associated with a sitting phase of the plurality of motion phases, and the plurality of skeletal motion data segments include a fourth skeletal motion data segment associated with a standing phase of the plurality of motion phases, wherein the processor is configured to perform: calculating the motion time of the sitting phase based on the number of frames of the plurality of posture frames in the third skeletal motion data segment; determining the sitting motion parameters of the sitting phase by analyzing human skeletal information of the plurality of posture frames of the third skeletal motion data segment, wherein the plurality of motion ability assessment indicators include the motion time of the sitting phase and the sitting motion parameters; calculating the motion time of the standing phase based on the number of frames of the plurality of posture frames in the fourth skeletal motion data segment; and determining the standing motion parameters of the standing phase by analyzing human skeletal information of the plurality of posture frames of the fourth skeletal motion data segment, wherein the plurality of motion ability assessment indicators include the motion time of the standing phase and the standing motion parameters.

20. The human motion function assessment system of claim 14, wherein the processor is configured to perform: determining the body center of gravity position in each of the plurality of posture frames of the plurality of skeleton motion data segments based on the human skeleton information of the plurality of posture frames of at least one of the plurality of skeleton motion data segments; acquiring a center of gravity movement trajectory based on the body center of gravity position in each of the plurality of posture frames of the plurality of skeleton motion data segments; and determining a stability index based on the degree of change of the center of gravity movement trajectory relative to the horizontal plane.

Images