Estimation device, estimation method, and program recording medium

1 . An estimation device comprising:

an inertial measurement sensor installed on a foot portion, the inertial measurement sensor comprising a three-axis accelerometer and a three-axis gyroscope configured to measure spatial acceleration and spatial angular velocity and output time-series data based on the spatial acceleration and the spatial angular velocity;

at least one memory storing instructions; and

at least one processor connected to the at least one memory and configured to execute the instructions to:

detect a terminal stance period from the time-series data based on a physical quantity related to movement of a foot measured by the inertial measurement sensor provided at the foot portion, the detecting the terminal stance period from the time-series data comprising converting the time-series data from a local coordinate system to a world coordinate system and detecting the terminal stance period as a period from heel lift to opposite heel strike based on an inflection point of roll angular acceleration;

extract a feature amount from an angular waveform in a coronal plane during the terminal stance period, the angular waveform being a pitch angle about a Y-axis in the coronal plane (ZX plane), and the feature amount comprising at least one of an average value, a weighted average value, and an integral value within the terminal stance period;

estimate a degree of pronation/supination of the foot by inputting the feature amount extracted from the angular waveform in the coronal plane to a presumption model, the presumption model generated by learning a data set in which the feature amount extracted from the angular waveform in the coronal plane is an explanatory variable and a center of pressure excursion index (CPEI) obtained from a foot pressure distribution measured by a pressure sensor is an objective variable; and

output the estimation result regarding the degree of pronation/supination of the foot.

2 . The estimation device according to claim 1 , wherein

the at least one processor is configured to execute the instructions to

output the estimation result indicating one of pronation/supination of the foot and a normal foot according to a value of the center of pressure excursion index.

3 . The estimation device according to claim 2 , wherein

the at least one processor is configured to execute the instructions to

output the estimation result indicating supination when the center of pressure excursion index value is equal to or more than 20,

output the estimation result indicating normality when the center of pressure excursion index value is equal to or more than 9 and less than 20, and

output the estimation result indicating pronation when the center of pressure excursion index value is less than 9.

4 . The estimation device according to claim 1 , wherein

the at least one processor is configured to execute the instructions to

extract a gait waveform for one gait cycle starting from heel strike, from the time-series data of the sensor data, and

detect a period of 30 to 50% of the extracted gait waveform as the terminal stance period.

5 . The estimation device according to claim 1 , wherein

the at least one processor is configured to execute the instructions to

detect timing of heel lift and timing of an opposite heel strike from the time-series data of the sensor data, and

detect a period from the timing of heel lift to the timing of the opposite heel strike is detected as the terminal stance period.

6 . An estimation system comprising:

the estimation device according to claim 1 ; and

a data acquisition device configured to measure a spatial acceleration and a spatial angular velocity, generate sensor data based on the measured spatial acceleration and spatial angular velocity, and transmit the generated sensor data to the estimation device.

7 . The estimation device according to claim 1 ,

wherein the presumption model is a machine learning model, and

wherein the processor is further configured to:

perform classification of a degree of pronation/supination based on an output of the machine learning model, and

based on the classification indicating supination or pronation, to output recommendation information that supports user decision-making to visit a hospital.

8 . An estimation method that causes a computer to execute processes comprising:

measuring, by an inertial measurement sensor including a three-axis accelerometer and a three-axis gyroscope, spatial acceleration and spatial angular velocity and outputting time-series data based on the spatial acceleration and the spatial angular velocity;

detecting a terminal stance period from the time-series data based on a physical quantity related to a movement of a foot measured by the inertial measurement sensor provided at the foot portion, the detecting the terminal stance period from the time-series data comprising converting the sensor data from a local coordinate system to a world coordinate system and detecting the terminal stance period as a period from heel lift to opposite heel strike based on an inflection point of roll angular acceleration;

extracting a feature amount from an angular waveform in a coronal plane during the terminal stance period, the angular waveform being a pitch angle about a Y-axis in the coronal plane (ZX plane), and the feature amount comprising at least one of an average value, a weighted average value, and an integral value within the terminal stance period;

estimating a degree of pronation/supination of the foot by inputting the feature amount extracted from the angular waveform in the coronal plane to a presumption model, the presumption model generated by learning a data set in which the feature amount extracted from the angular waveform in the coronal plane is an explanatory variable and a center of pressure excursion index (CPE) obtained from a foot pressure distribution measured by a pressure sensor is an objective variable; and

outputting the estimation result regarding the degree of pronation/supination of the foot.

9 . The estimation method according to claim 8 , further comprising:

outputting the estimation result indicating one of pronation/supination of the foot and a normal foot according to a value of the center of pressure excursion index.

10 . The estimation method according to claim 9 , further comprising:

outputting the estimation result indicating supination when the center of pressure excursion index value is equal to or more than 20,

outputting the estimation result indicating normality when the center of pressure excursion index value is equal to or more than 9 and less than 20, and

outputting the estimation result indicating pronation when the center of pressure excursion index value is less than 9.

11 . The estimation method according to claim 8 , further comprising:

extracting a gait waveform for one gait cycle starting from heel strike, from the time-series data of the sensor data, and

detecting a period of 30 to 50% of the extracted gait waveform as the terminal stance period.

12 . The estimation method according to claim 8 ,

wherein the presumption model is a machine learning model, and

wherein the estimation method further comprises:

performing classification of a degree of pronation/supination based on an output of the machine learning model, and

based on the classification indicating supination or pronation, outputting recommendation information that supports user decision-making to visit a hospital.

13 . A non-transitory program recording medium that stores a program configured to cause a computer to execute one or more processes comprising:

measuring, by an inertial measurement sensor including a three-axis accelerometer and a three-axis gyroscope, spatial acceleration and spatial angular velocity and outputting time-series data based on the spatial acceleration and the spatial angular velocity;

detecting a terminal stance period from the time-series data based on a physical quantity related to a movement of a foot measured by the inertial measurement sensor provided at the foot portion, the detecting the terminal stance period from the time-series data comprising converting the sensor data from a local coordinate system to a world coordinate system and detecting the terminal stance period as a period from heel lift to opposite heel strike based on an inflection point of roll angular acceleration;

extracting a feature amount from an angular waveform in a coronal plane during the terminal stance period, the angular waveform being a pitch angle about a Y-axis in the coronal plane (ZX plane), and the feature amount comprising at least one of an average value, a weighted average value, and an integral value within the terminal stance period;

estimating a degree of pronation/supination of the foot by inputting the feature amount extracted from the angular waveform in the coronal plane to a presumption model, the presumption model generated by learning a data set in which the feature amount extracted from the angular waveform in the coronal plane is an explanatory variable and a center of pressure excursion index (CPE) obtained from a foot pressure distribution measured by a pressure sensor is an objective variable; and

outputting the estimation result regarding the degree of pronation/supination of the foot.

14 . The non-transitory program recording medium according to claim 13 , wherein the one or more processes further comprises:

outputting the estimation result indicating one of pronation/supination of the foot and a normal foot according to a value of the center of pressure excursion index.

15 . The non-transitory program recording medium according to claim 14 , wherein the one or more processes further comprises: outputting the estimation result indicating supination when the center of pressure excursion index value is equal to or more than 20, outputting the estimation result indicating normality when the center of pressure excursion index value is equal to or more than 9 and less than 20, and outputting the estimation result indicating pronation when the center of pressure excursion index value is less than 9.

16 . The non-transitory program recording medium according to claim 13 , wherein the one or more processes further comprises:

extracting a gait waveform for one gait cycle starting from heel strike, from the time-series data of the sensor data, and

detecting a period of 30 to 50% of the extracted gait waveform as the terminal stance period.

17 . The non-transitory program recording medium according to claim 8 ,

wherein the presumption model is a machine learning model, and

wherein the one or more processes further comprises:

performing classification of a degree of pronation/supination based on an output of the machine learning model, and

based on the classification indicating supination or pronation, outputting recommendation information that supports user decision-making to visit a hospital.