Walking aid and system and method of gait monitoring

Patent metadata

Field	Value
Publication number	US-10799154-B2
Application number	US-201615068317-A
Country	US
Kind code	B2
Filing date	Mar 11, 2016
Priority date	Mar 11, 2015
Publication date	Oct 13, 2020
Grant date	Oct 13, 2020

Abstract

A clinical assessment tool coupled to a walking aid for enhancing a therapist's observation-based gait assessment with use of additional objective and quantitative data such as acceleration, angular velocity, and applied forces. The assessment tool facilitates appropriate assistive gait device prescription, provides patients and therapists feedback during gait training, and reduces wrist and shoulder injuries with walking aid usage. The assessment tool is configured to detect timing and speed of walking aid, placement, angular acceleration of the walking aid, and amounts of weight borne on the walking aid.

First claim

What is claimed is:

1. A gait monitoring system for determining a state of action of a patient, the system comprising:
a walking aid including:
an elongated housing having a first end and a second end;
a handle coupled to the first end of the elongated housing;
a base coupled to the second end of the elongated housing distal from the handle, wherein the base is configured to bear weight applied to the handle; and
a power source coupled to the elongated housing;

a plurality of first sensors disposed on the handle, each of the first sensors configured to detect a first force applied to the handle;
a second sensor unit coupled to the handle and configured to detect an acceleration value and an angular velocity value at the handle;
a third sensor coupled to the base and configured to detect a second force applied to the base;
a fourth sensor unit coupled to the base and configured to detect an acceleration value and an angular velocity value at the base, wherein the second sensor unit and the fourth sensor unit are positioned at two different points on the walking aid such that different patterns of acceleration and velocity are detected at the two different points depending on a phase of walking; and
a microcontroller in electrical communication with the first sensors, the second sensor unit, the third sensor, and the fourth sensor unit, the microcontroller configured to electronically transmit the first forces from the first sensors, the acceleration value and the angular velocity value from the second sensor unit, the second force from the third sensor, and the acceleration value and the angular velocity value from the fourth sensor unit to a computing device for electronic processing and outputting of information related to a gait pattern of a patient, the computing device including
a memory configured to store instructions associated with a gait pattern prediction application;
one or more processors coupled to the memory, the one or more processors executing the gait pattern prediction application in conjunction with the instructions stored in the memory, wherein the one or more processors are configured to:
receive data from the plurality of first sensors, the second sensor unit, the third sensor, and the fourth sensor unit;
preprocess the data from the plurality of first sensors, the second sensor unit, the third sensor, and the fourth sensor unit to obtain derived data;
apply a sliding window technique to segment the derived data into fixed length intervals;
convert the data from the plurality of first sensors, the second sensor unit, the third sensor, and the fourth sensor unit into a frequency domain data set and a time domain data set using feature extraction techniques;
extract features of the time domain data set and from the frequency domain data set;
apply a feature selection technique to the extracted features to rank the extracted features;
determine a preferred data set from the extracted features in the time domain data set and the frequency domain data set that are representative of a state of action of the patient;
generate a model defining a set of states of action of the patient, the model based on a set of controlled walking aid data; and
predict the state of action of the patient by comparing the preferred data set of the patient's walking aid data to the model based on the controlled walking aid data; and
output the prediction on a display device.

2. The gait monitoring system of claim 1 , wherein the microcontroller is positioned with the elongated housing or the handle or the base.

3. The gait monitoring system of claim 1 , further comprising a housing coupled to the elongated housing or the base, the housing defining a first recess configured to receive the third sensor.

4. The gait monitoring system of claim 3 , wherein the third sensor is a load cell.

5. The gait monitoring system of claim 1 , further comprising a housing coupled to the elongated housing, the housing configured to support a power source, an analog-to-digital converter, and an accelerometer.

6. The gait monitoring system of claim 5 , wherein the housing includes a first recess configured to receive the power source, a second recess configured to receive the analog-to-digital converter, and a third recess configured to receive the accelerometer.

7. The gait monitoring system of claim 1 , further comprising an analog to digital converter (ADC) in electronic communication with the microcontroller, the ADC configured to receive the first force data from the first sensors.

8. The gait monitoring system of claim 1 , wherein at least eight of the first sensors are positioned on the handle of the walking aid.

9. The gait monitoring system of claim 1 , wherein the states of action are standing in place, walking, stair ascent, or stair descent.

10. A non-transitory computer readable medium carrying a computer program comprising computer readable instructions configured to cause an electronic processor to carry out a method of determining a state of action of a patient, the method comprising:
receiving, as input to the processor, data from a plurality of force sensors distributed on a walking cane and a plurality of inertial sensors distributed on the walking cane, wherein the walking cane is able to bear weight and the patient is a user of the walking cane;
preprocessing the data from the force sensors and the inertial sensors to obtain derived data;
apply a sliding window technique to segment the derived data into fixed length intervals;
converting, by the processor, the data from the force sensors and the inertial sensors into a frequency domain data set and a time domain data set using feature extraction techniques;
extracting features of the time domain data set and from the frequency domain data set;
applying a feature selection technique to the extracted features to rank the extracted features;
extracting selected features in the sliding window of the time domain data set and from the frequency domain data set;
determining, by the processor, a preferred data set from the extracted features in the time domain data set and the frequency domain data set that are representative of a state of action of the patient;
generating, by the processor, a model defining a set of states of action of the patient, the model based on a set of controlled walking cane data; and
predicting, by the processor, the state of action of the patient by comparing the preferred data set of the patient's walking cane data to the model based on the controlled walking cane data; and
outputting, by the processor, the prediction on a display device in a graphical user interface of a clinical tool for gait analysis, gait device prescription, or gait training.

11. The non-transitory computer readable medium of claim 10 , wherein the controlled walking cane data is data collected by a microcontroller coupled to the walking cane during known states of action of the patient.

12. The non-transitory computer readable medium of claim 10 , wherein the states of action are standing in place, walking, stair ascent, or stair descent.

Assignees

• Univ Vanderbilt

Inventors

• Sarkar Nilanjan
• Withrow Thomas J
• Wade Joshua W
• Boyles Robert
• Myszka Alec
• Bekele Esube T
• Beccani Marco

Classifications

• A61B5/6887

  mounted on external non-worn devices, e.g. non-medical devices · CPC title

• A45B3/00

  Sticks combined with other objects · CPC title

• A61B5/112 Primary

  Gait analysis · CPC title

• G01L5/0004

  Force transducers adapted for mounting in a bore of the force receiving structure (G01L5/0009 takes precedence) · CPC title

• G01L5/00

  Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes · CPC title