Head-mounted devices for postural alignment correction

Patent metadata

Field	Value
Publication number	US-2024215866-A1
Application number	US-202218147307-A
Country	US
Kind code	A1
Filing date	Dec 28, 2022
Priority date	Dec 28, 2022
Publication date	Jul 4, 2024
Grant date	—

Abstract

Examples of head-mounted devices for postural alignment correction are provided. In one aspect, a head-mounted device is implemented to include a display, a plurality of sensors including an inertial measurement unit, and a controller. The controller includes instructions executable to control the head-mounted device to receive inertial measurement data from the inertial measurement unit, to input the received inertial measurements into a machine learning model, and to receive an estimated body posture from the neural machine learning model. In this aspect, additionally or alternatively, the machine learning model is an artificial neural network that has been trained to estimate the body posture by calculating a craniovertebral angle, where the craniovertebral angle is defined as an angle between a first line crossing a tragus point and a cervical vertebra point and a second line crossing the cervical vertebra point.

First claim

1 . A head-mounted device for postural alignment correction, the head-mounted device comprising:
a display; a plurality of sensors comprising an inertial measurement unit; and a controller comprising instructions executable to control the head-mounted device to:
receive inertial measurement data from the inertial measurement unit;
input the received inertial measurements into a machine learning model; and
receive an estimated body posture from the machine learning model.

2 . The head-mounted device of claim 1 , wherein the body posture comprises a head posture, and wherein the machine learning model is an artificial neural network that has been trained to estimate the head posture by calculating a craniovertebral angle.

3 . The head-mounted display of claim 1 , wherein the plurality of sensors further comprises one or more cameras, and wherein the artificial neural network has been trained to estimate the body posture using image data from the one or more cameras.

4 . The head-mounted device of claim 1 , wherein the neural network has been trained based on an average human population, and wherein estimating the body posture comprises:
computing a loss value using a loss function; and adjusting the trained neural network based on the computed loss value.

5 . The head-mounted device of claim 4 , wherein computing the loss value using the loss function comprises a supervised learning process that includes receiving an input from a user indicating accuracy of the estimated body posture.

6 . The head-mounted device of claim 1 , wherein the controller further comprises instructions executable to output information to a user using the display, wherein the information indicates the estimated body posture.

7 . The head-mounted device of claim 1 , wherein the presented information includes recommendations on corrective posture actions.

8 . A head-mounted device for postural alignment correction, the head-mounted device comprising:
a display; a plurality of sensors comprising a camera; and a controller comprising instructions executable to control the head-mounted device to:
receive image data from the camera;
input the received image data into an artificial neural network; and
receive an estimated body posture from the artificial neural network.

9 . The head-mounted device of claim 8 , wherein the plurality of sensors comprises one or more of a downward-facing camera and a forward-facing camera, and wherein the image data comprises stereoscopic image data.

10 . The head-mounted device of claim 8 , wherein:
the estimated body posture comprises an estimated head posture; the plurality of sensors further comprises an inertial measurement unit, and the neural network has been trained to estimate the body posture by calculating a craniovertebral angle using inertial measurement data from the inertial measurement unit and image data from the camera.

11 . The head-mounted device of claim 8 , wherein the artificial neural network has been trained to estimate the body posture using the image data based on a portion of a body of a user that is in view of the camera.

12 . The head-mounted device of claim 8 , wherein the artificial neural network has been trained to estimate a reclining position of a user using the image data from the camera.

13 . The head-mounted device of claim 8 , wherein the artificial neural network has been trained based on an average human population, and wherein estimating the body posture comprises:
computing a loss value using a loss function; and adjusting the trained neural network based on the calculated loss.

14 . The head-mounted device of claim 13 , wherein computing the loss value using the loss function comprises a supervised learning process that includes receiving an input from a user indicating accuracy of the estimated body posture.

15 . The head-mounted device of claim 8 , wherein the controller further comprises instructions executable to present information to a user using the display, wherein the information indicates the estimated body posture.

16 . The head-mounted device of claim 8 , wherein the presented information includes recommendations on corrective posture actions.

17 . On a head-mounted computing device, a method for postural alignment correction, the method comprising:
receiving inertial measurement data from an inertial measurement unit; inputting the received inertial measurements into an artificial neural network; receiving an estimated body posture from the artificial neural network; outputting the estimated body posture to a user interface; receiving input from a user indicating accuracy of the estimated body posture; computing a loss value using a loss function based on the estimated body posture and the received input from the user; and adjusting the artificial neural network based on the computed loss value.

18 . The method of claim 17 , wherein the estimated body posture comprises an estimated head posture, and wherein the artificial neural network has been trained to estimate the head posture by calculating a craniovertebral angle.

19 . The method of claim 17 , further comprising:
receiving image data from a camera comprising one or more of a downward-facing camera and a forward-facing camera; and inputting the image data into the artificial neural network.

20 . The method of claim 17 , further comprising:
outputting recommendations on corrective posture actions to the user interface.

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Demaster-Smith Rayna
• Ziat Mehdi

Classifications

• A61B5/742

  using visual displays (displays for heart-related electrical signals, e.g. ECG, A61B5/339) · CPC title

• A61B5/6803

  Head-worn items, e.g. helmets, masks, headphones or goggles · CPC title

• A61B5/1121

  Determining geometric values, e.g. centre of rotation or angular range of movement · CPC title

• G06V10/82

  using neural networks · CPC title

• G06V40/10

  Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands · CPC title