Granted Patent B2

US 12,629,516 · App. 18/046,773 · Granted May 19, 2026

Inventors: Pierluigi Alfredo Mantovani (Berkeley, CA); Juan Manuel Rodriguez (Berkeley, CA); Petr Karashchuk (Seattle, WA); Andrew Ekelem (Fairfield, CA); Mohammed Aashyk Mohaiteen Hebsur Rahman (San Francisco, CA)

Assignee: Evolution Devices, Inc.

A61N1/36003A61B5/0022A61B5/112A61B5/6828A61N1/0452A61N1/0456A61N1/0476A61N1/3603G06F3/011G06F3/017G06N3/08G06N20/00A61B2562/0219

Abstract

Claims (32)

1 . A wearable device for tracking or modulating a movement of a limb of a user, the device comprising:

one or more wearable articles configured to be worn on a limb of a user;

one or more electrode arrays configured to be in physical contact with the limb of the user; and

a control unit comprising a wireless communication module, one or more processors, one or more memory units, a portable power supply, an electrical muscle stimulation (EMS) generator in electrical communication with the one or more electrode arrays, and an inertial measurement unit (IMU), wherein a housing of the control unit is coupled to at least one of the one or more wearable articles, wherein the IMU comprises at least one of a gyroscope and an accelerometer,

wherein the one or more processors are programmed to execute instructions stored in the one or more memory units to:

retrieve gyroscope readings and accelerometer readings from the IMU, map the gyroscope readings to two periodic functions using a machine learning algorithm;

calculate a phase angle from the two periodic functions;

convert the phase angle to the gait cycle percentage; and

instruct the EMS generator to provide electrical stimulation to nerves and muscles of the limb via the one or more electrode arrays based in part on the gait cycle percentage calculated.

2 . The device of claim 1 , wherein the one or more wearable articles comprises a wearable article configured to be worn on a lower leg of the user below a knee and above an ankle of the user, and wherein the control unit is configured to be positioned on an anterior side of the lower leg of the user below the knee and above the ankle of the user.

3 . The device of claim 1 , wherein the one or more wearable articles comprises a first wearable article and a second wearable article, wherein the first wearable article is configured to be worn on an upper leg of the user and the second wearable article is configured to be worn on a lower leg of the user below a knee and above an ankle of the user.

4 . The device of claim 3 , wherein the one or more electrode arrays comprises at least a first upper leg array, a second upper leg array, a first lower leg array, and a second lower leg array, wherein the first upper leg array and the second upper leg array and wherein the first upper leg array and the second upper leg array are configured to be in physical contact with a skin surface in proximity to a hamstring or quadricep of the user, and wherein the first lower leg array and the second lower leg array are configured to be in physical contact with the skin surface in proximity to a tibialis anterior of the user.

5 . The device of claim 4 , wherein the one or more processors are programmed to execute additional instructions stored in the one or more memory units to instruct the EMS generator to generate a plurality of asymmetrical biphasic square pulses for transmission to electrodes of the one or more electrode arrays to provide the electrical stimulation to the nerves and muscles of the limb of the user.

6 . The device of claim 1 , wherein the machine learning algorithm comprises one or more multilayer perceptron networks.

7 . The device of claim 1 , wherein the one or more processors are programmed to execute further instructions stored in the one or more memory units to map at least one of the gyroscope readings and the accelerometer readings to three-dimensional angles of at least one of a hip, a knee, and a foot of the user throughout a gait cycle and determine at least one of a foot strike pattern, a foot inclination angle at initial contact, a tibia angle at loading response, a hip extension during late stance, a trunk lean, a heel eversion, a foot progression angle, a pelvic drop, a knee flexion during stance, a stride length, a knee window, a vertical displacement of the center mass, and a heel whip of the user based in part on the gait cycle percentage calculated, the gyroscope readings, the accelerometer readings, and the mapped three-dimensional angles.

8 . The device of claim 1 , wherein the machine learning algorithm is trained by correlating gyroscope readings with three-dimensional kinematic data of a model subject in motion.

9 . A method of modulating a movement of a limb of a user, comprising:

retrieving, using one or more processors, at least one of gyroscope readings from a gyroscope of an inertial measurement unit (IMU) and accelerometer readings from an accelerometer of the IMU, wherein the one or more processors and the IMU are part of a control unit further comprising a wireless communication module, one or more memory units, an electrical muscle stimulation (EMS) generator, and a portable power supply, and wherein a housing of the control unit is coupled to at least one of one or more wearable articles configured to be worn on the limb of the user;

mapping, using the one or more processors, the gyroscope readings to two periodic functions using a machine learning algorithm;

calculating a phase angle from the two periodic functions;

converting the phase angle to a gait cycle percentage;

instructing the EMS generator to provide electrical stimulation to nerves and muscles of the limb via one or more electrode arrays in electrical communication with the EMS generator based in part on the gait cycle percentage calculated; and

stimulating the nerves and muscles of the limb via the one or more electrode arrays in physical contact with the limb.

10 . The method of claim 9 , wherein the one or more wearable articles comprises a wearable article configured to be worn on a lower leg of the user below a knee and above an ankle of the user, and wherein the control unit is positioned on an anterior side of the lower leg of the user below the knee and above the ankle of the user.

11 . The method of claim 9 , wherein the one or more wearable articles comprises a first wearable article and a second wearable article, wherein the first wearable article is configured to be worn on an upper leg of the user and the second wearable article is configured to be worn on a lower leg of the user below a knee and above an ankle of the user.

12 . The method of claim 11 , wherein the one or more electrode arrays comprises at least a first upper leg array, a second upper leg array, a first lower leg array, and a second lower leg array, wherein the first upper leg array and the second upper leg array are configured to be in physical contact with a skin surface in proximity to a hamstring or quadricep of the user, and wherein the first lower leg array and the second lower leg array are configured to be in physical contact with the skin surface in proximity to a tibialis anterior of the user.

13 . The method of claim 12 , further comprising instructing the EMS generator, via the control unit, to generate a plurality of asymmetrical biphasic square pulses for transmission to electrodes of the one or more electrode arrays to provide the electrical stimulation to the nerves and muscles of the limb of the user.

14 . The method of claim 9 , wherein the machine learning algorithm comprises one or more multilayer perceptron networks.

15 . The method of claim 9 , further comprising:

mapping at least one of the gyroscope readings and the accelerometer readings to three-dimensional angles of at least one of a hip, a knee, and a foot of the user throughout a gait cycle; and

determining at least one of a foot strike pattern, a foot inclination angle at initial contact, a tibia angle at loading response, a hip extension during late stance, a trunk lean, a heel eversion, a foot progression angle, a pelvic drop, a knee flexion during stance, a stride length, a knee window, a vertical displacement of the center mass, and a heel whip of the user based in part on the gait cycle percentage calculated, the gyroscope readings, the accelerometer readings, and the mapped three-dimensional angles.

16 . The method of claim 9 , wherein the machine learning algorithm is trained by correlating gyroscope readings with three-dimensional kinematic data of a model subject in motion.

Assignments (1)

Continuity (3)

References Cited (17)