US 12,350,108 B2
Wearable imaging system for measuring bone displacement
Manjirnath Agrahayan Chatterjee, Alameda, CA (US); Faisal Mohammed Mirza, Aptos, CA (US); and Rajveen Rosie Sendher, Santa Cruz, CA (US)
Assigned to Kinisi Inc., Santa Cruz, CA (US)
Filed by Kinisi Inc, Santa Cruz, CA (US)
Filed on Jun. 15, 2022, as Appl. No. 17/841,489.
Claims priority of provisional application 63/211,378, filed on Jun. 16, 2021.
Prior Publication US 2022/0401079 A1, Dec. 22, 2022
Int. Cl. A61B 8/00 (2006.01); A61B 8/08 (2006.01); G16H 50/20 (2018.01)
CPC A61B 8/5223 (2013.01) [A61B 8/0875 (2013.01); A61B 8/4477 (2013.01); G16H 50/20 (2018.01); G06T 2207/10132 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A method comprising:
transmitting a digital signal to a distributed set of sensors positioned around a joint of a user, the digital signal comprising a first set of instructions for one or more sensors of the distributed set of sensors to emit ultrasound signals towards the joint of the user and a second set of instructions for sensors of the distributed set of sensors to detect ultrasound signals traveling through structures of the joint and echoes of ultrasound signals reflected off structures of the joint;
extracting physiological properties of bones connected at the joint and tissue surrounding the joint from features of the detected echoes of ultrasound signals, wherein the extracted physiological properties comprise one or more of: elasticity, density, strength, and vascularity;
generating a bone displacement measurement of the bones connected at the joint, wherein generating the bone displacement measurement comprises:
inputting the physiological properties of the joint and properties of the detected echoes of ultrasound signals to a machine-learning displacement model trained to generate a virtual representation of a space surrounding the joint, wherein the displacement model is trained using a training dataset of historical ultrasound signals collected from a population of users, each entry of the training dataset comprising historical ultrasound signals detected for a joint and labeled with an identifier of the joint and a known bone displacement measurement at the labeled joint; and
determining the bone displacement measurement based on a distance between a sensor of the distributed set of sensors emitting ultrasound signals and a sensor of the distributed set of sensors detecting echoes of ultrasound signals, wherein the bone displacement measurement describes a movement of bones connected at the joint; and
responsive to determining that the bone displacement measurement at the joint exceeds a threshold displacement, transmitting the bone displacement measurement to a medical provider of the user.
 
9. A wearable imaging system comprising:
a distributed set of sensors secured around a joint of a user, the distributed set of sensors comprising:
one or more sensors configured to emit ultrasound signals towards the joint of the user; and
sensors configured to detect ultrasound signals traveling through structures of the joint and echoes of ultrasound signals reflected off structures of the joint; and
a controller configured to:
transition each sensor of the distributed set of sensors between a first mode pertaining the one or more sensors emitting ultrasound signals and a second mode pertaining to sensors detecting ultrasound signals traveling through structures of the joint;
extract physiological properties of bones connected at the joint and tissue surrounding the joint from features of the detected echoes of ultrasound signals, wherein the extracted physiological properties comprise one or more of: elasticity, density, strength, and vascularity;
generating a bone displacement measurement of the bones connected at the joint, wherein the controller generates the bone displacement measurement by:
inputting the physiological properties of the joint and properties of the detected echoes of ultrasound signals to a machine-learning displacement model trained to generate a virtual representation of a space surrounding the joint, wherein the displacement model is trained using a training dataset of historical ultrasound signals collected from a population of users, each entry of the training dataset comprising historical ultrasound signals detected for a joint and labeled with an identifier of the joint and a known bone displacement measurement at the labeled joint; and
determining the bone displacement measurement based on a distance between a sensor of the distributed set of sensors emitting ultrasound signals and a sensor of the distributed set of sensors detecting echoes of ultrasound signals, wherein the bone displacement measurement describes a movement of bones connected at the joint; and
responsive to determining that the bone displacement measurement exceeds a threshold displacement, transmit the bone displacement measurement to a medical provider of the user.
 
16. A non-transitory computer readable storage medium, storing instructions encoded thereon that, when executed by one or more processors, cause the processors to:
transmit a digital signal to a distributed set of sensors positioned around a joint of a user, the digital signal comprising a first set of instructions for one or more sensors of the distributed set of sensors to emit ultrasound signals towards the joint of the user and a second set of instructions for sensors of the distributed set of sensors to detect ultrasound signals traveling through structures of the joint and echoes of ultrasound signals reflected off structures of the joint;
extract physiological properties of bones connected at the joint and tissue surrounding the joint from features of the detected echoes of ultrasound signals, wherein the extracted physiological properties comprise one or more of: elasticity, density, strength, and vascularity;
generate a bone displacement measurement of the bones connected at the joint, wherein instructions for generating the bone displacement measurement cause the processor to:
input the physiological properties of the joint and properties of the detected echoes of ultrasound signals to a machine-learning displacement model trained to generate a virtual representation of a space surrounding the joint, wherein the displacement model is trained using a training dataset of historical ultrasound signals collected from a population of users, each entry of the training dataset comprising historical ultrasound signals detected for a joint and labeled with an identifier of the joint and a known bone displacement measurement at the labeled joint; and
determine the bone displacement measurement based on a distance between a sensor of the distributed set of sensors emitting ultrasound signals and a sensor of the distributed set of sensors detecting echoes of ultrasound signals, wherein the bone displacement measurement describes a movement of bones connected at the joint; and
responsive to determining that the bone displacement measurement at the joint exceeds a threshold displacement, transmit the bone displacement measurement to a medical provider of the user.