Patents by Inventor Patrick Kaifosh
Patrick Kaifosh has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20230218202Abstract: A system comprising a plurality of neuromuscular sensors, each of which is configured to record a time-series of neuromuscular signals from a surface of a user's body; and at least one computer hardware processor programmed to perform: applying a source separation technique to the time series of neuromuscular signals recorded by the plurality of neuromuscular sensors to obtain a plurality of neuromuscular source signals and corresponding mixing information; providing features, obtained from the plurality of neuromuscular source signals and/or the corresponding mixing information, as input to a trained statistical classifier and obtaining corresponding output; and identifying, based on the output of the trained statistical classifier, and for each of one or more of the plurality of neuromuscular source signals, an associated set of one or more biological structures.Type: ApplicationFiled: March 21, 2023Publication date: July 13, 2023Inventors: Patrick Kaifosh, Tudor Giurgica-Tiron, Timothy Machado, Thomas Reardon, Erik Schomburg
-
Patent number: 11635736Abstract: A system comprising a plurality of neuromuscular sensors, each of which is configured to record a time-series of neuromuscular signals from a surface of a user's body; and at least one computer hardware processor programmed to perform: applying a source separation technique to the time series of neuromuscular signals recorded by the plurality of neuromuscular sensors to obtain a plurality of neuromuscular source signals and corresponding mixing information; providing features, obtained from the plurality of neuromuscular source signals and/or the corresponding mixing information, as input to a trained statistical classifier and obtaining corresponding output; and identifying, based on the output of the trained statistical classifier, and for each of one or more of the plurality of neuromuscular source signals, an associated set of one or more biological structures.Type: GrantFiled: October 19, 2018Date of Patent: April 25, 2023Assignee: Meta Platforms Technologies, LLCInventors: Patrick Kaifosh, Tudor Giurgica-Tiron, Timothy Machado, Thomas Reardon, Erik Schomburg
-
Publication number: 20220269346Abstract: The disclosed method may include receiving neuromuscular activity data over a first time series from a first sensor on a wearable device donned by a user receiving ground truth data over a second time series from a second sensor that indicates a body part state of a body part of the user, generating one or more training datasets by time-shifting at least a portion of the neuromuscular activity data over the first time series relative to the second time series, to associate the neuromuscular activity data with at least a portion of the ground truth data, and training one or more inferential models based on the one or more training datasets. Various other related methods and systems are also disclosed.Type: ApplicationFiled: May 10, 2022Publication date: August 25, 2022Inventors: Nadine Hussami, Patrick Kaifosh, Alexandre Barachant, Daniel Wetmore
-
Patent number: 11361522Abstract: Methods and apparatus for enabling a user or third-party to select or adjust parameters of one or more statistical models used to generate a musculoskeletal representation. The method comprises providing as input to the statistical model(s), a plurality of neuromuscular signals recorded by a plurality of neuromuscular sensors during performance of at least one gesture by a user, wherein the at least one gesture is performed by the user while wearing a wearable device having the plurality of neuromuscular sensors arranged thereon, rendering at least one visual representation based on an output of the statistical model(s), and receiving user or third-party input to adjust parameters of the statistical model(s) based on the rendered at least one visual representation, the user input including a selection of a particular statistical model of the statistical model(s) and/or an adjustment of parameters associated with the particular statistical model.Type: GrantFiled: October 18, 2019Date of Patent: June 14, 2022Assignee: Facebook Technologies, LLCInventors: Patrick Kaifosh, Adam Berenzweig, Adam Al-natsheh
-
Patent number: 11337652Abstract: A method for determining spatial information for a multi-segment articulated rigid body system having at least an anchored segment and a non-anchored segment coupled to the anchored segment, each segment in the multi-segment articulated rigid body system representing a respective body part of a user, the method comprising: obtaining signals recorded by a first autonomous movement sensor coupled to a body part of the user represented by the non-anchored segment; providing the obtained signals as input to a trained statistical model and obtaining corresponding output of the trained statistical model; and determining, based on the corresponding output of the trained statistical model, spatial information for at least the non-anchored segment of the multi-segment articulated rigid body system.Type: GrantFiled: July 25, 2017Date of Patent: May 24, 2022Assignee: Facebook Technologies, LLCInventors: Patrick Kaifosh, Timothy Machado, Thomas Reardon, Erik Schomburg, Calvin Tong
-
Patent number: 11331045Abstract: The disclosed systems and methods are generally directed to interpreting neuromuscular signals. The system includes (A) a plurality of neuromuscular sensors that detect a plurality of neuromuscular signals from a user and (B) at least one multiplexer in communication with the plurality of neuromuscular sensors and that is capable of dynamically adjusting neuromuscular sensor processing based on neuromuscular signal characteristics. A computer processor is programmed to (i) receive a set of neuromuscular signals from the plurality of neuromuscular sensors, (ii) determine, via a real-time system, at least one signal characteristic included in a neuromuscular signal, where the neuromuscular signal is associated with a first neuromuscular sensor included in the plurality of neuromuscular sensors; and (iii) dynamically reconfigure the processing of neuromuscular signals from the plurality of neuromuscular sensors based on an output from the multiplexer.Type: GrantFiled: April 29, 2020Date of Patent: May 17, 2022Assignee: Facebook Technologies, LLCInventors: Anthony D. Moschella, Daniel Wetmore, Patrick Kaifosh, Adam Al-natsheh, Tudor Giurgica-Tiron, Qiushi Mao
-
Patent number: 11327566Abstract: The disclosed method may include receiving neuromuscular activity data over a first time series from a first sensor on a wearable device donned by a user receiving ground truth data over a second time series from a second sensor that indicates a body part state of a body part of the user, generating one or more training datasets by time-shifting at least a portion of the neuromuscular activity data over the first time series relative to the second time series, to associate the neuromuscular activity data with at least a portion of the ground truth data, and training one or more inferential models based on the one or more training datasets. Various other related methods and systems are also disclosed.Type: GrantFiled: March 27, 2020Date of Patent: May 10, 2022Assignee: Facebook Technologies, LLCInventors: Nadine Hussami, Patrick Kaifosh, Alexandre Barachant, Daniel Wetmore
-
Publication number: 20220019284Abstract: Computerized systems, methods, and computer-readable storage media storing code for the methods enable feedback to be provided to a user based on neuromuscular signals sensed from the user. One such system includes neuromuscular sensors and at least one computer processor. The sensors, which are arranged on one or more wearable devices, are configured to sense neuromuscular signals from the user. The at least one computer processor is or are programmed to process the neuromuscular signals using one or more inference models, and to provide feedback to the user based on one or both of: the processed neuromuscular signals and information derived from the processed neuromuscular signals. The feedback includes visual feedback of information relating to one or both of: a timing of an activation of at least one motor unit of the user and an intensity of the activation of the at least one motor unit of the user.Type: ApplicationFiled: November 15, 2019Publication date: January 20, 2022Inventors: Patrick KAIFOSH, Alexandre BARACHANT, Mason REMALEY, Julien KILlAN, Kirak HONG, Nathan DANIELSON, Qiushi MAO, Daniel WETMORE
-
Patent number: 11216069Abstract: Systems and methods for using neuromuscular information to improve speech recognition. The system includes a plurality of neuromuscular sensors arranged on one or more wearable devices and configured to continuously record a plurality of neuromuscular signals from a user, at least one storage device configured to store one or more trained statistical models for determining text based on audio input and the plurality of neuromuscular signals, at least one input interface configured to receive the audio input, and at least one computer processor programmed to obtain the audio input and the plurality of neuromuscular signals, provide as input to the one or more trained statistical models, the audio input and the plurality of neuromuscular signals or signals derived from the plurality of neuromuscular signals, and determine based, at least in part, on an output of the one or more trained statistical models, the text.Type: GrantFiled: May 8, 2018Date of Patent: January 4, 2022Assignee: Facebook Technologies, LLCInventors: Adam Berenzweig, Patrick Kaifosh, Alan Huan Du, Jeffrey Scott Seely
-
Publication number: 20210405750Abstract: Methods and apparatus for providing a dynamically-updated computerized musculo-skeletal representation comprising a plurality of rigid body segments connected by joints. The method comprises recording, using a plurality of autonomous sensors arranged on one or more wearable devices, a plurality of autonomous signals from a user, wherein the plurality of autonomous sensors include a plurality of neuromuscular sensors configured to record neuromuscular signals. The method further comprises providing as input to a trained statistical model, the plurality of neuromuscular signals and/or information based on the plurality of neuromuscular signals.Type: ApplicationFiled: April 12, 2021Publication date: December 30, 2021Inventors: Patrick Kaifosh, Timothy Machado, Thomas Reardon, Erik Schomburg
-
Patent number: 11179066Abstract: Methods and apparatus for substantially real-time detection of spike events in neuromuscular data. The method comprises receiving a plurality of neuromuscular signals from a plurality of neuromuscular sensors arranged on one or more wearable devices worn by a user, detecting, based on the plurality of neuromuscular signals or information derived from the plurality of neuromuscular signals, at least one spike event corresponding to firing of an action potential in at least one motor unit, determining, based on the plurality of neuromuscular signals or the information derived from the plurality of neuromuscular signals, a biological source of the detected at least one spike event, and generating at least one output based, at least in part, on the detected at least one spike event and/or the determined biological source of the detected at least one spike event.Type: GrantFiled: August 13, 2019Date of Patent: November 23, 2021Assignee: Facebook Technologies, LLCInventors: Patrick Kaifosh, Alexandre Barachant, Michael Isaac Mandel, Daniel Wetmore
-
Patent number: 11163361Abstract: Methods and apparatus for calibrating performance of one or more statistical models used to generate a musculoskeletal representation. The method comprises controlling presentation of instructions via a user interface to instruct the user to perform the at least one gesture and updating at least one parameter of the one or more statistical models based, at least in part on a plurality of neuromuscular signals recorded by a plurality of neuromuscular sensors during performance of the at least one gesture by the user.Type: GrantFiled: October 31, 2019Date of Patent: November 2, 2021Assignee: Facebook Technologies, LLCInventors: Patrick Kaifosh, Tudor Giurgica-Tiron, Adam Berenzweig, Steven Kober, Adam Al-natsheh, Alexandre Barachant, Zhuo Wang
-
Patent number: 11036302Abstract: Systems and methods for using neuromuscular information to improve speech recognition. The system includes a plurality of neuromuscular sensors, arranged on one or more wearable devices, wherein the plurality of neuromuscular sensors is configured to continuously record a plurality of neuromuscular signals from a user, at least one storage device configured to store one or more trained statistical models, and at least one computer processor programmed to provide as an input to the one or more trained statistical models, the plurality of neuromuscular signals or signals derived from the plurality of neuromuscular signals, determine based, at least in part, on an output of the one or more trained statistical models, at least one instruction for modifying an operation of a speech recognizer, and provide the at least one instruction to the speech recognizer.Type: GrantFiled: February 10, 2020Date of Patent: June 15, 2021Assignee: Facebook Technologies, LLCInventors: Adam Berenzweig, Patrick Kaifosh, Alan Huan Du, Jeffrey Scott Seely
-
Patent number: 11000211Abstract: Methods and apparatus for adapting a control mapping associating sensor signals with control signals for controlling an operation of a device. The method comprises obtaining first state information for an operation of the device, providing the first state information as input to an intention model associated with an operation of the device and obtaining corresponding first intention model output, providing a plurality of neuromuscular signals recorded from a user and/or signals derived from the neuromuscular signals as inputs to a first control mapping and obtaining corresponding first control mapping output, and updating the first control mapping using the inputs provided to the first control mapping and the first intention model output to obtain a second control mapping.Type: GrantFiled: July 25, 2017Date of Patent: May 11, 2021Assignee: Facebook Technologies, LLCInventors: Patrick Kaifosh, Timothy Machado, Thomas Reardon, Erik Schomburg, Joshua Merel, Steven Demers
-
Patent number: 10990174Abstract: Methods and apparatus for providing a dynamically-updated computerized musculo-skeletal representation comprising a plurality of rigid body segments connected by joints. The method comprises recording, using a plurality of autonomous sensors arranged on one or more wearable devices, a plurality of autonomous signals from a user, wherein the plurality of autonomous sensors include a plurality of neuromuscular sensors configured to record neuromuscular signals. The method further comprises providing as input to a trained statistical model, the plurality of neuromuscular signals and/or information based on the plurality of neuromuscular signals.Type: GrantFiled: July 25, 2017Date of Patent: April 27, 2021Assignee: Facebook Technologies, LLCInventors: Patrick Kaifosh, Timothy Machado, Thomas Reardon, Erik Schomburg
-
Patent number: 10950047Abstract: Methods and apparatus for anonymizing neuromuscular signals used to generate a musculoskeletal representation. The method comprises recording, using a plurality of neuromuscular sensors arranged on one or more wearable devices, a plurality of neuromuscular signals from a user, providing as input to a trained statistical model, the plurality of neuromuscular signals and/or information based on the plurality of neuromuscular signals; and generating, the musculoskeletal representation based, at least in part, on an output of the trained statistical model, wherein the musculoskeletal representation is an anonymized musculoskeletal representation from which at least one personal characteristic of the user has been removed.Type: GrantFiled: October 16, 2019Date of Patent: March 16, 2021Assignee: Facebook Technologies, LLCInventors: Patrick Kaifosh, Tudor Giurgica-Tiron, Steven Kober, Adam Al-natsheh, Alexandre Barachant
-
Patent number: 10921764Abstract: Methods and apparatus for controlling a physical object in an environment based, at least in part, on neuromuscular signals. The method comprises recording a plurality of neuromuscular signals from a plurality of neuromuscular sensors arranged on one or more wearable devices worn by a user, receiving a selection of a physical object within the environment, and controlling, based at least in part on the plurality of neuromuscular signals and/or information based on the plurality of neuromuscular signals, an operation of the selected object within the environment.Type: GrantFiled: September 26, 2019Date of Patent: February 16, 2021Assignee: Facebook Technologies, LLCInventors: Qiushi Mao, Jasmine Stone, Adam Berenzweig, Patrick Kaifosh, Robert John DiMaiolo, Jason Reisman, Robert Cochran, Naor Brown, Nitzan Bartov, Joshua Duyan, Daniel Wetmore
-
Patent number: 10905350Abstract: Computerized systems, methods, and computer-readable storage media storing code for implementing the methods are described for providing dynamically-updated musculoskeletal information. One such system includes a processor is programmed to: provide, as an input to a trained inference model, information based on a plurality of neuromuscular signals from a user and information based on at least one image of the user; determine, based on an output of the trained inference model, position information describing a spatial relationship between two or more connected musculoskeletal segments of the user and/or force information describing a force exerted by at least one musculoskeletal segment of the user; and output the position information and/or the force information.Type: GrantFiled: August 30, 2019Date of Patent: February 2, 2021Assignee: Facebook Technologies, LLCInventors: Adam Berenzweig, Thomas Reardon, Christopher Osborn, Patrick Kaifosh, Brett Jurman, Daniel Wetmore
-
Patent number: 10842407Abstract: Computerized systems, methods, and computer-readable storage media storing code for implementing the methods are provided for training an inference model used to generate a musculoskeletal representation. One such system includes a processor programmed to: determine, based on information obtained from at least one image, position information describing a spatial relationship between two or more connected musculoskeletal segments of a user; determine, based on a plurality of neuromuscular signals, force information; associate the position information with the force information; train an inference model to output a musculoskeletal representation consistent with the position information and/or the force information when neuromuscular input signals provided to the inference model have at least one predetermined characteristic, to produce an updated inference model; and cause the updated inference model to be stored in a memory.Type: GrantFiled: August 30, 2019Date of Patent: November 24, 2020Assignee: Facebook Technologies, LLCInventors: Adam Berenzweig, Thomas Reardon, Christopher Osborn, Patrick Kaifosh, Brett Jurman, Daniel Wetmore
-
Patent number: 10817795Abstract: Methods and systems for dynamically reconstructing handstate information based on multiple inputs are described. The methods and systems use data from multiple inputs including a plurality of neuromuscular sensors arranged on one or more wearable devices and one or more cameras. The multimodal data is provided as input to a trained statistical model. The methods and systems determine, based on the data from the multiple inputs, an estimate and representation of the spatial relationship between two or more connected segments of the musculoskeletal representation and force information describing a force exerted by at least one segment of the musculoskeletal representation. The methods and systems further update the computerized musculoskeletal representation based, at least in part, on the position information and the force information.Type: GrantFiled: January 25, 2019Date of Patent: October 27, 2020Assignee: Facebook Technologies, LLCInventors: Patrick Kaifosh, Tudor Giurgica-Tiron, Brett Jurman