Optum, Inc.

Unitedhealth Group Incorporated

Jon Kevin Muse

Gregory J Boss

Marilyn L Gordon

Rama S Ravindranathan

Komal Khatri

Ramprasad Anandam Gaddam

Guerino Bonetti

Kristine Xu

Rick A Hamilton, Ii

Mark Gregory Megerian

Vijay S Nori

Paul J Godden

Yash Sharma

Charles P Schaller

Hadi D Halim

Martijn P Van Overbeek

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing optimized breathing therapy. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform optimized breathing therapy using at least one of interruption score generation machine learning models, observed inspiration-expiration pattern, expected inspiration-expiration patterns, expected musical patterns, and inferred musical patterns.

Machine learning techniques for optimized breathing therapy

A controller for an artificial heart enables activity-specific adjustments to the operation of an artificial heart by obtaining sensor data from a plurality of sensors monitoring characteristics of a patient's body, and using the sensor data as input to one or more control parameter models for identifying control parameters to be provided to the artificial heart to adjust the operational parameters of the artificial heart. The controller is in wireless communication with the artificial heart via an application program interface (API)-based communication channel that facilitates communication between the controller and the artificial heart. Moreover, a cloud-based management computing entity may be utilized to train and/or execute one or more models to enable real-time updates to the operational characteristics of the artificial heart to enable the artificial heart to appropriately accommodate activities of the patient.

Artificial heart control systems and methods

Various embodiments of the present disclosure provide methods, apparatuses, systems, computing devices, computing entities, and/or the like for detecting and preventing seizure events in real-time. For example, various embodiments provide techniques for detecting and preventing seizure events in real-time that use seizure prediction machine learning models.

Seizure prediction machine learning models

Various embodiments of the present disclosure provide methods, apparatuses, systems, computing devices, computing entities, and/or the like for monitoring a user's movement in real-time and providing or augmenting stimulation. For example, various embodiments provide techniques generating movement prediction profiles using movement prediction machine learning models and for use in conjunction with wearable devices.

Movement prediction machine learning models

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing optimized equipment allocation. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform optimized equipment allocation using at least one of equipment evaluation scoring machine learning models, equipment optimized allocation machine learning models, and equipment history events.

Machine learning techniques for optimized equipment allocation

A computing device may receive, from a remote computing device, a request to establish a communication session with the remote computing device and contextual information indicative of an urgency level of the communication session. The computing device may output, at one or more output devices, the contextual information indicative of the urgency level of the communication session and at least one of: haptic output or an audible alert that indicates the request to establish the communication session.

Providing health urgency context for incoming calls

Systems and methods are configured to perform machine-learning-based predictive behavioral response. In various embodiments, one or more behavioral monitoring data objects are identified and processed using a behavioral pattern prediction machine learning model to generate a behavioral pattern prediction model. The behavioral pattern prediction model is processed using a risk generation machine learning model to generate a risk model, wherein: (i) the risk generation machine learning model is generated based at least in part by one or more risk factors, and (ii) the risk model comprises a per-risk factor score for each risk factor of the one or more risk factors. The risk model is processed using an adjustment generation machine learning model to generate an adjustment model and one or more prediction-based actions are performed based on the adjustment model.

Machine-learning-based predictive behavioral monitoring

An example method includes obtaining, by processing circuitry, an image depicting a face of a person wearing eyewear and a reflection pattern generated by projecting light onto a lens of the eyewear and determining, by the processing circuitry, an optical parameter of the lens based on the image. The method further includes receiving, by the processing circuitry and from an electronic database, a personal characteristic data relating to a previously identified person and comparing, by the processing circuitry, the optical parameter of the lens to the personal characteristic data. The method further includes determining, by the processing circuitry and based on the comparison, a score indicative of a match between the person wearing eyewear and the previously identified person.

Authentication and facial recognition through analysis of optometric prescription data

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing splinting activity detection. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform splinting activity detection using at least one of splinting activity detection machine learning models, observed inspiration-expiration waveform pattern, and expected inspiration-expiration waveform patterns.

Machine learning techniques for detecting splinting activity

Embodiments of the invention provide apparatuses, systems, and methods for more accurate remote monitoring of a user's body. In some embodiments, a system for monitoring a user's body comprises a wearable device, a video sensor attached at a collar portion of the wearable device, a plurality of audio sensors spaced and attached at a body portion of a wearable device and a controller configured to determine a Jugular Venous Pressure (JVP) of the user, and determine audio characteristics of an output of the plurality of audio sensors to generate an audio heat map corresponding to at least one internal organ of the user.

Jugular venous pressure (JVP) measurement

A computing system includes a storage device and processing circuitry. The processing circuitry is configured to obtain an image frame that comprises a plurality of pixels that form a pixel array. Additionally, the processing circuitry is configured to determine that a region of the image frame belongs to a trigger content type. Based on determining that the region of the image frame belongs to the trigger content type, the processing circuitry is configured to modify the region of the image frame to adjust a luminance of pixels of the region of the image frame based on part on an ambient light level in a viewing area of the user; and output, for display by a display device in the viewing area of the user, a version of the image frame that contains the modified region.

Identifying regions of visible media data that belong to a trigger content type

Embodiments of the invention provide methods for identifying suspect heart or lung sounds and methods for generating an isolated sound sample. In some embodiments, a method for identifying suspect heart or lung sounds comprises obtaining, by processing circuitry, an instance of data samples comprising at least two audio samples and respiratory and/or cardiac cycle data, the audio samples comprising auscultation sounds of an individual, based on the respiratory and/or cardiac cycle data, processing the at least two audio samples to generate a plurality of sound samples, identifying a primary sound sample, generating at least one of a modified secondary sound or a modified baseline sample and combining the modified secondary sound sample or the modified baseline sample with the primary sound sample to generate an isolated sound sample.

Electronic auscultation and improved identification of auscultation audio samples

An example method comprises determining, by a computing system, that a message-sending computing device has received an indication of user input indicating an intent to send a message to a user of a message-recipient computing device; prior to the message being delivered to the message-recipient computing device, calculating, by the computing system, a screen sharing risk score (SSRS) that indicates a risk that the message-recipient computing device is currently sharing screen content of a screen of the message-recipient computing device with a screen-recipient computing device associated with an untrusted screen-recipient user; and based on the SSRS indicating that the risk is above a risk threshold, causing the message-sending computing device to output an alert indicating that the message-recipient computing device is possibly sharing the screen content of the message-recipient computing device with the screen-recipient computing device associated with the untrusted screen-recipient user.

Preventing sensitive information from being screen shared with untrusted users

Data ownership of a single record data object comprising a plurality of individual data elements may be distributed across a plurality of users, such that each individual user is capable of separately controlling access to those data elements for which the data owner has ownership privileges. These data ownership privileges, and corresponding access rights which may be individually provided by distinct data owners, is managed by a data management computing entity such that a single composite user interface may be generated for a user viewing a particular record data object such that the viewer is provided with viewing access to only those data elements for which the viewer has access. Thus, separate user interfaces may be generated and provided for different viewers accessing the same record data object.

Growing community of inventors

Thompson Station, TN, United States of America

Jon Kevin Muse