Abstract: Techniques for improved machine learning are provided. Patient data for a patient is received, the patient data relating to an action performed by a caregiver. A medical code is generated for the action by processing the patient data using a machine learning model, and the action is associated with the medical code. An event record, including the medical code, is generated for the patient.

Abstract: The present disclosure provides for the division of environmental sounds from speech sounds to extract and analyze the behaviors and activities occurring in the environment; thus expanding and improving the functionality of AI assistant devices. These environmental sounds can be securely uploaded to an electronic chart, and may be used to aid in the treatment of existing conditions or the prophylaxis/mitigation of conditions not yet experienced by a patient under observation. Accordingly, the present disclosure provides for improved functionality in assistant devices and devices linked to the AI assistant devices.

Abstract: Techniques for improved data services are provided. A check-in scan indicating a user device in a medical facility is received, and a medical professional associated with the user device is identified. A patient of the medical facility is identified based at least in part on the check-in scan. Access permission of the medical professional is validated, with respect to the patient, based at least in part on the check-in scan. In response to validating the access permission of the medical professional, credentials of the medical professional are not requested, a patient profile of the patient is automatically retrieved via one or more networks, and input describing the patient, from the medical professional, is automatically stored in the patient profile via the one or more networks.

Abstract: Techniques for improved sensor monitoring using machine learning are provided. An elopement likelihood indicating a probability of elopement for a patient in a care setting is generated by processing first patient data using a trained machine learning model. One or more sensor devices are enabled in response to determining that the elopement likelihood exceeds a threshold. Second patient data is collected for the patient using the enabled one or more sensor devices. An intervention for the patient is selected based on the second patient data, and the intervention is enacted.

Abstract: Methods and devices include determining that a prompt is to be displayed on a display of a user device to present information to a user of the user device, to receive an input from the user of the user device, or a combination. The methods and devices further include determining a classification of the prompt based, at least in part, on the information to be presented, the input to be received, context associated with current use of the user device by the user, or a combination thereof. The methods and devices further include causing, at least in part, a presentation of a graphical user interface element on the display of the user device based, at least in part, on the classification of the prompt.

Abstract: Certain aspects of the present disclosure provide techniques for predicting wound management treatment resources. This includes determining characteristics of a wound for a patient based on an image of the wound, including detecting the characteristics based on analyzing the image using a first machine learning (ML) model trained to detect wound characteristics from a captured image. The techniques further include predicting at least one of: (i) treatment resources or (ii) a treatment facility for treating the wound, including providing to a second trained ML model characteristics of the wound, patient medical data for the patient, and treatment facility data describing a plurality of available treatment facilities.

Abstract: Certain aspects of the present disclosure provide techniques for security management of health information using artificial intelligence assistant by receiving, at an artificial intelligence (AI) assistant device from a requestor in an environment, an utterance including a request to provide health information related to a patient and confirming, via a machine learning model hosted by the AI assistant device, whether an unauthorized person is present in the environment with the AI assistant device, where the unauthorized person is not permitted by the patient to receive the health information but is permitted to interact with the AI assistant device. Further, in response to determining that the unauthorized person is present, generating, by the AI assistant device, an audio deferral that does not include the health information that was requested.

Abstract: Techniques for improved machine learning are provided. Activity data describing an activity for a financial account of a resident in a residential care facility is received, and a set of attributes corresponding to the activity is extracted from the activity data, comprising determining a first attribute of the set of attributes by processing unstructured text associated with the activity using one or more natural language processing techniques. A validity score is generated by processing the set of attributes using a trained machine learning model, where the validity score indicates a probability that the activity is valid. In response to determining that the validity score is below a defined threshold, one or more interventions are initiated for the resident.

Abstract: Embodiments herein determine when to place a passive assistive call using personal artificial intelligence (AI) assistants. The present embodiments improve upon the base functionalities of the assistant devices by monitoring the usually discarded or filtered-out environmental sounds to identify when a person is in distress to automatically issue an assistive call in addition to or alternatively to monitoring user speech for active commands to place assistive calls. The assistant device may be in communication with various other sensors to enhance or supplement the audio assessment of the persons in the environment, and may be used in a variety of scenarios where prior call systems struggled to quickly and accurately identify distress in various monitored persons (e.g., patients) including falls, stroke onset, and choking.

Abstract: A sentiment analysis system is described that uses a machine learning model to predict a mood of an individual based on the motions used by the individual when performing sign language. The present disclosure uses image processing to monitor how a non-verbal individual communicates to, for example, predict overall health and mental well-being.

Abstract: Embodiments herein describe middleware that serves as an intermediary between two computing systems performing asynchronous communication. Rather than communicating directly with each other, the two computing system communicate with the middleware; however, the middleware can be introduced seamlessly so that the two system believe they are communicating with each other rather than the middleware—i.e., the middleware is invisible to the systems. Asynchronous communication with guaranteed delivery may require the two computing systems to perform a handshake operation where the computing system transmitting the object does not consider the task complete until receiving a confirmation from the receiving computing system that the object was processed successfully.

Abstract: Embodiments herein describe tracking the use of medical devices to update their remaining life. In addition to tracking the remaining life of the overall device, the embodiments herein can track the remaining life of each component which may be greater than the life of the medical device. Unlike typical modular devices which are replaced after they break or exceed their expected life, one-time use (or limited use) medical devices may be limited to use by a specific patient before they are discarded. The embodiments herein can use communication between healthcare providers and a tracker database to track the usage of a medical device and update the remaining life of its components to identify components that can be recycled after the medical device is no longer being used.

Abstract: Certain aspects of the present disclosure provide techniques for wound management and treatment. This includes determining characteristics of a wound for a patient based on an image of the wound, including detecting the characteristics based on analyzing the image using a first ML model. The techniques further include identifying patient medical data including characteristics relating to a medical history for the patient, and predicting a first care plan for the patient based on providing the characteristics of the wound and the patient medical data to a second ML model. The first care plan is configured to be used to treat the wound for the patient.

Abstract: Certain aspects of the present disclosure provide techniques for wound management and treatment. This includes determining characteristics of a wound for a patient based on an image of the wound, including detecting the characteristics based on analyzing the image using a first ML model trained to detect wound characteristics from a captured image. The techniques further include identifying patient medical data including characteristics relating to a medical history for the patient, and predicting a first care plan for the patient based on providing the characteristics of the wound and the patient medical data to a second ML model. The second ML model is trained to predict the first care plan using prior wound care outcome data including a prior wound care outcomes relating to prior patients. The first care plan is configured to be used to treat the wound for the patient.

Abstract: Certain aspects of the present disclosure provide techniques for wound management and treatment. This includes determining characteristics of a wound for a patient based on an image of the wound, including detecting the characteristics based on analyzing the image using a first ML model trained to detect wound characteristics from a captured image. The techniques further include identifying patient medical data including characteristics relating to a medical history for the patient, and predicting a first care plan for the patient based on providing the characteristics of the wound and the patient medical data to a second ML model. The second ML model is trained to predict the first care plan using prior wound care outcome data including a prior wound care outcomes relating to prior patients. The first care plan is configured to be used to treat the wound for the patient.