A sleep monitor for monitoring baby sleep uses sleep state classification based on heartbeat feature respiration features. The sleep monitor automatically retrains the classification during use of the sleep monitor. Training examples for use in this training process are generated automatically by detecting time instants whereat the baby in the bed is in a wake state, based on signals from the at least one of a sound feature detector a movement feature detector (112) and an open eye detector (114). The retraining may comprise using time sequence from the end of detection of wake states to assign a class to heartbeat feature and/or respiration feature values during that time sequence for the training process. In an embodiment, the retraining comprises clustering detected heartbeat feature and/or respiration feature values detected outside the detected wake states.

Various embodiments described herein relate to a method and related wearable device and non-transitory machine-readable medium including receiving sensor data from at least one sensor of the wearable device; comparing the sensor data to schedule format stored in a memory of the wearable device, wherein the schedule format specifies at least one characteristic of sensor readings previously associated with a predefined context; determining that the received sensor data matches the schedule format; determining, based on the received sensor data matching the schedule format, that the user is currently in the predefined context; identifying an action associated with the predefined context; and executing the action while the user is in the predefined con text.

An emotion analysis method and an electronic apparatus thereof are provided. The emotion analysis method is adapted to the electronic apparatus having a database or connected to the database in order to analyze an emotion of an examinee. The emotion analysis method includes: obtaining a heart rate signal of the examinee; defining a plurality of candidate emotions from the database; analyzing the heart rate signal to obtain a plurality of target emotion parameters; and analyzing the target emotion parameters to determine one of the candidate emotions corresponding to the heart rate signal by applying an emotion analysis model.

Systems and methods for predicting exposure to an agent. One or more features are extracted from physiological data. For each respective classifier, (i) the respective classifier is identified, wherein the respective classifier is trained using training data for a respective physiological state, (ii) the respective classifier is applied to the one or more features to obtain a classifier output that represents a likelihood of exposure, (iii) a respective first threshold is applied to the classifier output to determine a patient state classification, and (iv) the patient state classifications are aggregated across a number of time intervals to obtain an aggregate patient state classification for each classifier. The aggregate patient state classifications are combined across the plurality of classifiers to obtain a combined classification, and an indication that the patient has been exposed to the agent is provided when the combined classification exceeds a second threshold.

Systems and methods for processing sensor data and self-calibration are provided. In some embodiments, systems and methods are provided which are capable of calibrating a continuous analyte sensor based on an initial sensitivity, and then continuously performing self-calibration without using, or with reduced use of, reference measurements. In certain embodiments, a sensitivity of the analyte sensor is determined by applying an estimative algorithm that is a function of certain parameters. Also described herein are systems and methods for determining a property of an analyte sensor using a stimulus signal. The sensor property can be used to compensate sensor data for sensitivity drift, or determine another property associated with the sensor, such as temperature, sensor membrane damage, moisture ingress in sensor electronics, and scaling factors.

A computer implemented method for providing pain management using a wearable device determines a predictive model estimating an intensity level of pain as a function of at least one physiological parameter of a user of the wearable device and at least one activity of the user. The activity of the user includes one or combination of a type of the activity, a level of the activity, a location of the activity, and a duration of the activity. The method determines measurements of physiological and activity sensors of the wearable device to produce values of the physiological parameter and the activity of the user and predicts the intensity level of the pain based on the predictive model and the values of the physiological parameter and the activity of the user. The method executes actions based on the predicted intensity level of pain.

A technology for a wearable medical device for monitoring medical parameters. Medical measurement data can be received at the wearable medical device from a medical measurement sensor attached to the wearable medical device or a medical measurement sensor in communication with the wearable medical device. A calibration coefficient can be determined for calibrating the wearable medical device based on the medical measurement data. The wearable medical device can be calibrated based on the calibration coefficient.

A patient health management platform accesses a metabolic profile for a patient and biosignals recorded for the patient during a current time period comprising sensor data and/or lab test data collected for the patient. The platform receives patient data recorded during the current time period comprising food items consumed, medications taken, and symptoms experienced by the patient. The platform implements a machine-learned metabolic model to determine a metabolic state of the patient at a conclusion of the current time period by comparing a true representation of the metabolic state and a prediction of the metabolic state. The true representation and the prediction are determined based on the recorded biosignals and the recorded patient data, respectively. The platform generates a patient-specific treatment recommendation outlining instructions for the patient to improve their metabolic state and provides the patient-specific treatment recommendation to the patient device for display to the patient.

Methods and apparatus provide physiological movement detection, such as gesture, breathing, cardiac and/or gross motion, such as with sound, radio frequency and/or infrared generation, by electronic devices such as vehicular processing devices. The electronic device in a vehicle may, for example, be any of an audio entertainment system, a vehicle navigation system, and a semi-autonomous or autonomous vehicle operations control system. One or more processors of the device, may detect physiological movement by controlling producing sensing signal(s) in a cabin of a vehicle housing the electronic device. The processor(s) control sensing, with a sensor, reflected signal(s) from the cabin. The processor(s) derive a physiological movement signal with the sensing signal and reflected signal and generate an output based on an evaluation of the derived physiological movement signal. The output may control operations or provide an input to any of the entertainment system, navigation system, and vehicle operations control system.

System for performing fully automatic brain tumor and tumor-aware cortex reconstructions upon receiving multi-modal MRI data (T1, T1c, T2, T2-Flair). The system outputs imaging which delineates distinctions between tumors (including tumor edema, and tumor active core), from white matter and gray matter surfaces. In cases where existing MRI model data is insufficient then the model is trained on-the-fly for tumor segmentation and classification. A tumor-aware cortex segmentation that is adaptive to the presence of the tumor is performed using labels, from which the system reconstructs and visualizes both tumor and cortical surfaces for diagnostic and surgical guidance. The technology has been validated using a publicly-available challenge dataset.