Abstract: A method comprises obtaining ear modeling data representing a 3-dimensional (3D) impression of an ear surface of an ear of a patient; determining, based on the ear modeling data, values of landmarks of the ear, wherein the landmarks include ear canal landmarks of an ear canal, and determining the values of the landmarks comprises: predicting an ear aperture plane of an aperture of the ear; determining a plurality of cross-sectional planes that are aligned with the ear aperture plane; for each of the cross-sectional planes: determining an intersection boundary of the cross-sectional plane representing a line of intersection between the cross-sectional plane and the ear canal; and determining a centroid of the intersection boundary of the cross-sectional plane; and determining values of the ear canal landmarks based on the centroids.

Abstract: A hearing device includes a microphone that produces an audio input signal and a loudspeaker that outputs an amplified audio signal into an ear canal. A signal processing path is coupled to the microphone and the loudspeaker. The signal processing path includes a deep neural network configured to predict an instantaneous gain margin of the hearing device based on a set of inputs. The set of inputs includes a first parameter of the audio input signal, a second parameter of the amplified audio signal, and a gain of the signal processing path.

Abstract: A method comprises obtaining ear modeling data, wherein the ear modeling data includes a 3D model of an ear canal; applying a shell generation to generate a shell shape based on the ear modeling data, wherein the shell-generation model is a machine learning model and the shell shape is a 3D representation of a shell of an ear-wearable device; applying a set of one or more component-placement models to determine, based on the ear modeling data, a position and orientation of a component of the ear-wearable device, wherein the component-placement models are independent of the shell-generation model and each of the component-placement models is a separate machine learning model; and generating an ear-wearable device model based on the shell shape and the 3D arrangement of the components of the ear-wearable device.

Abstract: An ear-wearable electronic device comprises a motion sensor configured to generate motion information and a first respiration rate estimated using the motion information. A PPG sensor is configured to generate PPG data and a second respiration rate estimate using the PPG data. A processor is configured to produce a respiration rate estimate using the first and second respiration rate estimates. A communication device is configured to communicate the respiration rate estimate to one or both of an external electronic device and a cloud database.

Abstract: The present invention provides a system and a method for denoising motion artifacts from vital signs signals comprising steps of receiving one or more reflected signals from at least one subject, generating a time sequence buffer of reflected signals of predefined duration for vital signals processing, source separation and a component selection, and estimating and tracking vital signs of the source separation and a component selection.

Abstract: A method for reconstructing high signal-to-noise ratio (SNR) magnetic resonance imaging (MRI) slices, including: receiving a thick MRI slice of bodily tissue acquired using a single MRI scan, wherein the thick slice has a high SNR; receiving two thin MRI slices of the bodily tissue acquired using a single MRI scan, wherein each of the two thin MRI slices has a low SNR; and reconstructing multiple high SNR thin slices of the bodily tissue using the thick slice and the two thin slices.

Abstract: A method for reconstructing high signal-to-noise ratio (SNR) magnetic resonance imaging (MRI) slices, including: receiving a thick MRI slice of bodily tissue acquired using a single MRI scan, wherein the thick slice has a high SNR; receiving two thin MRI slices of the bodily tissue acquired using a single MRI scan, wherein each of the two thin MRI slices has a low SNR; and reconstructing multiple high SNR thin slices of the bodily tissue using the thick slice and the two thin slices.

Abstract: A method and a device for the acoustic detection of a one lung intubation situation in a human subject are disclosed. According to some embodiments, the disclosed method includes computing an autoregressive moving average (ARMA) or an autoregressive function of electrical signals received from acoustic detectors placed at different locations on the body of the subject. Appropriate locations for acoustic detectors include the back region and the chest region. The disclosed method and apparatus are insensitive to uncancelled, random background noise with a loudness associated with an operating room or intensive care ward. The disclosed device is configurable so that the relative occurrence rate of missed detections or false negatives and false positive alarms can be modified. In one exemplary embodiment, the device is adapted such that at most 9% of identifications are false positive identifications, and at most 2% of identifications are false negative identifications.