The present disclosure provides systems and methods that generating health diagnostic information from an audio recording. A computing system can include a machine-learned health model comprising that includes a sound model trained to receive data descriptive of a patient audio recording and output sound description data. The computing system can include a diagnostic model trained to receive the sound description data and output a diagnostic score. The computing system can include at least one tangible, non-transitory computer-readable medium that stores instructions that, when executed, cause the processor to perform operations. The operations can include obtaining the patient audio recording; inputting data descriptive of the patient audio recording into the sound model; receiving, as an output of the sound model, the sound description data; inputting the sound description data into the diagnostic model; and receiving, as an output of the diagnostic model, the diagnostic score.

A plurality of individually addressable electrodes is supported by a housing. The individually addressable electrodes are for at least one of applying stimulation electrical signals to skin of a user and detecting biometric electrical signals from the skin of the user. At least one of a signal detector is provided for detecting the biometric electrical signals and a signal generator is provided for generating the stimulation electrical signals. An electrode multiplex circuit is provided for addressing the plurality of individually addressable electrodes by at least one of routing the biometric electrical signals from the skin of the user through more than one of the plurality of individually addressable electrodes to the signal detector and routing the stimulation electrical signals from the signal generator through more than one of the plurality of individually addressable electrode to the skin of the user. A microprocessor is provided for controlling at least one of the signal detector, the signal generator, the electrode multiplex circuit.

The exemplified system and method facilitate an objective, non-invasive measurement of myelin quality and integrity in living brains based on isolation of myelin-specific magnetic relaxation constants in k-space. The system uses magnetic resonance (MR) signals to select signatures specific to, or associated with, myelin and its structure and to then encode the selected signatures into an image or model to which the quantitative myelin health information can be co-registered with 2D, 3D visualization, and tractography of the myelin-signal isolated MR information. The system also sets a model for digital hierarchical learning of biomedical signals in MR, and beyond, based on experimental data in which it executes the herein described signal isolation operations.

A method and system for remote diagnosis of a cognitive disorder in humans are provided. The system comprises receive, over a network, at least one facial image of a patient; extracting, from the at least one facial image, at least one facial feature indicative of a cognitive decline; classifying the at least one extracted facial feature using a classifier, wherein the classifier maps a plurality of candidate facial features to a score indicating a stage of a cognitive decline; and determining a positive diagnosis of a cognitive decline of the patient, based on the score provided by the classifier in response to the at least one extracted facial feature.

A controller and associated multi-axis sensor system for augmenting the automatic intelligent delivery of one or more drugs is provided. The controller and associated multi-axial sensor system are based on the detection and determination of particular physical lifestyle events. As a specific example, a pump augmentation system includes a six-axis accelerometer sensor, a gyroscopic pitch sensor and a controller. The controller is configured to receive motion data from the six-axis accelerometer sensor and orientation data from the gyroscopic pitch sensor. The controller provides a pump instruction signal for changing a delivery rate of a drug to a user based on the motion data and the orientation data. The system and methods are particularly suited for treating a user with Parkinson's disease.

A method includes engaging Parkinson's Disease (PD) subjects in a continuous motor performance task that elicits natural motor variability, quantifying natural motor variability of each PD subject with an array of motor metrics at short timescales, applying a machine-learning classification or regression algorithm to determine weights for each of these metrics to maximally differentiate each patient's motor performance from that of controls performing the same task, and combining the weights to determine a scalar metric of motor performance for each short epoch of motor behavior.

A method of providing voice therapy or training to a patient or other person over a communications network. The method includes delivering by a server computer communicatively connected to the communications network, a digital exercise instruction to a user client device communicatively connected to the communications network. The method includes receiving over the communications network from the user client device a digital voice signal representing an analog voice signal input to the user client device by the patient or other person, and storing the digital voice signal in a database communicatively connected to the server computer. The method also includes delivering a website to a Speech Language Pathologist (SLP) device or other administrator device communicatively connected to the communications network and providing access to the digital voice signal to the administrator device.

A technical solution is described for implementing a computer-executed signal processing algorithm to search for time domain segments of a recorded electroencephalogram (EEG) that are highly correlated, either positively or negatively, to one or more, individual, synthetically generated, single-period single-frequency (SPSF) sinusoids. The SPSFs are motivated by the combined concepts of individual Striatal Beat Frequencies (SBF) used to model cortical neuron activity, Frequency Domain Reflectometry used to study Voltage Standing Wave Ratios (VSWR), Geophysics Seismograms, and ghosting effects of multipath passing through periodic sinusoids. This computationally intense approach is only recently realizable through the advent of high performance computing. The SPSF approach, since it is not constrained to the error-laden one-window-fits-all approach of the Time-Frequency Spectrogram, offer's a more detailed basis to assess, and truer visualization of, the health of brain's electrical activities. This approach is a push-back against the Uncertainty Principal.

Provided herein are, inter alia, methods for identifying subjects suffering from depression that will respond to treatment with an antidepressant.

A patient therapy method and program product are provided for communicating with a patient monitor to obtain data from a patient engaged in physical therapy. A library of musical features includes a plurality of coordinated musical tracks associated with respective functional therapeutic outcomes. The functional therapeutic outcomes include gait characteristics comprising at least at strike, cadence, stride length deviations. Based on a selected functional therapeutic outcome, musical tracks are selected and mixed together at a desired tempo. The tracks are adjusted based on changes in patient data.