ENSEMBLE MACHINE-LEARNING MODELS TO DETECT RESPIRATORY SYNDROMES

1 - 20 . (canceled)

21 . A tangible, non-transitory, machine-readable medium storing instructions that when executed by one or more processors effectuate operations comprising:

obtaining, with a computer system, a trained machine learning model configured to infer a presence, absence, stage, or risk of a respiratory condition based on input data comprising audio captured by mobile computing device of a user, wherein the trained machine learning comprises a double parallel feedforward neural network and wherein the trained machine learning model is trained by:

obtaining a training set comprising a plurality of training records, wherein:

each training record in the training set includes a plurality of parameters and corresponding values for a respective person;

each training record in the training set includes audio of the respective person's voice or cough; and

each training record in the training set includes an indicator indicating whether or not the respective person has been diagnosed with a respiratory illness; and

training the machine learning model on the training set to infer whether a user has the respiratory illness based on the audio;

receiving, with the computer system, a first user record of a first user, the first user record comprising an audio file or stream of a voice of the first user;

inferring, with the computer system, that the first user has a presence, absence, stage, or risk of a respiratory condition based on the audio file or stream of a voice of the first user; and

storing, with the computer system, the presence, absence, stage, or risk of a respiratory condition in memory.

22 . The medium of claim 1 , wherein the plurality of training records comprises at least two of the following:

textual questionnaire answers;

data indicative of breathing; or

temporal data.

23 . The medium of claim 1 , the operations further comprising:

steps for training the machine learning model.

24 . The medium of claim 1 , wherein:

training the machine learning model comprises computing partial derivatives of parameters of the machine learning model with respect to an objective function and adjusting the parameters of the machine learning model in directions the partial derivatives indicate locally optimize the machine learning model.

25 . The medium of claim 1 , wherein the machine learning model comprises:

at least two outputs, with a first output indicating respiratory infection and a second output indicating the stage of respiratory infection.

26 . The medium of claim 1 , wherein:

the machine learning model comprises means for ensembling output of a plurality of sub-models.

27 . The medium of claim 1 , wherein the operations further comprise:

configuring lossy compression of the audio file or stream of the person's voice to preserve non-human perceivable data that affects accuracy of the trained machine learning model.

28 . The medium of claim 1 , wherein:

training the machine learning model is performed by a different set of computers of the computer system than a set of computers of the computer system that perform inferring the first user has a presence, absence, stage, or risk of a respiratory condition.

29 . The medium of claim 1 , wherein:

inferring that the first user has the respiratory illness is performed by a smartphone of the first user that is part of the computer system.

30 . The medium of claim 1 , wherein:

the trained machine learning model comprises an ensemble of at least three different machine learning algorithms having outputs combined with means for ensembling multiple sub-models.

31 . The medium of claim 1 , the operations further comprising:

performing a pre-processing of an audio voice sample, before inputting to the trained machine learning model, to clean the voice sample and select segments of the voice sample to be input to the trained machine learning model.

32 . The medium of claim 1 , wherein the operations further comprise:

extracting cepstral coefficients from the audio file or stream of the voice of the first user.

33 . The medium of claim 32 , wherein extracting the cepstral coefficients comprises:

constructing a spectrogram from the audio file or stream of the voice of the first user;

calculating a log-power for each frame from the spectrogram;

applying a filter to a magnitude of the log-power;

performing a log-compression and transform to a cepstral domain of an output of the filter; and

forming a vector of cepstral coefficients for each frame.

34 . The medium of claim 1 , wherein the operations comprise:

extracting mel-frequency cepstral coefficients derived from a power spectrum of audio of a voice sample of a second user.

35 . The medium of claim 1 , further comprising:

computing quality metrics of the input data, the quality metrics comprising at least one of a signal-to-noise ratio, a blur level, or a lighting level, wherein the trained machine learning model is configured to adjust weighting of the input data in response to the quality metrics during inferring of the respiratory condition.

36 . The medium of claim 1 , wherein:

the input data further comprises an image captured by a mobile computing device;

each training record in the training set further includes an image of at least part of the respective person;

training the machine learning model on the training set to infer whether a user has the respiratory illness is based on the audio and images;

the first user record further comprises an image of at least part of the first user; and

inferring that the first user has a presence, absence, stage, or risk of a respiratory condition is based on the audio file or stream of a voice of the first user and an image of at least part of the first user.

37 . The medium of claim 36 , wherein the plurality of training records comprises at least two of the following:

textual questionnaire answers;

data indicative of breathing; or

temporal data;

facial image;

fingertip video; or

biometric images of skin, stool, mucus, urine, or vomit.

38 . The medium of claim 36 , wherein the plurality of training records comprises each of the following:

textual questionnaire answers;

data indicative of breathing; and

temporal data;

facial image;

fingertip video; and

biometric images of skin, stool, mucus, urine, or vomit.

39 . The medium of claim 36 , wherein the trained machine learning model is arranged to transform acoustic input data into a first set of feature values and visual input data into a second set of feature values, and wherein the first set of feature values and the second set of feature values are combined by one of: joining the feature values into a single representation prior to classification, merging classification results derived separately from the feature values, or adjusting the relative influence of the feature values through context-dependent weighting.

40 . A method, comprising:

obtaining, with a computer system, a trained machine learning model configured to infer a presence, absence, stage, or risk of a respiratory condition based on input data comprising audio captured by mobile computing device of a user, wherein the trained machine learning comprises a double parallel feedforward neural network and wherein the trained machine learning model is trained by:

obtaining a training set comprising a plurality of training records, wherein:

each training record in the training set includes a plurality of parameters and corresponding values for a respective person;

each training record in the training set includes audio of the respective person's voice or cough; and

each training record in the training set includes an indicator indicating whether or not the respective person has been diagnosed with a respiratory illness; and

training the machine learning model on the training set to infer whether a user has the respiratory illness based on the audio;

receiving, with the computer system, a first user record of a first user, the first user record comprising an audio file or stream of a voice of the first user;

inferring, with the computer system, that the first user has a presence, absence, stage, or risk of a respiratory condition based on the audio file or stream of a voice of the first user; and

storing, with the computer system, the presence, absence, stage, or risk of a respiratory condition in memory.