Speech-based pulmonary assessment
Pulmonary assessment based on speech can include identifying one or more audio features and speech patterns of a user's speech. A cognitive burden associated with the user's speech can be determined. A pulmonary condition of the user can be determined based on predetermined correlations between the one or more audio features and speech patterns of the user's speech, the cognitive burden, and a respiratory airway condition.
1. A computer-based method, comprising:
extracting from a user's speech, using computer hardware, audio features including Mel-frequency cepstral coefficients (MFCCs) specifying speech patterns of the user's speech;
calculating, by the computer hardware, a metric specifying cognitive burden associated with the user's speech based on at least one of complexity of words or grammatical structure of the user's speech; and
determining, by the computer hardware, a pulmonary condition of the user by processing the audio features and the metric of cognitive burden through a predictive model including one or more first convolutional neural network layers trained to extract spatiotemporal features pertaining to airway anomalies and one or more second long short-term memory layers trained to correlate the spatiotemporal features with lung functions.
2. The method of claim 1 , further comprising:
generating a customized script specific to the user, wherein the user's speech is a vocalization of the customized script.
3. The method of claim 1 , further comprising:
generating a customized script specific to the user, wherein the customized script comprises words and grammatical structure selected based on a likelihood of causing the user to improve the pulmonary condition in response to multiple readings of the customized script by the user.
4. The method of claim 1 , further comprising:
determining a respiratory airway condition of the user multiple times over a predetermined time interval; and
conveying a pulmonary assessment to at least one of the user or a healthcare professional in response to detecting a predetermined change in the pulmonary condition of the user during the predetermined time interval.
5. The method of claim 1 , further comprising:
determining whether the pulmonary condition of the user corresponds to an airway anomaly of the user, wherein the airway anomaly includes at least one of an airway obstruction or an airway restriction.
6. The method of claim 1 , wherein
determining the pulmonary condition of the user comprises determining one or more lung function parameters.
7. The method of claim 6 , wherein
the one or more lung function parameters include at least one of an estimated forced expiratory volume in one second (FEV1) of the user or a forced vital capacity (FVC) of the user.
8. A system, comprising:
one or more processors configured to initiate operations including:
extracting from a user's speech audio features including Mel-frequency cepstral coefficients (MFCCs) specifying speech patterns of the user's speech;
calculating a metric specifying cognitive burden associated with the user's speech based on at least one of complexity of words or grammatical structure of the user's speech; and
determining a pulmonary condition of the user by processing the audio features and the metric of cognitive burden through a predictive model including one or more first convolutional neural network layers trained to extract spatiotemporal features pertaining to airway anomalies and one or more second long short-term memory layers trained to correlate the spatiotemporal features with lung functions.
9. The system of claim 8 , wherein the processor is configured to initiate operations further including:
generating a customized script specific to the user, wherein the user's speech is a vocalization of the customized script.
10. The system of claim 8 , wherein the processor is configured to initiate operations further including:
generating a customized script specific to the user, wherein the customized script comprises words and grammatical structure selected based on a likelihood of causing the user to improve the pulmonary condition in response to multiple readings of the customized script by the user.
11. The system of claim 8 , wherein the processor is configured to initiate operations further including:
determining a respiratory airway condition of the user multiple times over a predetermined time interval; and
conveying a pulmonary assessment to at least one of the user or a healthcare professional in response to detecting a predetermined change in the pulmonary condition of the user during the predetermined time interval.
12. The system of claim 8 , wherein the processor is configured to initiate operations further including:
determining whether the pulmonary condition of the user corresponds to an airway anomaly of the user, wherein the airway anomaly includes at least one of an airway obstruction or an airway restriction.
13. The system of claim 8 , wherein
determining the pulmonary condition of the user comprises determining one or more lung function parameters.
14. A computer program product, the computer program product comprising:
one or more computer-readable storage media and program instructions collectively stored on the one or more computer-readable storage media, the program instructions executable by a processor to cause the processor to initiate operations including:
extracting from a user's speech audio features including Mel-frequency cepstral coefficients (MFCCs) specifying speech patterns of the user's speech;
calculating a metric specifying cognitive burden associated with the user's speech based on at least one of complexity of words or grammatical structure of the user's speech; and
determining a pulmonary condition of the user by processing the audio features and the metric of cognitive burden through a predictive model including one or more first convolutional neural network layers trained to extract spatiotemporal features pertaining to airway anomalies and one or more second long short-term memory layers trained to correlate the spatiotemporal features with lung functions.
15. The computer program product of claim 14 , wherein the program instructions are executable by the processor to cause the processor to initiate operations further including:
generating a customized script specific to the user, wherein the user's speech is a vocalization of the customized script.
16. The computer program product of claim 14 , wherein the program instructions are executable by the processor to cause the processor to initiate operations further including:
generating a customized script specific to the user, wherein the customized script comprises words and grammatical structure selected based on a likelihood of causing the user to improve the pulmonary condition in response to multiple readings of the customized script by the user.
17. The computer program product of claim 14 , wherein the program instructions are executable by the processor to cause the processor to initiate operations further including:
determining a respiratory airway condition of the user multiple times over a predetermined time interval; and
conveying a pulmonary assessment to at least one of the user or a healthcare professional in response to detecting a predetermined change in the pulmonary condition of the user during the predetermined time interval.
18. The computer program product of claim 14 , wherein the program instructions are executable by the processor to cause the processor to initiate operations further including:
determining whether the pulmonary condition of the user corresponds to an airway anomaly of the user, wherein the airway anomaly includes at least one of an airway obstruction or an airway restriction.
19. The computer program product of claim 14 , wherein
determining the pulmonary condition of the user comprises determining one or more lung function parameters.
20. The computer program product of claim 19 , wherein
the one or more lung function parameters include at least one of an estimated forced expiratory volume in one second (FEV1) of the user or a forced vital capacity (FVC) of the user.