Screening of individuals for a respiratory disease using artificial intelligence

A method for predicting a likelihood of having a respiratory infection in an individual, the method implemented within a screening system, the screening system has a processor and a memory, the method comprising:

• receiving, by the screening system, a sound sample from the individual, the sound sample includes a cough sound, nasal breathing sounds, and vocalization sounds;
• pre-processing the sound sample, by a pre-processing module, to remove noise and extract a set of unique latent features, the pre-processing module implemented within the screening system;
• processing the set of unique latent features by a plurality of machine learning based classifiers;
• computing, by a mediator module implemented within the system an average weightage of a plurality of outputs from the plurality of classifiers, based on the weightage assigned to each classifier of the plurality of classifiers to obtain a result, the plurality of outputs are obtained via processing the set of unique latent features of the sound samples by the plurality of classifiers, based on the result classifying the individual into a category indicative of the likelihood of having the respiratory infection or into another category indicative of a likelihood of not having the respiratory infection, wherein each classifier is configured to classify the individual into the two categories, wherein each classifier is assigned a weightage;
• the weightage is assigned to each classifier based on a parameter of a plurality of parameters used for generating each classifier, wherein the plurality of parameters are selected from a group consisting of biomarkers, metadata, neurologic manifestations including central nervous system manifestations and peripheral nervous system manifestations, and images of cutaneous manifestations of the respiratory infection in a form of anomalies in skin;
• and
• communicating, by the mediator module, the result to the individual.

A screening system for predicting a likelihood of having a disease in an individual, the disease effecting respiratory sounds, the screening system having a processor and a memory, the system configured to implement a method comprising:

• receiving, by the screening system, cough sound samples from subjects suffering from the disease and subjects not suffering from the disease;
• pre-processing the cough samples, by a pre-processing module implemented within the screening system, into labeled features obtaining a training dataset;
• randomly partitioning the training dataset into a plurality of training datasets;
• generating a plurality of classifiers, by the screening system using machine learning, based on the plurality of training datasets, wherein at least one classifier of the plurality of classifiers is a deep learning classifier;
• assigning a weightage, by the screening system, to each classifier of the plurality of classifiers, the weightage is assigned to each classifier based on a parameter of a plurality of parameters used for generating each classifier, wherein the plurality of parameters are selected from a group consisting of biomarkers, metadata, neurologic manifestations including central nervous system manifestations and peripheral nervous system manifestations, and images of cutaneous manifestations of the respiratory infection in a form of anomalies in skin;
• computing, by a mediator module implemented within the system an average weightage of a plurality of outputs from the plurality of classifiers, based on the weightage assigned to each classifier of the plurality of classifiers to obtain a result, the plurality of outputs are obtained via processing a set of unique latent features of the sound samples by the plurality of classifiers, wherein the result classifies the individual into a category indicative of the likelihood of having the disease or into another category indicative of the likelihood of not having the disease, wherein each classifier is configured to distinguish the individual into one of the two categories, the categories being positive and negative;
• communicating, by the mediator module, the result to the individual.