Pattern recognition system for quantifying the likelihood of the contribution of multiple possible forms of chronic disease to patient reported dyspnea

Abstract

Systems and methods for quantifying the likelihood of the contribution of multiple possible forms of chronic disease to patient reported dyspnea can include the testing protocol having a flow/volume loop, performed at rest, flowed by the measurement of cardiopulmonary exercise gas exchange variables during rest, exercise and recovery as unique data sets. The data sets are analyzed using feature extraction steps to produce a pictorial image consisting of disease silos displaying the likelihood of the contribution of various chronic diseases to patient reported dyspnea. In some embodiments, the silos are split into subclass silos. In some embodiments, multiple chronic disease indexes are used to differentiate between sub-types of a particular chronic disease (e.g., differentiating WHO 1 PH from WHO 2 or WHO 3 PH). Test results are plotted serially to provide feedback to the physician on the efficacy of therapy provided to the patient.

What is claimed is: 
     
         1 . A method for characterizing contributions of physiological conditions to dyspnea in a patient comprising:
 receiving gas exchange measurements that are based on breath-by-breath data captured during a gas exchange test on a patient;   determining, using a machine learning classifier, a plurality of contribution values that are each associated with a different physiological condition, wherein the plurality of contribution values are numeric values based on the gas exchange measurements, wherein at least one of the contribution values is based on the gas exchange data and each of the plurality of contribution values corresponds to a strength of evidence that the associated physiological condition contributes to dyspnea in the patient; and   outputting the plurality of contribution values to indicate the strength of evidence of the associated physiological conditions.   
     
     
         2 . The method of  claim 1 , wherein the gas exchange test includes an exercise phase during which the patient performs exercise and a rest phase that precedes the exercise phase and at least one contribution value of the plurality of contribution values is determined based on gas exchange measurements captured on the breath-by-breath basis during the rest phase and the exercise phase. 
     
     
         3 . The method of  claim 1 , wherein the gas exchange test includes an exercise phase during which the patient performs exercise and a recovery phase that follows the exercise phase and at least one contribution value of the plurality of contribution values is determined based on gas exchange measurements captured on the breath-by-breath basis during the recovery phase and the exercise phase. 
     
     
         4 . The method of  claim 1 , wherein the receiving gas exchange measurements includes receiving gas exchange measurements based on breath-by-breath data captured by a flow sensor and an analyzer. 
     
     
         5 . The method of  claim 1 , further comprising repeatedly evaluating a respiratory exchange rate based on the received gas exchange measurements to determine when a predetermined target metric corresponding to exertion level is reached. 
     
     
         6 . The method of  claim 1 , further comprising receiving spirometric measurements including at least one of forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and inspiratory capacity (IC); and determining an FEV1/FVC ratio and determining a contribution value associated with obstructed or restrictive lung disease based on the FEV1/FVC ratio. 
     
     
         7 . The method of  claim 1 , wherein the physiological conditions associated with the plurality of contribution values are selected from a group of physiological conditions consisting of heart failure, pulmonary arterial hypertension, pulmonary venous hypertension, obstructed lung, restricted lung, obesity, general deconditioning, and acute decompensated heart failure. 
     
     
         8 . The method of  claim 1 , further comprising generating a user interface that includes a first visual indicator and a second visual indicator, wherein a property of the first visual indicator is determined based on a first contribution value of the plurality of contribution values and a property of the second visual indicator is based on a second contribution value of the plurality of contribution values, wherein the properties of the first and second visual indicators are selected from a group of properties consisting of:
 height; and   color.   
     
     
         9 . The method of  claim 8 , wherein the user interface further includes at least one historical visual indicator, wherein a property of the at least one historical visual indicator is determined based on a historical contribution value from a historical gas exchange test of the patient. 
     
     
         10 . The method of  claim 1 , wherein at least one contribution value of the plurality of contribution values is determined based on at least one physiological parameter selected from a group of physiological parameters consisting of age, gender, and Body Mass Index (BMI). 
     
     
         11 . The method of  claim 1 , wherein the machine learning classifier includes a Bayesian classifier or a support vector machine. 
     
     
         12 . A method comprising:
 performing, using a cardiopulmonary exercise gas exchange analyzer, a gas exchange test on a patient;   obtaining a plurality of gas exchange measurements, including end tidal CO2, on a breath-by-breath basis during the gas exchange test;   determining, using a machine learning classifier, a first contribution value associated with a first physiological condition, wherein the first contribution value is based on at least one of the plurality of gas exchange measurements;   determining a second contribution value associated with a second physiological condition, wherein the second contribution value is based on at least one of the plurality of gas exchange measurements; and   causing the first contribution value and the second contribution value to be displayed.   
     
     
         13 . The method of  claim 12 , wherein causing the first contribution value and the second contribution value to be displayed includes generating a user interface having a first visual indicator and a second visual indicator, wherein a property of the first visual indicator is determined based on the first contribution value and a property of the second visual indicator is determined based on the second contribution value. 
     
     
         14 . The method of  claim 13 , further comprising retrieving historical test data relating to a previous gas exchange test on the patient; and wherein the user interface includes a first historical visual indicator and a second historical visual indicator, wherein a property of the first historical visual indicator is determined based on a first historical contribution value and a property of the second historical visual indicator is determined based on a second historical contribution value, the first historical contribution value being associated with the first physiological condition and being based on the historical test data, and the second historical contribution value being associated with the second physiological condition and being based on the historical test data. 
     
     
         15 . The method of  claim 12 , further comprising:
 receiving results of a spirometry test on the patient, wherein the spirometry test includes a flow-volume loop; and   obtaining a plurality of spirometric measurements, including at least one of forced vital capacity (FVC), inspiratory capacity (IC), and forced expiratory volume in one second (FEV1);   wherein the determining the second contribution value includes determining the second contribution value based on at least one of the plurality of spirometric measurements and the at least one of the plurality of gas exchange measurements.   
     
     
         16 . The method of  claim 12 , wherein determining a first contribution value includes determining the first contribution value based on at least one of the plurality of gas exchange measurements and at least one physiological parameter selected from a group of physiological parameters consisting of age, gender, and Body Mass Index (BMI). 
     
     
         17 . The method of  claim 12 , wherein the machine learning classifier includes a Bayesian classifier or a support vector machine. 
     
     
         18 . A system comprising:
 a flow sensor configured to sense a respiratory flow of a patient;   an analyzer configured to determine a composition of at least a portion of the respiratory flow of the patient; and   a computing device configured to:   receive gas exchange measurements that are based on breath-by-breath data captured by the flow sensor and the analyzer during a gas exchange test on a patient;   determine, using a machine learning classifier, a plurality of contribution values that are each associated with a different physiological condition, wherein the plurality of contribution values are numeric values based on the gas exchange measurements, wherein at least one of the contribution values is based on the gas exchange data and each of the plurality of contribution values corresponds to a strength of evidence that the associated physiological condition contributes to dyspnea in the patient; and   output the plurality of contribution values to indicate the strength of evidence of the associated physiological conditions.   
     
     
         19 . The system of  claim 18 , wherein the computing device being configured to receive gas exchange measurements includes the computing device being configured to receive gas exchange measurements that are based on breath-by-breath data captured by the flow sensor and the analyzer during a submaximal exercise phase of the gas exchange test during which the patient performs submaximal exercise. 
     
     
         20 . The system of  claim 18 , wherein the computing device is configured to determine at least one contribution value of the plurality of contribution values based on at least one physiological parameter selected from a group of physiological parameters consisting of age, gender, and Body Mass Index (BMI).