Systems and methods for improving disease diagnosis

Abstract

The present invention relates to systems and methods for improving the accuracy of disease diagnosis and to associated diagnostic tests involving the correlation of measured analytes with binary outcomes (e.g., not-disease or disease), as well as higher-order outcomes (e.g., one of several phases of a disease). Methods of the present invention use biomarker sets, preferably those with orthogonal functionality, to obtain concentration and proximity score values for disease and non-disease states. The biomarker a set's proximity scores are graphed on an orthogonal grid, with one dimension for each biomarker. The proximity scores and orthogonal gridding is then used to calculate a disease state or non-disease state diagnosis for the patient.

1 . A computer-implemented method for determining a probability of a disease state for a disease in a patient under examination, the method comprising:
 receiving a first set of concentration values of a first analyte from a first set of samples from patients with a not-disease diagnosis for the disease;   receiving a second set of concentration values of the first analyte from a second set of samples from patients with a disease diagnosis for the disease, wherein the first set and second set of samples comprise a training set of samples;   calculating a mean value of concentration of the first analyte from the first set of concentration values;   calculating a mean value of concentration of the first analyte from the second set of concentration values;   calculating a first proximity score representing the mean value of concentration of the first set of analytes;   calculating a second proximity score representing the mean value of concentration of the second set of analytes; and   applying a machine learning algorithm to map the concentrations of the training set of samples into a range of proximity scores between the first proximity score and the second proximity score to provide an assessment of the probability of the disease state of a patient under examination.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein the first and second analytes are low abundance proteins. 
     
     
         3 . The computer-implemented method of  claim 1 , wherein the concentration values of the first and second analytes are correlated to one or more population distribution characteristics. 
     
     
         4 . The computer-implemented method of  claim 1 , further comprising performing an adjustment to the evaluative model at the neural network to compensate for topology instability. 
     
     
         5 . The computer-implemented method of  claim 1 , further comprising performing a support vector machine regression analysis at the neural network to complete the evaluative model. 
     
     
         6 . The computer-implemented method of  claim 1 , wherein the training set of samples includes at least one of blood samples, urine samples, and tissue samples. 
     
     
         7 . The computer-implemented method of  claim 1 , wherein the training set of samples includes an equal number of disease samples and not-disease samples. 
     
     
         8 . The computer-implemented method of  claim 1 , wherein the neural network adjusts the concentrations of the first and second analytes to reduce proteomic variance using compression, expansion, inversion, reversal, or folding functions. 
     
     
         9 . The computer-implemented method of  claim 1 , wherein the disease is a metabolic disease. 
     
     
         10 . The computer-implemented method of  claim 1 , wherein the disease is one of Alzheimer's disease, macular degeneration, rheumatoid arthritis, multiple sclerosis, Parkinson's disease, an auto-immune disease, or cardiomyopathy. 
     
     
         11 . A computer system for determining a probability of a disease state for a disease in a patient under examination comprising a server that:
 receives a first set of concentration values of a first analyte from a first set of samples from patients with a not-disease diagnosis for the disease;   receives a second set of concentration values of the first analyte from a second set of samples from patients with a disease diagnosis for the disease, wherein the first set and second set of samples comprise a training set of samples;   calculates a mean value of concentration of the first analyte from the first set of concentration values;   calculates a mean value of concentration of the first analyte from the second set of concentration values;   calculates a first proximity score representing the mean value of concentration of the first set of analytes;   calculates a second proximity score representing the mean value of concentration of the second set of analytes; and   applies a machine learning algorithm to map the concentrations of the training set of samples into a range of proximity scores between the first proximity score and the second proximity score to provide an assessment of the probability of the disease state of a patient under examination.   
     
     
         12 . The computer system of  claim 11 , wherein the first and second analytes are low abundance proteins. 
     
     
         13 . The computer system of  claim 11 , wherein the concentration values of the first and second analytes are correlated to one or more population distribution characteristics. 
     
     
         14 . The computer-implemented method of  claim 1 , wherein the neural network performs an adjustment to the evaluative model to compensate for topology instability. 
     
     
         15 . The computer system of  claim 11 , wherein the neural network performs a support vector machine regression analysis to complete the evaluative model. 
     
     
         16 . The computer system of  claim 11 , wherein the training set of samples includes at least one of blood samples, urine samples, and tissue samples. 
     
     
         17 . The computer system of  claim 11 , wherein the training set of samples includes an equal number of disease samples and not-disease samples. 
     
     
         18 . The computer system of  claim 11 , wherein the neural network adjusts the concentrations of the first and second analytes to reduce proteomic variance using compression, expansion, inversion, reversal, or folding functions. 
     
     
         19 . The computer system of  claim 11 , wherein the disease is a metabolic disease. 
     
     
         20 . The computer system of  claim 11 , wherein the disease is one of Alzheimer's disease, macular degeneration, rheumatoid arthritis, multiple sclerosis, Parkinson's disease, an auto-immune disease, or cardiomyopathy.