Discrete bayesian analysis of data
OwnerCarlos E. Padilla, Valeri I. Karlovinventors
Life of this patent
Estimating the term…
Abstract
A probabilistic approximation of a data distribution is provided, wherein uncertain measurements in data are fused together to provide an indication of whether a new data item belongs to a given disease model. The probabilistic approximation is provided in accordance with a Bayesian analysis technique that examines the relationship of probability distributions for observable events x and multiple hypotheses H regarding those events.
Claims38
A method of determining clinically relevant information from gene expression data, comprising:
- conducting a statistical analysis of the gene expression data to identify trends and dependencies among the data;
- and
- deriving a probabilistic model from the gene expression data, the probabilistic model being indicative of a probable classification of the data into clinically relevant information.
A method for generating an a posteriori tree of clinically relevant information for a subject, wherein the method comprises:
- performing an analysis of gene expression data for a population of individuals, wherein the data comprises a matrix of discriminations between clinically relevant information selected from a predetermined list of clinically relevant information;
- performing a Bayesian statistical analysis to estimate a series of hypothesis-conditional probability density functions p(x|Hi) where a hypothesis Hi is one of a set H of the clinically relevant information;
- determining a prior probability density function p(Hi) for each of the hypotheses Hi;
- determining a posterior test-conditional probability density function p(Hi|x) for each of the hypotheses Hi test data records;
- and
- generating a posterior tree of possible clinically relevant information for a test subject in accordance with test results for the test subject.
A method of developing a test to screen for one or more inapparent diseases, comprising:
- conducting a statistical analysis of data in order to identify trends and dependencies among the data, wherein the data comprises gene expression data from a subject;
- deriving a probabilistic model from the data, the probabilistic model being indicative of a probable disease diagnosis for a patient, wherein the probabilistic model is derived using a discrete Bayesian analysis;
- identifying from among the input data, the data that contributes to the diagnosis;
- and
- identifying the genes that generated the data that contributes to the diagnosis.
A method of diagnosing a disease condition of a patient, the method comprising:
- receiving a set of gene expression data comprising gene expression data from a population X of individuals;
- estimating a hypothesis-conditional probability density function p(x|H 1) where the hypothesis H1 relates to a diagnosis condition for a test patient x, and estimating a hypothesis-conditional probability density function p(x|H2) where the hypothesis H2 relates to a non-diagnosis condition for a test patient;
- determining a prior probability density function p(H) for the each of the hypotheses H1 and H2;
- determining a posterior test-conditional probability density function p(H|x) for each of the hypotheses H1 and H2 on the test data x;
- and
- providing a diagnosis probability of a new patient for the H disease condition, based on the determined posterior test-conditional probability density function p(H1|x) as compared to the posterior test-conditional probability density function p(H2|x) and gene expression data of the new patient.
A method of diagnosing a disease from data, comprising:
- conducting a statistical analysis of the data in order to identify trends and dependencies among the data, wherein the data comprises gene expression data from a subject;
- deriving a probabilistic model from the data, the probabilistic model being indicative of a probable disease diagnosis for a patient, wherein the disease is an inapparent disease.
Dependencies are inferred from the claim text.