Method of diagnosing biological states through the use of a centralized, adaptive model, and remote sample processing

• US 7,761,239 B2
• Filed: 12/10/2004
• Issued: 07/20/2010
• Est. Priority Date: 12/11/2003
• Status: Active Grant

• Current Assignee: Aspira Women's Health, Inc.
• Original Assignee: Correlogic Systems, Inc. (Aspira Women's Health, Inc.)
• Inventors: Chen, Tzong-Hao, Levine, Peter J., Hitt, Ben A.
• Primary Examiner(s): Clow; Lori A
• Application Number:
  US11/008,784
• Publication Number:
  US 20050209786A1
• Time in Patent Office: 2,048 Days
• Field of Search: None
• US Class Current: 702/19
• CPC Class Codes: G06F 18/217 Validation; Performance evaluation; Active pattern learning techniques
• CPC Class Codes: G06F 18/2321 using statistics or function optimisation, e.g. modelling of probability density functions
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
• CPC Class Codes: G16B 40/10 Signal processing, e.g. from mass spectrometry [MS] or from PCR
• CPC Class Codes: G16B 50/00 ICT programming tools or database systems specially adapted for bioinformatics
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: Y10T 436/24 Nuclear magnetic resonance, electron spin resonance or other spin effects or mass spectrometry

Abstract

A model of a particular biological state can be developed. The model may be used to determine if an unknown biological sample exhibits a particular biological state. This can be done by receiving either a biological sample or data associated with the biological sample. After the data is received, the data may be input into the model. In one embodiment, the acquisition of the data associated with the biological sample is performed at a first location and the imputing of the data into the model is performed at a second location different than the first location. Unless the data maps identically to the model, the data would have an inherent effect on the position of the particular clusters within the discriminatory pattern, if it is allowed to affect the model. The modeling software can keep track of the net effect on the model that each sample received has on the position of the model. If the model has drifted outside of a predetermined tolerance, the model can be updated. Various business relationships may be developed to undertake various steps of the overall method for providing a diagnosis to a patient.

Claims

1. A computer implemented method of determining whether a diagnostic model is accurately applicable to a population of subjects to which the model is applied, the model being configured to determine a biological state of a subject, the model being based on a set of data streams, each of the data streams being obtained by performing an analysis of biological samples taken from subjects of known biological states, the model having at least one diagnostic cluster located in a vector space having at least three dimensions, each dimension corresponding to a vector common to the data streams in the set of data streams, the diagnostic cluster having a centroid located at an initial centroid location based on locations of vectors from the set of data streams that correspond to the diagnostic cluster, comprising:

receiving a vector set, the vector set including at least three vectors from a data stream obtained by performing the analysis of a biological sample taken from a subject of an unknown biological state;

mapping the unknown vector set into the vector space using a suitably programmed computer;

if the unknown vector set maps into the diagnostic cluster, calculating an updated location of the cluster centroid based on a combination of the location of the vectors from the unknown vector set and the set of data streams that correspond to the diagnostic cluster;

determining the distance between the initial centroid location and the updated centroid location; and

if the distance between the initial centroid location and the updated centroid location is greater than a predetermined threshold, providing an output indicating that the threshold has been exceeded which is indicative that the diagnostic model is not accurately applicable to the population of subjects to which the model is applied.

2. The method of claim 1, further comprising;

if the distance between the initial centroid location and the updated centroid location is greater than the predetermined threshold, creating a new diagnostic model based on a combination of the unknown vector set and the data streams that correspond to the biological samples of known biological states from the population of subjects to which the model is applied.

3. The method of claim 1, wherein the output is a visual output.

4. The method of claim 1, the output being a first output, further comprising:

if the distance between the initial centroid location and the updated centroid location is less than the threshold, providing a second output indicating that the threshold has not been exceeded which is indicative that the diagnostic model is accurately applicable to the population of subjects to which the model is applied.

5. A method of determining whether a diagnostic model is sufficiently representative of a first population of subjects to which the model is applied, the model being configured to determine a biological state of a subject from the first population of subjects, the model being based on a set of data streams, each of the data streams being obtained by performing an analysis of biological samples taken from a second population of subjects, the second population of subjects being different than the first population of subjects, the second population of subjects being of known biological states, the model having at least one diagnostic cluster located in a vector space, the diagnostic cluster having an initial centroid located at an initial centroid location based the set of data streams obtained by performing the analysis of the biological samples taken from the second population of subjects, comprising:

receiving a data stream obtained by performing the analysis of a biological sample taken from a first subject of the first population, the first subject of the first population being of an unknown biological state;

receiving a data stream obtained by performing the analysis of a biological sample taken from a second subject of the first population, the second subject of the first population being of an unknown biological state;

mapping the first data stream and the second data stream into the vector space using a suitably programmed computer;

if the first data stream and the second data stream map into the diagnostic cluster, calculating an updated location of the cluster centroid based on a combination of the first data stream, and second data stream, and the set of data streams obtained by performing the analysis of the biological samples taken from the second population of subjects;

determining the distance between the initial centroid location and the updated centroid location; and

if the distance between the initial centroid location and the updated centroid location is greater than a predetermined threshold, providing an output indicating that the threshold has been exceeded which is indicative that the diagnostic model is not sufficiently representative of the first population of subjects.

6. The method of claim 5, wherein the output is a visual output.

7. The method of claim 5, further comprising a step in which the determining is performed in a self monitoring process.