Methods for identifying human hereditary disease patterns

• US 5,642,936 A
• Filed: 01/29/1996
• Issued: 07/01/1997
• Est. Priority Date: 01/29/1996
• Status: Expired due to Fees

• Current Assignee: Oncormed, Inc. (Steel Partners Holdings LP)
• Original Assignee: Oncormed, Inc. (Steel Partners Holdings LP)
• Inventors: Evans, Steven
• Primary Examiner(s): Kamm, William E.
• Assistant Examiner(s): EVANISKO, GEORGE ROBERT
• Application Number:
  US08/593,446
• Time in Patent Office: 519 Days
• Field of Search: 364/413.01, 364/413.02, 128/630, 128/897, 128/898, 128/906
• US Class Current: 600/300
• CPC Class Codes: G16B 20/00 ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: Y10S 128/924 using artificial intelligence

Abstract

The present invention relates to a method of determining human hereditary disease risk factors, a method of determining hereditary cancer patterns presenting in cancer family histories, and a method of determining whether a cancer family history represents a hereditary pattern.

Claims

1. A method for determining the existence of a hereditary disease risk in a patient, comprising the steps of:

assembling a computer a database made up of a plurality of records each pertaining to an individual and containing a history of at least one specific disease in a family of that individual, said history including a plurality of parameters relating to each family member identified in said history;

defining a plurality of functions each pertaining to said parameters and assigning predetermined weights to said functions based on values of said parameters;

for each record in said database, summing the weights obtained for each of said functions to obtain a total value for each of said functions, identifying said record as presenting a hereditary pattern if the total value is above a predetermined threshold, and grouping said identified record into a subset of records;

for each record in said database, applying expert knowledge generated rules to independently identify records as presenting hereditary patterns;

comparing said independently identified records with said subset of records, and validating defined functions if a predetermined minimum percentage of records in said subset are consistent with said independently identified records; and

using and displaying said validated functions as a recognizer of hereditary disease patterns in a family history of said patient.

2. A method according to claim 1, wherein said plurality of records contain histories of instances of cancer.

3. A method according to claim 2, wherein said instances of cancer include breast, ovarian, endometrial, prostate, malignant melanoma and colon cancer.

4. A method according to claim 3, further including the step of assigning weights to particular attributes used in said functions, and defining attributes as significant in the definition of specific hereditary disease patterns if a minimum percentage of records in said subset are consistent with said independently identified records.

5. A method according to claim 1, further comprising creating an additional record in said database for each relative of said individual who is identified as having had said disease.