×

Annotating medical data represented by characteristic functions

  • US 9,138,143 B2
  • Filed: 08/17/2010
  • Issued: 09/22/2015
  • Est. Priority Date: 08/17/2010
  • Status: Active Grant
First Claim
Patent Images

1. A method for annotating large volumes of medical data represented by binary decision diagrams comprising, by one or more processors associated with one or more computer systems:

  • accessing a plurality of set of samples of sensor data recording a plurality of medical measurements taken by one or more medical sensors, wherein each set of the one or more sets of samples corresponds to an annotation used to categorize the sensor data;

    representing each data value of the set of samples as a minterm to yield a set of minterms, each minterm within the set of minterms comprising a logical expression of one or more variables allocated to the data value of the sample;

    generating a characteristic function from the set of minterms, the characteristic function being represented by a reduced ordered binary decision diagram indicating whether a given minterm is a member of the set of minterms, the reduced ordered binary decision diagram being stored compactly in a BDD library;

    receiving a search query for a search of a set of samples of sensor data wherein the query indicates one or more requested values of one or more sensor parameters, the search query being represented by a query function corresponding to the requested values;

    identifying one or more minterms of the set of minterms that are associated with a given medical annotation by combining the query function and the characteristic function through logical operations, the identified minterms counting the number of instances where the data value represented by the minterm of the set of minterms is equal to the one or more requested values; and

    reporting the results of said identifying as search results, through an interface.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×