Self-improving classification system
AssigneeMedtronic, Inc.
InventorOrhan Soykan
Life of this patent
Estimating the term…
Abstract
A self-improving classification system classifies specimens based on class identifiers. The system stores specimen profiles in a database that is updated with additional specimen profiles and with follow-up data that corrects classification of specimens that were initially incorrectly classified. Algorithms use the updated database to discover new class identifiers, modify thresholds of known class identifiers, and drop unnecessary class identifiers to improve classification of specimens.
Claims13
A self improving method of classifying a specimen from a patient at possible risk of ventricular tachyarrhythmia/ventricular fibrillation (VT/VF), the method comprising:
- using an algorithm, by a computer, for classifying specimens based on at least one class identifier, wherein said class identifier includes a Single Nucleotide Polymorphism (SNP), the SNP selected from the group consisting of SEQ ID NO.'s 1-83 and 102;
- generating a specimen profile from the specimen by analyzing the genomic information of the specimen and storing, by a computer, the specimen profile in a database;
- classifying, based on the specimen profile, the specimen into a class of being from a patient at possible risk of VT/VF;
- the classifying being carried out using the algorithm;
- determining if reclassification of the specimen is necessary;
- refining the algorithm based on a refinement of the at least one class identifier, and
- reclassifying the specimen if necessary, by a computer.
Dependencies are inferred from the claim text.