| US 7,457,731 B2 | ||
| Early detection of disease outbreak using electronic patient data to reduce public health threat from bio-terrorism | ||
| R. Bharat Rao, Berwyn, Pa. (US) | ||
| Assigned to Siemens Medical Solutions USA, Inc., Malvern, Pa. (US) | ||
| Filed on Dec. 13, 2002, as Appl. No. 10/319,365. | ||
| Claims priority of provisional application 60/340634, filed on Dec. 14, 2001. | ||
| Prior Publication US 2003/0177038 A1, Sep. 18, 2003 | ||
| Int. Cl. G06F 17/10 (2006.01) | ||
| U.S. Cl. 703—2 [705/7; 705/2; 705/3; 600/300; 707/100; 707/104.1; 707/9] | 53 Claims |
| 1. A method for automated identification of disease outbreak, the method comprising:
obtaining patient records for patients being treated, at least some of the patient records containing information derived
with a processor by data mining from individual data collections of the patients being treated, the information derived with
the processor from an unstructured data source including text format, image information, waveform information or combinations
thereof, the patient records being a machine readable structured dataset including the information derived with the processor
from the unstructured data source and information from a structured data source;
correlating selected patient data from the structured dataset, including data from the unstructured data source, contained
in the patient records with disease indicia for at least one disease;
identifying an anomaly at least in part based on the correlations;
outputting an alert about the anomaly;
extracting information from a plurality of clinical data sources to create a set of probabilistic assertions;
combining the set of probabilistic assertions to create one or more unified probabilistic assertions; and
inferring patient symptoms from the one or more unified probabilistic assertions.
|