发明名称 Associating records in healthcare databases with individuals
摘要 At least one attribute type associated with an individual in a first record of a first database and one or more similar attribute types in record(s) of other second database(s) are located. The attribute types compared. Based on the comparison, a first weighted score for the first record and another weighted score for each other records are computed. The weighted score indicates a likelihood the particular attribute type is associated with a same individual as other attribute type(s) located in other record(s) and further accounts for a time delta between measurements of the attribute types. Confidence score outputs for the records are also computed. The confidence score output of a record is based on all weighted scores of that record and indicates a likelihood that record is associated with the same individual. At least one database is updated based on at least one of the confidence score outputs.
申请公布号 US9230060(B2) 申请公布日期 2016.01.05
申请号 US201414507393 申请日期 2014.10.06
申请人 International Business Machines Corporation 发明人 Friedlander Robert R.;Khan Anwer M.
分类号 G06F7/00;G06F17/30;G06F19/00;G06Q10/00;G06Q50/24 主分类号 G06F7/00
代理机构 Russell Ng PLLC 代理人 Russell Ng PLLC ;Pivnichny John
主权项 1. A method of associating healthcare records in healthcare databases with an individual, the method comprising: locating at least one attribute type in a first record of a first database and one or more similar attribute types in one or more other database records of one or more second databases, the at least one attribute type being associated with the individual; comparing the at least one attribute type and each of the one or more similar attribute types; in response to comparing the at least one attribute type and each of the one or more similar attribute types, determining, based on the comparison, a first weighted score for each one of the at least one attribute type in the first record and another weighted score for each one of the one or more similar attribute types in the one or more other records, wherein a weighted score for a particular attribute type indicates a likelihood that the particular attribute type is associated with a same individual as at least one other attribute type located in another record, and wherein a weighted score for a particular attribute type further accounts for a time delta between measurements of the particular attribute type in a particular record and measurements of one or more similar attribute types in one or more other records; calculating a first confidence score output for the first record and one or more other confidence score outputs for each of the one or more other records, wherein a confidence score output of a record is based on all weighted scores of that record and indicates a likelihood that record is associated with the same individual; and updating at least one of the first and second databases based on at least one of the first confidence score output and the one or more other confidence score outputs.
地址 Armonk NY US