Linking Health Records for Federated Query Processing

Abstract

A federated query portal in an electronic health record infrastructure enables large epidemiology studies by combining data from geographically dispersed medical institutions. However, an individual's health record has been found to be distributed across multiple carrier databases in local settings. Privacy regulations may prohibit a data source from revealing clear text identifiers, thereby making it non-trivial for a query aggregator to determine which records correspond to the same underlying individual. In this paper, we explore this problem of privately detecting and tracking the health records of an individual in a distributed infrastructure. We begin with a secure set intersection protocol based on commutative encryption, and show how to make it practical on comparison spaces as large as 10 10 pairs. Using bigram matching, precomputed tables, and data parallelism, we successfully reduced the execution time to a matter of minutes, while retaining a high degree of accuracy even in records with data entry errors. We also propose techniques to prevent the inference of identifier information when knowledge of underlying data distributions is known to an adversary. Finally, we discuss how records can be tracked utilizing the detection results during query processing.

Research and realisation of similar information high precision purification and mining technology

“…The experiment dataset, similar information purification and mining adopts the proposed method in the paper based on time series updating method, FCA method (in Huang et al, 2015) and multiple purification method(in Cohen et al, 2015), scan purification and mining results with the period of 0.02s, 0.04s, 0.06s. F * value in different periods should be equal in principle, but there are bias in actual calculation, so the average of three cycles is taken as F * value's final result (Dewri et al, 2016;Sethi et al, 2015), Tables 3, 4 and 5 recall F * value of the three purification and mining methods in different datasets. The proposed method gains the highest F * value and the mining performance is the best.…”

A graph traversal attack on Bloom filter-based medical data aggregation

“…Commutative encryption appears to hold some promise for solving PPRL. Some recent results [2] suggest that, what always was deemed to be computationaly too expensive, efficient implementation are indeed possible for realistic size data sets.…”

Towards a New Paradigm of Federated Electronic Health Records in Palestine

“…It can also provide information for drugs discovery and clinical good practices and research, as well as provide information for policy makers and others who need access to national health indicators and simulations. An example by Dewri Rinku et al explains: "A federated query portal in an electronic health record infrastructure enables large epidemiology studies by combining data from geographically dispersed medical institutions" [12]. The FEHR may also accommodate travelers and tourists' healthcare services.…”