Development and use of a medication history service associated with a health information exchange: architecture and preliminary findings.

We describe our early experience with use in emergency department settings of a standards-based medication history service integrated into a health information exchange (HIE). The service sends queries from one Exchange's emergency department interface both to a local ambulatory care system and to the medication hub services provided by a second HIE. This second HIE in turn sends requests to SureScripts and returns histories for incorporation into the first Exchange's clinical interface. The service caches all requests to avoid costly duplicate query charges and maintains an account of queries, registered users, charges, and results obtained. Usage may be increasing as additional retail pharmacy data become available. Early results suggest that research and development emphasis requirements will of necessity shift from obtaining prescription medication history to finding new means to ensuring effective use.

A regional health information exchange: architecture and implementation.

The MidSouth eHealth Alliances health information exchange in Memphis, Tennessee provides access to data on almost 1 million individuals. The effort is the product of a comprehensive, integrated approach to technology and policy that emphasizes patient-centered use, low-cost, flexibility, and rigorous privacy and confidentiality policies and practices It is used in emergency departments and other major clinical settings. This paper provides a high-level overview of the system and its use. The early anecdotal success of this effort and preliminary formal clinical and financial evaluation suggest that health information exchanges can improve care at relatively low cost.

The MidSouth eHealth Alliance: use and impact in the first year.

The MidSouth e-Health Alliance is a health information exchange that has been in use in the Memphis, Tennessee region since May, 2006. This health information exchange took two years to develop from the time it was initially conceived. Following on the work done by the Indianapolis project, the MidSouth e-Health Alliance focused initially on implementations in emergency departments throughout this region. A total of 321 clinicians have used the system in the 5 emergency departments since s initial deployment. This paper reports on the processes users are engaged in to use the system as well as the demographics and patient characteristics associated with system use to date.

Data mining for gene networks relevant to poor prognosis in lung cancer via backward-chaining rule induction.

We use Backward Chaining Rule Induction (BCRI), a novel data mining method for hypothesizing causative mechanisms, to mine lung cancer gene expression array data for mechanisms that could impact survival. Initially, a supervised learning system is used to generate a prediction model in the form of "IF THEN " style rules. Next, each antecedent (i.e. an IF condition) of a previously discovered rule becomes the outcome class for subsequent application of supervised rule induction. This step is repeated until a termination condition is satisfied. "Chains" of rules are created by working backward from an initial condition (e.g. survival status). Through this iterative process of "backward chaining," BCRI searches for rules that describe plausible gene interactions for subsequent validation. Thus, BCRI is a semi-supervised approach that constrains the search through the vast space of plausible causal mechanisms by using a top-level outcome to kick-start the process. We demonstrate the general BCRI task sequence, how to implement it, the validation process, and how BCRI-rules discovered from lung cancer microarray data can be combined with prior knowledge to generate hypotheses about functional genomics.

Using prior knowledge and rule induction methods to discover molecular markers of prognosis in lung cancer.

An iterative computational scientific discovery approach is proposed and applied to gene expression data for resectable lung adenocarcinoma patients. We use genes learned from the C5.0 rule induction algorithm, clinical features and prior knowledge derived from a network of interacting genes as represented in a database obtained with PathwayAssist to discover markers for prognosis in the gene expression data. This is done in an iterative fashion with machine learning techniques seeding the prior knowledge. This research illustrates the utility of combining signaling networks and machine learning techniques to produce simple prognostic classifiers.