Protempa ${project.version} -- A Temporal Query Framework for Clinical Databases
Department of Biomedical Informatics, Emory University

OVERVIEW

Welcome to Protempa! It is a software framework that supports the construction 
of software systems for querying clinical databases, summarizing patient data, 
detecting clinically significant data patterns in groups of patients or
individual patients, and extract, transform and load (ETL). Its
special feature, temporal abstraction, supports queries and
summarization of data in terms of temporal patterns.  Temporal
patterns enable robust querying and extraction of clinical history.
These patterns include frequencies (at least two, three, etc. of some
event or observation); sequences (chemotherapy followed by surgery
within 180 days, a hospital encounter followed by another hospital
encounter within 30 days); and overlaps (falling platelet counts in
the setting of HELLP syndrome).  Protempa is the foundation for the
Eureka! Clinical Analytics tool, available from
http://aiw.sourceforge.net.

WHAT'S NEW

Version 2.3 now requires Java 7 compatibility. Ability to search for 
propositions from knowledge sources was added.  A new module for using 
BioPortal as a knowledge source was added.

Version 2.2 eliminates the need to run separate queries to hook up references
between propositions, significantly reducing the number of quries generated
and execution time.

Version 2.0.3 includes the following new features and changes:
* We added a new abstraction type called compound low-level abstraction. This
lets you infer abstract parameters from one or more other low-level
abstractions or compound low-level abstractions.
* We made toString()'s behavior more uniform across the framework. We use the
Apache ToStringBuilder and friends in most places.
* We moved the packages to our new public Maven repository (see below).
* Various other bug fixes.

Note that Protempa's public APIs are subject to change. A stable ABI will be 
part of a future release.

REQUIREMENTS

The current version is known to work with Oracle Java 7. It is expected to work
with OpenJDK 7 as well.

Protempa uses the Maven build system, available from
http://maven.apache.org.  You must install maven, if you do not
already have it, to work with the source code. We use Maven 3.

INSTALLATION

The easiest way to get Protempa is using Maven. It is available from
the http://maven.cci.emory.edu/nexus/content/repositories/aiw/ repository. 
Use the following dependency declaration:

<dependency>
    <groupId>org.protempa</groupId>
    <artifactId>protempa</artifactId> 
    <version>${project.version}</version>
</dependency>

The source code is also available in tarball and zip file formats from
http://aiw.sourceforge.net/dev/protempa-${project.version}.

Please see the technical documentation linked below for detailed
instructions on what the framework does and how to use it.

CONTACTING US

Please send an email to help@eurekaclinical.org for support or questions.

LICENSING

Protempa is freely available as open source under the Apache License,
Version 2.0. A copy of the license is in the LICENSE file provided in
this distribution.

Protempa is Copyright (C) 2012-2013 Emory University.

MORE INFORMATION

Please see http://aiw.sourceforge.net/dev/protempa-${project.version} for technical
documentation and more information.