BCIG Special Event: "Four Great Talks by Four Great Summer Students!”"
THE EVENT: Carl Leonard and Jim DeLeo in the NIH Clinical Center
Department of Clinical Research Informatics Scientific Computing Section have
been extremely honored to work with four great summer student interns this
summer. They would like to give you an opportunity to meet these students and
experience first hand the excellent work that they have done. As you will see
their work clearly promote the Scientific Computing Section’s mission which is
to advance the application of modern scientific computing methodology in basic
biomedical research and clinical medicine with particular attention given to
providing tools to more rapidly meet the objectives of translational medicine,
namely to move knowledge from basic research to clinical practice more quickly
and efficiently.
Who Should Attend? Anyone interested in content of these talks and in
meeting the students and helping them to get ready for Poster Day which is the
day after this very special BCIG event.
Directions: (1) Enter the north entrance of the NIH Clinical Center, (2)
stay to the left and pass the coffee shop, (3) proceed to the “I East Lab
Corridor” on the left, (4) go all the way down the corridor to the four
elevators, (5) take an elevator to the 2nd floor, (6) exit elevator and walk to
the hallway, (7) turn left, (8) take a few steps, (9) Room 2-3330 is on the
right.
BCIG WEB SITE: www.nih-bcig.org
NIH
CONTACTS:
Carl Leonard, 301-496-0191,
cleonard@lired.com
Jim DeLeo, 301-496-3858, jdeleo@nih.gov |
3:00 pm - 4:00 pm August 6, 2008
TITLE: “SMART-MARTTM
- a Desktop Tool for Clinical Data Mining”
PRESENTER: James Woo, Johns Hopkins
University
OBJECTIVE:
To build a computational module in Smart-MartTM that imports
different formatted data files, e.g. Excel spread sheets, database views, and
tabular text files into a standardized format compatible with Smart-MartTM
application programs.
----------------------------
TITLE: “Facilitating Data
Analysis through Ruby & R Interaction”
PRESENTER: Rishi Gharpuray, Stanford
University
OBJECTIVE:
To create and develop a software graphical module that uses Ruby code in Smart-MartTM,
a software environment presently under development by the NIHCC Scientific
Computing Section. Plots produced with this module should facilitate data
analysis for biomedical research and promote the objectives of data mining and
translational medicine.
----------------------------
TITLE: “Understanding
Behavior and Brain Activity Using fMRI Imaging”
PRESEBNTER: Simone Campbell, Gaithersburg
High School
OBJECTIVE:
To study achievements in cognitive neuroscience that have led to better
understanding of how mental representations map into neural activity patterns
and to suggest that new advancements in computer science may be useful in
furthering this work, particularly with regard to bridging the gap between basic
research and patient care - the main goal of translational medicine.
Based on their work together in the summer of 2008, Morgan Clinton
and Jim DeLeo have submitted a paper to the Journal of Movement Disorders
The title of this paper is
Corticobasal Syndrome Patient
Subtyping with HubuHTM – a New Clustering Algorithm
ABSTRACT
Corticobasal Syndrome (CBS) is a class of neurodegenerative
disorders that includes several major subtypes. A computerized clinical
decision support system that could provide improved subtyping of this disorder
could potentially improve diagnoses and outcome for CBS patients. We developed
HubuHTM, a new computer clustering algorithm that appears useful for
this purpose. We applied HubuHTM to a clinical data set consisting
mostly of clinical psychology test scores related to 107 CBS patients and show
some early findings about CBS that the algorithm has revealed. We discuss how
we plan to enhance this data set and we explain how we will extend the basic
HubuHTM algorithm for use in a more generalized data mining mode in
which it will automatically generate reports containing significant findings.
We plan to develop and apply HubuHTM in the expanded data mining
mode to the expanded CBS database as well as to other kinds of clinical data.
Our longer range goal is to use HubuHTM to develop practical
diagnostic and treatment selection comparative effectiveness support tools that
have practical value in improving patient outcome.
Key Words
Neurodegenerative disorders, clustering algorithm, data mining,
neuropathology, corticobasal syndrome
If you would
like more information about this paper and particularly about the operational
and availability of the HubuHTM algorithm, please contact Jim DeLeo,
jdeleo@nih.gov, 301-496-3848.
----------------------------
TITLE: “Artificial Neural
Network Subgrouping of Corticobasal Syndrome Patients”
PRESENTER: Morgan Clinton, University of
South Carolina
OBJECTIVE:
To explore using artificial neural network methodology to differentiate
corticobasal syndrome (CBS) patient subgroups with a longer range goal of
developing practical diagnostic and treatment selection comparative
effectiveness tools that would improve patient outcome. |
