Grid Architecture for Healthcare Preparedness Infosphere
Abijit Deshmukh, Ph.D., Mechanical and Industrial Engineering,
UMass Amherst
Richard Wait, M.D., Ph.D., Chairman, Department of Surgery, Baystate
Medical Center
The objective of this project is to evaluate the applicability
of the computational grid architecture as an integrative framework
for the Healthcare Preparedness Infosphere (HPI), which is being
developed by researchers at the Baystate Medical Center with the
cooperation and assistance of members of the Massachusetts Department
of Health, the Naval Undersea Warfare Center, FEMA, MEMA and SEPRI
(UMass).
The proposed project will develop a scaled model of the
HPI using Globus grid architecture, which will collect information
on resources (beds, bed uses, operating rooms, emergency facilities,
staff, etc) for a set of hospitals and medical centers affiliated
with Baystate Medical Center. The HPI information will be used
to distribute emergency victims as well as healthcare assets (e.g.
healthcare workers, supplies, respirators, etc.) among the available
facilities in an optimum manner and also under normal operating
conditions to improve overall efficiency of the hospitals.
Integrated
real-time information from different sources provides a powerful
resource for making optimal decisions in healthcare applications.
If implemented successfully, it will not only be of immense importance
in case of emergencies but also in regular day-to-day operations.
By creating a scaled down version of the concept, UMass Amherst
and Baystate Medical Center will be able to demonstrate the efficacy
of using grid architecture for integrating information in the HPI.
Effective Date- September 1, 2004
A New Approach to MR Imaging in Dementia
Edward Riseman, Ph.D., Computer Science, UMass Amherst
Joseph
Horowitz, Ph.D., Mathematics and Statistics, UMass Amherst
Benjamin
Liptzin, Ph.D., Chairman, Department of Psychology, Baystate
Medical Center
Neuroimaging techniques have been used in the diagnosis and differential
diagnosis of dementia to identify and localize brain tumors, subdural
hematomas, and major strokes, as well as to estimate the amount
of brain atrophy that is present. Regrettably, these techniques
have not provided adequate sensitivity or specificity for the diagnosis
of Alzheimer's Disease (AD) or related dementias. This issue has
become increasingly important as therapeutic research identifies
more specific targets for intervention and makes it imperative
that neuroimaging research focus on patterns which better define
AD and other dementias.
In recent years two approaches have been
used to improve the yield from neuroimaging studies. The first
is to assess atrophy in different regions of the brain, particularly
those thought to be most affected by AD, rather than to rely on
brain atrophy as a whole. The second approach has been to use serial
measures to detect the progression of regional atrophy over time.
The
proposed study will apply state-of-the-art medical imaging techniques
from the field of computer vision and mathematics. This pilot study
is intended to provide initial advances to the field by using computer
analysis of MR images to obtain volumetric measures of atrophy
in selected brain regions, and to support future studies to determine
serial measures of change over time. These methods initially are
intended to be semi- automated, hence will minimize the time required
by physicians or other personnel to evaluate volumetric information
in the images. Our pilot study is intended to provide initial advances
that are sufficient to form the foundation of future NIH proposals.
Effective Date- September 1, 2004
NHLH2 Gene Mutations in Obese Humans
Deborah Good, Ph.D., Vet and Animal Sciences, UMass Amherst
Holley
Allen, M.D., Chief, Pediatric Endocrinology, Baystate Medical Center
The obesity epidemic in both children and adults is of enormous
public health significance. While lifestyle issues are responsible
for weight gain, the risk for obesity and timing of obesity are,
at least partially, genetically programmed. Many genetic elements
play a role, with no single gene, to date, explaining a large proportion
or the risk.
Dr. Deborah Good, at UMass, studies the NLH2 gene,
which is a member of the basic helix-loop-helix family of transcription
factors, plays a key role in regulation of genes controlling body
weight in mice. This gene is expressed in the hypothalamic regions
that control appetite by POMC neurons, in both mouse and human
embryos.
The objectives of the described work are to evaluate the
presence of mutations in the NHLH2 gene in obese individuals, to
determine whether there is an association between mutations in
the NHLH2 gene and obesity, and to explore whether such mutations
are present in both adult and pediatric onset obesity.
This study
is a collaborative effort between the clinical services at Baystate
(Pediatric Weight Management Clinic (H. Allen and C. Wittcopp)
and Bariatric Surgery Program (I. Munshi) and the basic science/genetics
lab at UMass-Amherst (D. Good). The study cohort consists of a
total of two hundred patients: 50 obese children (recruited from
Pedi Wt Mgmt Clinic), 50 normal weight children (recruited from
general Pedi Endocrine Clinic), 50 obese adults (recruited from
Bariatric Surgery Prog) and 50 normal weight adults (normal weight
spouses or friends of Bariatric Surgery pts). After informed consent,
each subject will complete a short questionnaire, have a brief
PE, and have a 10 cc venous blood draw. Patients will be given
the option of additionally consenting for DNA banking for future
studies.
Genomic DNA isolation will be performed at Baystate. Samples
will be batched and transferred to Dr. Good's lab at UMass. The
coding regigene will be amplified using PCR, and then the gene
will be sequenced. In patients with confirmed polymorphisms, the
patient will be asked to return for a second blood draw, to reisolate
and independently confirm the polymorphism.
Data analyses will compare
mutations in obese and non-obese subjects and between obese children
and adults. We will evaluate for confounders such as gender, ethnicity,
dietary and activity patterns.
Effective Date- September 1, 2004
Sensitizing ER-Negative Breast Cancer Cells to Tamoxifen
with Rhodiola Crenulate
Sallie Smith Schneider, Ph.D., Pioneer Valley Life Sciences Institute
Kalidas
Shetty, Ph.D., Food Science, UMass Amherst
Richard Arenas, M.D., Department
of Surgery, Baystate Medical Center
Tamoxifen has been proven as a long term adjuvant therapy for
breast cancer. It is amongst a class of anti-estrogens that are
only indicated for the treatment of estrogen receptor positive
breast cancers. However, 30% of human breast cancers are estrogen
receptor negative or unresponsive to tamoxifen. The development
of novel drugs for long term therapy of ER-negative breast tumors
are of interest. Our preliminary data support a potential sensitization
of ER-negative breast cancers to tamoxifen by extracts from a plant
long used in tibetan herbal medicine, Rhodiola crenulate . The
ability to use tamoxifen, a well-accepted and characterized drug,
against ER-negative tumors through phytochemical modification could
revolutionize and expand our current understanding of hormonal
therapy for all breast cancers. The specific aims of the proposal
are to look at whether Rhodiola present in the diet can effect
tumor growth in vivo in the presence or absence of tamoxifen and
to understand the mechanism of sensitization better by examining
changes in the antioxidant enzyme response, estrogen receptor expression
and TGF beta expression.
Effective Date- September 1, 2004
Back to top |
