 |
"Now, I use PV-WAVE to interactively run simulations and plot the
data, and I have results in a matter of minutes instead of hours."
-Brian Young, Senior Software Engineer,
Marquette Medical Systems
QUICK
FACTS
Engineers at Marquette Medical Systems used PV-WAVE in order to
create a signal processing application that could analyze data points
taken by a patients heartbeat signals. The powerful visualization
and mathematical capabilities of PV-WAVE allowed for a sophisticated
and powerful analysis program that could examine specific data points
and identify any irregular heartbeats in the data.
THE PROBLEM
Marquette Medical Systems is a GE Medical Systems Company. The extensive
product offerings at Marquette include a line of noninvasive cardiac
diagnostic systems. The MARS® (Multi-parameter Arrhythmia Review
Station) 8000 unity workstation is a sophisticated patient monitoring
system that combines Marquette's innovative electrocardiogram (ECG)
monitoring technology with the processing power of a Sun Microsystems
workstation operating on the UNIX platform.
The MARS 8000 records a patient's heartbeat signals for 24 to 48
hours to check for any irregularities. After all the data has been
collected, it must be analyzed and presented in a way that enables
ECG technicians to easily scan the recordings and catch all relevant
data. From these key data points, the technicians create a thorough
and accurate ECG report to be reviewed by the patient's doctor.
An adult human heart beats approximately 100,000 times each day,
so the amount of data captured when monitoring a person's ECG for
24 hours is enormous. An average file runs about 80 Mbytes in size.
Marquette software engineers had to create signal processing software
that would "examine" each data point, determine if it
was indeed an irregular heartbeat, and then present this data in
a format that would aid technicians and physicians in looking at
only the pertinent data (i.e., the irregular heartbeats).
 |
Several factors complicate this signal processing analysis. The
MARS 8000 system can collect inputs from a variety of different
sources. For instance, the ECG signals can be captured from a battery-operated
unit that the patient wears, from hospital equipment at the patient's
bedside or from a hospital network. Each source has distinct characteristics
that can affect the signals themselves. Even greater variations
occur from patient to patient since each individual's heartbeats
are somewhat unique. The software needed to be able to factor those
differences into the analysis of each individual's signals.
Additionally, many other irregular signals are picked up when a
patient wears a monitor. Certain movements create abnormal-looking
signals that may be mistaken for irregular heartbeats. The system
had to eliminate these irrelevant signals without eliminating any
heartbeat data. And lastly, it was very difficult to take these
massive data files and present them in a meaningful, easy-to-understand
format.
Software engineers had been using self-authored tools for signal
processing analysis and presentation, but as the system's level
of sophistication grew, so did the need for more powerful analysis
and visualization software.
THE SOLUTION
PV-WAVE from Visual Numerics delivered the powerful visualization
and mathematical capabilities Marquette needed. Visual Numerics
provides industry-leading, mission-critical software solutions to
business and technical professionals around the world. Scientists,
researchers, educators, engineers, developers, Intranet managers,
testers and analysts use Visual Numerics' decision support tools
to solve problems, identify trends and share results.
Senior software engineer Brian Young compared various analytical
and visualization software packages. "Our needs had grown beyond
the capabilities of our in-house software tools, so we began to
research our options" says Young. "The type of signal
processing we do is very sophisticated and we needed more computing
power. PV-WAVE offered the most powerful visualization and graphics
tools and most comprehensive mathematical libraries of any of the
software we considered."
RETURN ON INVESTMENT
"Before we used PV-WAVE, I had to spend hours writing a C program
each time I wanted to try out a new idea, and then use a separate
tool to plot the data and measure the performance," Young says.
"It was very frustrating and time-consuming. Now, I use PV-WAVE
to interactively run simulations and plot the data, and I have results
in a matter of minutes instead of hours. My main objective was to
find software that enables me to use my time as efficiently as possible.
PV-WAVE has served me well." He adds, "My experience with
Visual Numerics as a company has been very positive. I have had
very few technical problems, but when I have had issues, both the
technical support and the sales staffs have been very responsive.
At Marquette Medical Systems, we see ourselves as our customers'
strategic partner in improving patient care and controlling costs
through innovative diagnostic, monitoring and clinical information
systems. Visual Numerics has partnered with us by providing not
only the best technology, but also service that is unparalleled
in the industry."
WORLD CLASS PRODUCTS, SERVICES,
AND SUPPORT
Visual Numerics has provided technical software
solutions for numerical analysis and visualization for over 30 years.
The company's software products help users understand complex data
from a variety of sources and build business-critical applications.
Visual Numerics offers two product lines: the IMSL® Numerical
Libraries for powerful mathematical and statistical analysis and
the PV-WAVE® visual data analysis development environment. Visual
Numerics also offers customized consulting services for applications
that involve mathematical, statistical, or visual data analysis
to meet today’s business analytical needs.
The IMSL Numerical Libraries which include
the IMSL C Library, IMSL Fortran Library and JMSL Library
for Java applications are the industry standard for numerical
analysis. They deliver developers with the breadth and depth of
core algorithms allowing for the rapid development of any application.
Whether developing applications in C, Fortran, or Java, or on UNIX,
Windows or Linux, the robust IMSL Libraries provide the reliable
foundation and the building blocks developers need.
The PV-WAVE family of products which includes
PV-WAVE, TS-WAVE, and JWAVE delivers engineers with the development
tools to efficiently and accurately meet their data analysis needs.
PV-WAVE solutions allow users to rapidly import, manipulate, analyze
and visualize data. The PV-WAVE family also includes robust time
series analysis software as well as the ability to share analysis
results across the enterprise with a Java-based solution. And, unlike
other products, PV-WAVE Advantage includes a sophisticated set of
analysis routines based on the industry-standard IMSL Libraries.
In addition, Visual Numerics Consulting
Services combine technical expertise, decades of hands-on experience
and a combination of powerful products to create the highest quality
solutions possible for your visual data analysis needs.
Visual Numerics unique combination of products
and services rapidly enhance ROI by delivering the highest efficiency,
greatest accuracy and maximum performance.
|
 |
 |
 |

Industry
Life Sciences
Application
Cardiac Diagnostic Analysis
Product
PV-WAVE
Marquette Medical Systems, a GE Medical Systems Company, is a
leading developer and supplier of innovative diagnostic, monitoring
and clinical information systems for use in national and regional
healthcare systems, hospitals, clinics and teaching and research
facilities worldwide. Founded in 1965, the company is headquartered
in Milwaukee, WI, and employs 3,000 people throughout North America,
Europe and Asia.
Key Benefits
| > |
Capable of interactively running simulations
and plotting data |
 |
| > |
Powerful visualization and graphics tools |
 |
| > |
Comprehensive mathematical libraries |
 |
| > |
Able to present massive data files in meaningful
format |
 |
| > |
Responsive Technical Support and Sales Team |
 |
| |
|
 |
| |
|
 |
| |
|
| |
| |
|
| |
|