The Applied Physics
Laboratory has signed an exclusive agreement with
Zargis Medical Corp., of Princeton, N.J., to license
computer-assisted heart sound analysis technology developed
jointly by Johns Hopkins researchers at APL and the
School of Medicine.
The technology — a pathologic murmur screening
system — was developed over the last six years as
part of the Lab's work in signal processing techniques and
algorithms originally created for detecting and classifying
submarines.
"Just as the Navy has been listening to ocean sounds
for years to distinguish a whale from a submarine, so have
cardiologists been listening to distinguish between normal
heart sounds and pathological ones," said APL physicist
Scott Hayek. "Signal processing technology can be used for
both tasks."
APL's involvement in this area began in the mid-1990s,
when the National Security Technology Department's Joe
Lombardo set about applying time-frequency analysis to the
detection of pathologic heart conditions with a
stethoscope. Hayek extended this work, joining forces with
cardiologist Reid Thompson and his large and growing set of
comprehensive heart sound files collected at the School of
Medicine. Together, Hayek and Thompson developed a
wavelet-based, time-frequency murmur diagnostic
instrument.
Studies have shown that primary care physicians often
refer patients to cardiologists at the first sign of a
suspicious heart sound, and often many of these patients
don't have a pathological condition. These needless
referrals result in wasted money, needless worry and
administrative burdens.
According to Hayek, an automated analysis of digitized
clinical information, such as heart sounds, could have
major implications for healthcare delivery systems using
telemedicine, for primary care physicians and for settings
where a trained health professional is not always
available, i.e., pre-sports participation physicals and
exams performed in remote areas.
The license with the Lab will give Zargis exclusive
commercial rights to APL's signal processing algorithms and
other intellectual property, including an extensive cardiac
research database of some 6,500 recordings from more than
1,200 individuals. Under the agreement, Zargis also will
support further research at APL and the School of Medicine
to continue development of the technology.