Experienced cardiologists can usually recognize pathologic heart murmurs wi
th high sensitivity and specificity, although nonspecialists with less clin
ical experience may have more difficulty. Harsh, pansystolic murmurs of int
ensity grade greater than or equal to3 at the left upper sternal border (LU
SB) are likely to be associated with pathology. In this study, we designed
a system for automatically detecting systolic murmurs due to a variety of c
onditions and examined the correlation between relative murmur intensity an
d likelihood of pathology. Cardiac auscultatory examinations of 194 childre
n and young adults were recorded, digitized, and stored along with correspo
nding echocardiographic diagnoses, and automated spectral analysis using co
ntinuous wavelet transforms was performed. Patients without heart disease a
nd either no murmur or an innocent murmur (n = 95) were compared to patient
s with a variety of cardiac diagnoses and a pathologic systolic murmur pres
ent at the LUSB (n = 99). The sensitivity and specificity of the automated
system for detecting pathologic murmurs with intensity grade greater than o
r equal to2 were both 96%, and for grade greater than or equal to3 murmurs
they were 100%. Automated cardiac auscultation and interpretation may be us
eful as a diagnostic aid to support clinical decision making.