G. Ferrigno et P. Carnevali, PRINCIPAL COMPONENT ANALYSIS OF CHEST-WALL MOVEMENT IN SELECTED PATHOLOGIES, Medical & biological engineering & computing, 36(4), 1998, pp. 445-451
A method is presented for assessing a compact set of parameters charac
teristic of respiratory system functional status. 3D movements of poin
ts in the chest wall and the volumes of chest wall compartments (pulmo
nary rib cage, abdominal rib cage and abdomen) are considered. The cc-
ordinates of these points are measured using an optoelectronic system
for 3D motion analysis. Principal component analysis is applied to the
se data. The behaviour of the eigenvectors of the covariance matrix of
the 3D co-ordinates of the points on the chest wall shows close agree
ment with the pathology characteristics. The same is found for the per
centage of total variance explained by the principal components of the
volume variations. In this case, the higher values of variance percen
tage explained indicate independent motions (active or passive) in the
degrees of freedom of the system identified by partitioning the total
volume into compartments.