Respiratory condition analysis apparatus, respiratory condition display apparatus, processing method therein, and program

摘要

A respiratory condition analysis apparatus includes a respiratory signal acquisition section configured to acquire a time-series respiratory signal including respiratory sound of a living body, a respiratory segment identification section configured to identify a respiratory segment that is a time segment including the respiratory sound in the respiratory signal, a feature value generation section configured to generate a predetermined feature value of the respiratory signal for the identified respiratory segment, and a respiratory abnormality degree generation section configured to generate a degree of abnormality of the respiratory sound included in the respiratory signal on the basis of the feature value.

主权项

1. A respiratory condition analysis apparatus comprising:
one or more processors configured to: acquire a time-series respiratory signal including respiratory sound of a living body, wherein the time-series respiratory signal is measured by one or more sensors coupled to the living body; generate a degree of correlation between a respiratory sound model set in advance and the respiratory signal; generate a degree of reliability of the respiratory sound representing respiratory sound likeness of the respiratory signal on the basis of the correlation degree; identify a respiratory segment that is a time segment including the respiratory sound in the respiratory signal, wherein the respiratory segment is identified on the basis of the respiratory sound reliability degree; generate a predetermined feature value of the respiratory signal for the identified respiratory segment; and generate a degree of abnormality of the respiratory sound included in the respiratory signal on the basis of the feature value, wherein a time frequency signal is generated by performing a time frequency conversion on the respiratory signal, the time frequency signal is projected into a basis vector of the respiratory sound model and a basis conversion signal is generated, and a predetermined nonlinear transformation is performed on the basis conversion signal to output the correlation degree.