Automatic classification of spirometric data.
Explore the source record for details and available documents.
Computers↗
Biomedical subjects
J F Donohue
Publications and source records attributed to J F Donohue.
Forced expiratory spirometric parameters derived by feature-extraction techniques.
Spirometric data from 200 patients seen in the pulmonary function laboratory were analyzed retrospectively. Using standard criteria, the spirometric patterns of the subjects were classified as indeterminate, normal, restrictive, or obstructive; obstruction was further classified as mild, moderate, severe, or very severe. A formal feature-extraction technique was used to define 2 new parameters. These were linear combinations of the 5 standard spirometric parameters and explained 94% of the variation in the original parameters. The derived parameters provided better separation of the various classes than the combination of the normalized forced vital capacity and the ratio of the forced expiratory volume in 1 s to the forced vital capacity.
Forced Expiratory Flow Rates↗