Improving the extraction of temporal motion strength signals from video recordings of neonatal seizures

Author

Karayiannis, Nicolaos B. ; Tao, Guozhi

Author_Institution

Dept. of Electr. & Comput. Eng., Houston Univ., TX, USA

fYear

2003

fDate

21-22 July 2003

Firstpage

87

Lastpage

92

Abstract

The paper presents a procedure developed to extract quantitative information from video recordings of neonatal seizures in the form of temporal motion strength signals. These signals are obtained by applying nonlinear filtering, segmentation, and morphological filtering on the differences between adjacent frames. The experiments indicate that temporal motion strength signals constitute an effective representation of videotaped clinical events and can be used for seizure recognition and characterization.

Keywords

biomedical optical imaging; feature extraction; image motion analysis; image segmentation; mathematical morphology; neurophysiology; nonlinear filters; paediatrics; patient monitoring; pattern classification; video signal processing; central nervous system disorders; clinical events; morphological filtering; neonatal seizures; nonlinear filtering; segmentation; seizure characterization; seizure recognition; temporal motion strength signal extraction; video recordings; Area measurement; Character recognition; Data mining; Filtering; Information analysis; Leg; Motion detection; Pediatrics; Surveillance; Video recording;

fLanguage

English

Publisher

ieee

Conference_Titel

Advanced Video and Signal Based Surveillance, 2003. Proceedings. IEEE Conference on

Print_ISBN

0-7695-1971-7

Type

conf

DOI

10.1109/AVSS.2003.1217906

Filename

1217906