[abstract] THE AUTOMATED EXTRACTION OF THE CHARACTERISTICS OF BLOOD CELL AND BUBBLE MOVEMENT FROM DOPPLER ULTRASOUND RECORDINGS

Rubicon Research Repository/Manakin Repository

[abstract] THE AUTOMATED EXTRACTION OF THE CHARACTERISTICS OF BLOOD CELL AND BUBBLE MOVEMENT FROM DOPPLER ULTRASOUND RECORDINGS

Show full item record


Title: [abstract] THE AUTOMATED EXTRACTION OF THE CHARACTERISTICS OF BLOOD CELL AND BUBBLE MOVEMENT FROM DOPPLER ULTRASOUND RECORDINGS
Author: Chappell, C; Payne, SJ
Abstract: INTRODUCTION: Doppler Ultrasound signals are widely used to grade the quantity of circulating gas emboli in divers. Current techniques rely on trained observers, making the grading process both time-consuming and subjective. The automated detection of bubbles, however, is confounded by the presence of signals arising from heartbeats. The aim of this study was thus to parameterise the features of both the heartbeat and bubbles in time-frequency space, such that information about the nature and number of bubbles could automatically be extracted from the signal. MATERIALS AND METHODS: Doppler Ultrasound signals from recreational divers, post-decompression, were provided by DAN Europe. The spectrogram of each signal was used to determine features pertaining to both the heartbeat and bubble signals. These were both parameterised in terms of Gaussian Mixture Models (GMM). RESULTS: It was found that the time-frequency pattern associated with heartbeats is parameterised by a GMM with approximately 7 kernels (a total of 21 free parameters), and that both intra- and inter-subject variation can be modeled using these parameters. A similar approach can be applied to locate bubbles, which occupy a wider frequency range, but a shorter time scale than the blood cell motions. The GMM for the blood cell motions was used to locate individual heartbeats in the Doppler Ultrasound signal. This is achieved with high sensitivity and specificity, even in the presence of bubbles. The bubbles are then separately identified, and the parameters of the GMM related to the underlying physiological nature of the bubble. CONCLUSIONS: The GMM offers a method by which Doppler Ultrasound data may be analysed and its separate components (bubbles and blood cell motion) identified, leading to the automated extraction of useful clinical information and hence towards an automatic grading system.
Description: Undersea and Hyperbaric Medical Society, Inc. (http://www.uhms.org )
URI: http://archive.rubicon-foundation.org/1567
Date: 2004

Files in this item

Files Size Format View
abstract.txt 259bytes Text file View/Open

This item appears in the following Collection(s)

  • UHMS Meeting Abstracts
    This is a collection of the published abstracts from the Undersea and Hyperbaric Medical Society (UHMS) annual meetings.

Show full item record

Browse

My Account