Foetal motion classification using optical flow displacement histograms

Authors:
Cristina Surlea;Fatih Kurugollu;Peter Milligan;Stephen Ong
Affiliations:
Queen's University Belfast, UK;Queen's University Belfast, UK;Queen's University Belfast, UK;Royal Jubilee Maternity Hospital, Belfast, UK
Venue:
Proceedings of the 4th International Symposium on Applied Sciences in Biomedical and Communication Technologies
Year:
2011

Citing 7
Cited 0

Robot vision

Robot vision
Original Contribution: A scaled conjugate gradient algorithm for fast supervised learning

Neural Networks
A perceptually based physical error metric for realistic image synthesis

Proceedings of the 26th annual conference on Computer graphics and interactive techniques
Event Detection from Video Surveillance Data Based on Optical Flow Histogram and High-level Feature Extraction

DEXA '09 Proceedings of the 2009 20th International Workshop on Database and Expert Systems Application
Artificial Intelligence

Artificial Intelligence
Quantifying motion in video recordings of neonatal seizures by regularized optical flow methods

IEEE Transactions on Image Processing
Estimating Just-Noticeable Distortion for Video

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

Foetal movement has been linked with foetal well-being. In the absence of medical input, its estimation depends exclusively on the mother's subjective opinion. Automatic classification of foetal movement from segmented two dimensional ultrasound scans is a step towards automatic estimation of foetal well-being through foetal motion evaluation. In this paper, optical flow displacement histograms are used to train a backpropagation neural network for classifying foetal movement by means of manually segmented frames that were evaluated independently by two medical experts. Results are promising towards developing an automated foetal motion analysis system but there is still the need for further testing of the hypothesis on a larger number of input samples.