Automated detection of important organs of a foetus in ultrasonic videos
Principal Supervisor: Dr. Huiyu Zhou
Second Supervisor: Prof. Neil Robertson
+ Project Description
It was found that women presenting with a sudden decrease of foetal movements (DFM) are exposed to a higher risk of foetal death. Foetal distress is not the only possible cause for DFM, other factors include external stimuli, drugs and sleep periods of the foetus. Detection of foetal motions that cannot be perceived by the mother will instil a sense of reassurance into her and decrease the number of unplanned hospital admissions or procedures. To describe the motion of a foetus, it is very important for one to identify the positions of the head, limbs and body of the foetus in a noisy environment. In this project, we intend to develop an automated system to detect the head, limbs and body of a foetus from videos so that the movements of these important organs can be fully analysed in a later stage.
This investigation mainly contains the following objectives:
(1) To extract spatio-temporal image features with bag of words to describe the head, limbs and body of a foetus.
(2) To effectively separate the foetal tissue from that of the mother using deep learning techniques.
(3) To experimentally demonstrate that the proposed detection system can produce more accurate and efficient results than the available systems reported in the literature.
Project collaborator: Dr. Stephen Ong, Royal Jubilee Maternity Hospital, Belfast.
+ How to Apply
Applicants should apply electronically through the Queen’s online application portal at: https://dap.qub.ac.uk/portal/
+ Contact Details
|Supervisor Name:||Dr. Huiyu Zhou|
School of Electronics
+44 (0)28 90971753