MPhil Thesis Defence


"Ultrasound Image Processing and Transmission for Medical Diagnosis"

By

Miss Xing Zheng


Abstract

Nowadays, the Internet connects hospitals, medical institutions and people
together. In this thesis, we focus on a medical application on ultrasound
image processing and transmission. We consider that a pregnant woman takes
her ultrasound examination in a local hospital, and the diagnosis and
analysis of her ultrasound video (or image sequences) could be done
remotely via the Internet. It is also possible for medical professionals
and her family members from a distance to simultaneously view her
ultrasound video using a multicast-capable network. In order to facilitate
clinicians' diagnosis and improve medical information distribution, we
address the issues in the clinical system related to the two following two
components.

The first component in the system focuses on automatic detection of fetal
head features in ultrasound images. In clinics, fetal biometric
measurement is done manually in the qualified ultrasound scanning images,
in which some important predefined features exist. We present simple and
new methods for automatic detection of these features in a single head
ultrasound image. We also describe our system and criteria for ranking
multiple images to facilitate clinicians' subsequent image selection and
measurements in a sequence of ultrasound images. The results show that the
methods can successfully identify features, and our system's ranking and
the clinicians' ranking are well matched.

The second component in the system is to offer quality real-time medical
video to users over multicast networks. Since the video packets may be
lost, they have to be mostly recovered at clients. We present a
feedback-free recovery scheme for layered video, in which the server
multicasts FEC (Forward Error Correction) and pseudo-ARQ packets (source
packets) in parallel with the original video packets. The receivers,
depending on their local losses, autonomously and dynamically join the
layers to minimize their errors after correction. Via analysis, we study
the optimal allocation of layer bandwidths, and the optimal combination of
FEC and pseudo-ARQ packets that the server should provide and a receiver
should join. We show that our scheme can substantially reduce the residual
error rate as compared with pure FEC or pure pseudo-ARQ.


Date:				Monday, 5 May 2003

Time:				4:00p.m.-6:00p.m.

Venue:				Room 2304
				Lifts 17-18

Committee Members:		Dr. Gary Chan (Supervisor)
				Dr. Albert Chung (Supervisor)
				Dr. Long Quan (Chairman)
				Dr. Michael Brown


**** ALL are Welcome ****