2.gif){kind=small}
Medical Image Processing, Visualisation and Analysis
Tuesday 13th
The second days lectures began with an interesting look at the use of x-ray images for measurement with David Hawkes (Kings College London). The accuracy of using 2D images to measure distances in 3D was assessed. Blood vessel area measurement and vessel tracking were used to highlight the use of image intensity measurements, which also have applications in bone densitometry and dual energy measurements. The use of x-rays to analyse bone structure, leading to accurate and reproducible measurements of joint space width in osteoarthritis and the use of texture analysis on trabecular structure were illustrated. After coffee Mark Henklemen (Toronto) presented an impressive overview of Magnetic Resonance imaging. The need for an optimal resolution in MR imaging to achieve a reasonable SNR per pixel and the differences between MR and photon limited techniques was discussed. The discussion then moved onto artefacts and ways of eliminating them at source in preference to post processing methods. In the afternoon Jeff Bamber (Royal Marsden) led us through ultrasound, starting with signal processing of the transmit pulse wave, the shaping and delaying of the pulse, which has new applications with the use of gas microbubble contrast media. The use of RF processing, amplifiers and filters was shown. Scan conversion methods were discussed including multi-line envelope processing used to extend the field of view in a lateral direction. Post-processing which allows colour and brightness to the displayed, the combination of information e.g. B-mode and colour Doppler and other analysis of texture and boundaries to be applied was explained.
Workshop
This session was followed by the first poster session and computing workshop, giving an opportunity for everyone to get involved and another chance to discuss each others work. The Matlab tutorials were organised on three topics: Fourier transforms in imaging, looking at artefacts in K-space, filtering and sampling. CT and SPECT imaging using projections with centre of rotation and uniformity corrections. Image segmentation using thresholding for CT images and combined with erosion/ dilation for MR images. After supper a group went off to see some Shakespeare at a local college and others found a good pub.
Taken from IPEM SCOPE 8(4) December 1999
Tuesday's Lecture Notes Online
X-rays for Measurement. Prof. David Hawkes
Images That Are Easy to Process: Managing Noise and Artifacts in MR Imaging.(pdf 45kb) Prof. Mark Henkelman
Image Formation and Image Processing in Ultrasound(pdf 28kb) Dr. Jeff Bamber
Wednesday 14th
1999 Summer School
To read the PDF document with a screen reader please link to the Adobe Accessibility Resource Center