British Journal of Radiology (2005) 78, 245-251
© 2005 British Institute of Radiology
doi: 10.1259/bjr/33540239
Techniques and trouble-shooting in high spatial resolution thin slice MRI for rectal cancer
G Brown, MD, MRCP, FRCR
1
I R Daniels, FRCS
2
C Richardson, DCRD
1
P Revell, DCR
3
D Peppercorn, FRCR
2 and
M Bourne, DMRD, FRCR
4
1 Department of Radiology, Royal Marsden Hospital, Fulham Road, London SW3 6JJ, 2 Pelican Cancer Foundation, Pelican Centre, North Hampshire Hospital, Aldermaston Road, Basingstoke, Hants RG24 9NA, 3 Siemens Medical, Siemens House, Oldbury, Bracknell, Berkshire RG12 8FZ and 4 Department of Radiology, University Hospital of Wales, Heath Park, Cardiff CF14 4XW, UK
View larger version (51K):
[in a new window]
|
Figure 1. (a) Endorectal coil and (b) T2 weighted image using the endorectal coil. The bowel wall layers are shown but there is insufficient detail from the surrounding mesorectum to adequately stage rectal tumours.
|
|
View larger version (169K):
[in a new window]
|
Figure 2. Planning the high resolution images. Images are acquired orthogonal to the long axis of the rectum. Note that for a low tumour, an additional oblique coronal scan (top right image) is also acquired, which is planned along the long axis of the anal canal.
|
|
View larger version (168K):
[in a new window]
|
Figure 3. Poor coil positioning. The images shown (top left and right) have been obtained with the lower edge of the pelvic coil placed at the level of the symphysis pubis. Consequently there is poor signal to noise ratio from the lower rectum and anal sphincter region. Repositioning the coil restores adequate signal to noise ratio.
|
|
View larger version (164K):
[in a new window]
|
Figure 4. It is important to centre the coil correctly to the whole rectum. Thus the lower edge of the coil needs to be placed so that it lies at least 10 cm below the symphysis pubis in order to ensure adequate signal is obtained from the lower rectum and anorectal junction. The upper most limit for coil placement is the sacral promontory.
|
|
View larger version (111K):
[in a new window]
|
Figure 5. A T1 weighted image (repetition time (TR) 500 ms, echo time (TE) 10 ms) has been obtained with a high spatial resolution 1 mm3 voxel but the bowel wall layers cannot be distinguished.
|
|
View larger version (159K):
[in a new window]
|
Figure 6. T1 fat saturated image with contrast enhancement. The bright fat signal intensity from perirectal fat (asterisk) has been suppressed and is of low signal intensity, there is florid enhancement of the entire bowel wall (arrow) and perirectal vessels (arrowheads) which would not be readily distinguishable from tumour if it were present.
|
|
View larger version (45K):
[in a new window]
|
Figure 7. (a) T1 weighted (repetition time (TR)=500 ms, echo time (TE) 10 ms) and (b) T2 fast spin echo (FSE) (TR>3000 ms, TE 105 ms) axial images of rectal tumour. The bowel wall layers cannot be seen on the T1 weighted images. Vessels (arrowhead), lymph node (arrow), tumour (open arrow) and muscle coat (curved arrow) all return very similar signal intensities making distinction between structures difficult on the T1 weighted image. This compares with the T2 weighted image, which shows clear separation of tumour and bowel wall layers due to high inherent tissue contrast.
|
|
View larger version (141K):
[in a new window]
|
Figure 8. Cross-talk. The image has been degraded by noise resulting from cross-talk. This varies with machines and manifests as loss of signal. This is overcome by interleaving slices during the high resolution acquisition.
|
|
View larger version (73K):
[in a new window]
|
Figure 9. Motion artefact. (a) This characteristically produces a series of bands that degrade the image. (b) This can often be overcome by ensuring the patient is not in discomfort, by repositioning the patient and ensuring adequate abdominal compression and repeating the sequence again after swapping phase and frequency directions.
|
|
Copyright © 2005 by the British Institute of Radiology.
