 |
 |
Well, one have to be mindful of basic neurobiology when imaging of primate infants. Generally, infant's brain is not myelenated in anything other that cortico-spinal track. In such a case, T1-w will not give you significant contrast.However, T2-w contrast is quite high. So I just run a true-fist sequence which can be used for tissue segmentation. ----- Original Message ----- From: "Matt Glasser" <[log in to unmask]> To: <[log in to unmask]> Sent: Friday, December 28, 2007 8:25 AM Subject: Re: [FSL] Re-forwarding Segmentation of brains of infant chimpanzees Hi, I wonder if there is not enough contrast between unmeylinated white matter and grey matter? We have gotten FAST to work pretty well in adult chimpanzees, only making some errors in motor cortex and the sub cortical grey nuclei. One thing that can help increase the image contrast is to square the image (fslmaths -sqr), if you do it a couple of times, you may have to reduce the values in the image by dividing by some large number (e.g. 1e6) and additionally set an upper threshold to remove any blood vessels (which will become so high in value that they will be classified in a separate tissue category). Peace, Matt. -----Original Message----- From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Tomoko Sakai Sent: Friday, December 28, 2007 4:08 AM To: [log in to unmask] Subject: [FSL] Re-forwarding Segmentation of brains of infant chimpanzees Dear FSL list (I’m sorry to send a garbled e-mail. I send you an e-mail again.) I am attempting segmentation of brains of infant chimpanzees using FAST. I succeeded to segment gray and white matter regions in the older subjects. However, the white matter regions seemed to be overestimated while the gray matter regions seemed to be underestimated in younger infants under one year old. I would like to ask you to give me any suggestion that I should try. I would also like to learn the method to segment the unmyelinated white, and myelinated white matter regions using FAST? The brains of infant chimpanzees were scanned under T1 weighted, and the data were processed as follows. (1) Scans were corrected for fluctuations in average intensity using SUSAN. (2) Brains were aligned to anterior commisure‐posterior commisure orientation in transverse view using MRIcro. (3) Cerebral portions of brains were extracted using BET (option: -R -m - s -r 50 -f 0.3 -g 0.2) and FSLview. (4) Gray and white matter regions were segmented using FAST (option: -t 1 - c 3 -a -A -od, -os -ob -i 16 -l 200 -v 1- 5). I appreciate for your help in advance. Tomoko Sakai Primate Research Institute Kyoto University ____________________________________________________________________ Tomoko Sakai Primate Research Institute, Kyoto University Brain Research Section・Department of Behavioral and Brain Sciences Inuyama, Aichi 484-8506, JAPAN TEL. +81-568-63-0567 (Administrative Office) FAX. +81-568-63-0085 E-mail: [log in to unmask] ____________________________________________________________________