Dear Andy,
I'm aware of Roderns' work, we used it in the past, but we didn't find it good enough for our needs in our setup.
Rorderns' algorithm was based on the old normalise algorithm and, given the very weak signal of gray matter in CT, let alone in low dose ones, this is almost invisible. Therefore its registration was mainly skull matching, especially after 8 mm smoothing. Therefore its power was limited. Maybe with very high dose images its strategy of expanding HU values worked better. In presence of noise it just confounded things.
I guess it depends on what you need to do. With my last solution, explained in my last email to John, I've found something that works for me. I was just looking at a way to find a good transformation to the MNI space from low-dose CT (~20 mAs). I can achieve that now (apparently... I'll do in depth testing now, comparing to MRI). It appears to be working better than FDG-PET based normalization!
What I guess will be extremely hard to do, instead, would be to perform VBM or even only segment grey matter well. I've tried on a clinical CT (500 mAs!) and, even there you can clearly see that the gray matter is segmented much thicker than it really is. But that's because it looks that way in the original image!
Hope you can find something of this useful!
Luca
Rispetta l’ambiente: non stampare questa mail se non è necessario.
Respect the environment: if it's not necessary, don't print this mail.
IL TUO 5XMILLE AL SAN RAFFAELE DI MILANO
Devolvi il tuo 5 per mille all’Ospedale San Raffaele: perché al centro della Ricerca ci sei TU. CODICE FISCALE: 07636600962, nel riquadro RICERCA SANITARIA. Non c’è cura, senza ricerca. Non c’è ricerca, senza il tuo 5xmille. Scopri come su http://www.5xmille.org
[http://www.5xmille.org/img/731x129.jpg]
|