The number of functional magnetic resonance imaging (fMRI) studies performed on the human spinal cord (SC) has considerably increased in recent years. The lack of a validated processing pipeline is, however, a significant obstacle to the spread of SC fMRI. One component likely to be involved in any such pipeline is the process of SC masking, analogous to brain extraction in cerebral fMRI. In general, SC masking has been performed manually, with the incumbent costs of being very time consuming and operator dependent. To overcome these drawbacks, we have developed a tailored semiautomatic method for segmenting echoplanar images (EPI) of human spine that is able to identify the spinal canal and the SC. The method exploits both temporal and spatial features of the EPI series and was tested and optimized on EPI images of cervical spine acquired at 3 T. The dependence of algorithm performance on the degree of EPI image distortion was assessed by computing the displacement warping field that best matched the EPI to the corresponding high-resolution T(2) images. Segmentation accuracy was above 80%, a significant improvement over values obtained with similar approaches, but not exploiting temporal information. Geometric distortion was found to explain about 50% of the variance of algorithm classification efficiency. (C) 2011 Elsevier Inc. All rights reserved.

Semiautomated segmentation of the human spine based on echoplanar images / Giovanni, Giulietti; E., Summers Paul; Diana, Ferraro; Carlo A., Porro; Maraviglia, Bruno; Giove, Federico. - In: MAGNETIC RESONANCE IMAGING. - ISSN 0730-725X. - STAMPA. - 29:10(2011), pp. 1429-1436. [10.1016/j.mri.2011.08.006]

Semiautomated segmentation of the human spine based on echoplanar images

Giovanni Giulietti;MARAVIGLIA, Bruno;GIOVE, FEDERICO
2011

Abstract

The number of functional magnetic resonance imaging (fMRI) studies performed on the human spinal cord (SC) has considerably increased in recent years. The lack of a validated processing pipeline is, however, a significant obstacle to the spread of SC fMRI. One component likely to be involved in any such pipeline is the process of SC masking, analogous to brain extraction in cerebral fMRI. In general, SC masking has been performed manually, with the incumbent costs of being very time consuming and operator dependent. To overcome these drawbacks, we have developed a tailored semiautomatic method for segmenting echoplanar images (EPI) of human spine that is able to identify the spinal canal and the SC. The method exploits both temporal and spatial features of the EPI series and was tested and optimized on EPI images of cervical spine acquired at 3 T. The dependence of algorithm performance on the degree of EPI image distortion was assessed by computing the displacement warping field that best matched the EPI to the corresponding high-resolution T(2) images. Segmentation accuracy was above 80%, a significant improvement over values obtained with similar approaches, but not exploiting temporal information. Geometric distortion was found to explain about 50% of the variance of algorithm classification efficiency. (C) 2011 Elsevier Inc. All rights reserved.
2011
epi; fmri; k-means; segmentation; spinal cord; temporal standard deviation
01 Pubblicazione su rivista::01a Articolo in rivista
Semiautomated segmentation of the human spine based on echoplanar images / Giovanni, Giulietti; E., Summers Paul; Diana, Ferraro; Carlo A., Porro; Maraviglia, Bruno; Giove, Federico. - In: MAGNETIC RESONANCE IMAGING. - ISSN 0730-725X. - STAMPA. - 29:10(2011), pp. 1429-1436. [10.1016/j.mri.2011.08.006]
File allegati a questo prodotto
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/446766
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? 1
  • Scopus 9
  • ???jsp.display-item.citation.isi??? 7
social impact