1076 The Effect of Head Motion on Data Quality for Neonatal Dti. (October 2012)

• Record Type: Journal Article
• Authors: Plaisier, A; Heemskerk, AM; Reiss, IKM; Lequin, MH; Leemans, A; Dudink, J
• Is Part Of: Archives of disease in childhood. Volume 97(2012)Supplement 2
• Page Start: A309
• Page End: A309
• Publication Date: 2012-10
• Subjects: Children -- Diseases -- Periodicals; Infants -- Diseases -- Periodicals; 618.920005
• DOI: 10.1136/archdischild-2012-302724.1076
• Languages: English
• ISSNs: 0003-9888

Abstract

Abstract : Background and Aims: Diffusion Tensor Imaging (DTI) has become valuable for quantitative evaluation of white matter maturation in preterm infants. Because of the occurrence of head movement, gathering good quality data is challenging in neonatal neuroimaging. This is especially of concern for DTI, where motion can result in severe signal drop-out and therefore miscalculation of DTI parameters if data outliers are not handled correctly. This study was aimed to quantify the occurrence of motion artefacts in neonatal DTI and to evaluate different methods for tensor estimation. Methods: We prospectively collected DTI data of 27 preterm infants that were scanned at 30 weeks gestational age. DTI data was acquired in 25 directions. Percentage outliers per slice was calculated. With Explore DTI, we assessed the effect of motion artefacts on tensor estimation using different methods. Results: 60% of subjects had slightly corrupted data (>15 slices with >30% outliers) of which 40% had severely corrupted data (>10 slices with >50% outliers). Corrupted data resulted in erroneous DTI parameters. This was especially true for the tensor estimation (ordinary least squares) typically performed by vendors and popular DTI software. More advanced tensor estimations showed more reliable data. Conclusions: Motion artefacts are a major problem in neonatal DTI as it can compromise accurate calculation of DTI parameters. These results press the need for careful data inclusion and the use of reliable methods for tensor estimation. Targeted acquisition, processing and quality assessment is needed in this population to obtain reliable evaluation of white matter maturation.