Meeting Banner
Abstract #0246

Testing the Sensitivity of Tract-Based Spatial Statistics to Simulated Treatment Effects in Preterm Neonates

Gareth Ball1, Serena J. Counsell1, Tomoki Arichi1, 2, Nazakat Merchant1, 2, Daniel Rueckert3, A David Edwards1, 2, James P. Boardman, 14

1Centre for the Developing Brain, Imaging Sciences Department, MRC Clinical Sciences Centre, Hammersmith Hospital, Imperial College London, London, United Kingdom; 2Division of Neonatology, Imperial College London Healthcare NHS Trust, London, United Kingdom; 3Biomedical Image Analysis Group, Department of Computing, Imperial College London, London, United Kingdom; 4Simpson Centre for Reproductive Health, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom

Preterm birth is associated with poor neurodevelopmental outcome. Microstructural alterations in the developing white matter are thought to represent key components of preterm brain injury. Developing early imaging biomarkers that are sensitive to these alterations, such as TBSS, would be beneficial to the evaluation of early neuroprotective strategies in preterm infants. Here, the sensitivity of TBSS to detect changes in white matter microstructure in neonates is tested by simulating global treatment effects, represented by increased fractional anisotropy, in groups of different sizes. Simulations were found to predict well a real biological effect, represented by increasing age at scan.

Abstract PDF Presentation Video

Keywords

according allow alterations anisotropy appropriate artificially assessing background ball biomedical brain carry clinical cognition college common computing concept constructed coordination corrected datasets deficits detect detected developing development difficult disabilities division dotted earlier early efficacy equivalent evaluate even expected fewer fractional future global health hospital ideal imperial increasing indicate infant infants infirmary injury intervention kingdom major maps marks measure merchant neonatal neonates optimized outcome patients pediatrics percentage pipeline porter potential powerful predicted preliminary profound proof proposed randomly readily real registered relatively represent represented reproductive risk royal scanned sciences sensitivity severe simulate simulated simulating simulations skeleton smith space spatial spectrum split statistical statistics strategies subjects template term therapeutic tract transformation treated treatment trials trust underlie untreated weeks white whole young