Kim Annink

268 Chapter 12 Predicting outcome There are currently many biomarkers for brain injury investigated in infants with HIE, e.g. MRI, (amplitude integrated) electric encephalography, monitoring cerebral oxygenation, cerebral ultrasound and evoked potentials (11,69). Research also focuses on clinical parameters as Apgar score, lactate and Thompson score and their relation with the severity of brain injury (69), as well as plasma biomarkers such as S100β, Tau, BNPD and metabolomics (70,71). For neonates with HIE, most of these data are collected as part of clinical care. Combining these biomarkers can increase the accuracy of outcome prediction in HIE and thereby optimizing the counseling of parents and clinical decision making (72). Machine learning can be a helpful method to develop the most accurate prediction model for infants with HIE. Ultra-high field neuroimaging needs optimization to further increase the additional diagnostic value. Afterwards, 7.0T MRI might serve as an additional biomarker to predict long-term outcome. It might help us to improve visualization of the anatomy, which might allows us to diagnose ulegyria in the cerebellum. It might also increase our knowledge about brain development in infants with brain injury. DWI might provide additional information about the development of microstructural connectivity, MRS about metabolic maturation and chemical exchange saturation transfer can give information about myelination (73). An accurate prediction of outcome might become even more important in the future, if the focus will be more on value-based health care.