Estimating the age of healthy infants from quantitative myelin water fraction maps

Abstract

The trajectory of the developing brain is characterized by a sequence of complex, nonlinear patterns that occur at systematic stages of maturation. Although significant prior neuroimaging research has shed light on these patterns, the challenge of accurately characterizing brain maturation, and identifying areas of accelerated or delayed development, remains. Altered brain development, particularly during the earliest stages of life, is believed to be associated with many neurological and neuropsychiatric disorders. In this work, we develop a framework to construct voxel-wise estimates of brain age based on magnetic resonance imaging measures sensitive to myelin content. 198 myelin water fraction (VFM) maps were acquired from healthy male and female infants and toddlers, 3 to 48 months of age, and used to train a sigmoidal-based maturational model. The validity of the approach was then established by testing the model on 129 different VFM datasets. Results revealed the approach to have high accuracy, with a mean absolute percent error of 13% in males and 14% in females, and high predictive ability, with correlation coefficients between estimated and true ages of 0.945 in males and 0.94 in females. This work represents a new approach toward mapping brain maturity, and may provide a more faithful staging of brain maturation in infants beyond chronological or gestation-corrected age, allowing earlier identification of atypical regional brain development. Hum Brain Mapp 36:1233–1244, 2015. © 2015 The Authors Human Brain Mapping Published by Wiley Periodicals, Inc.

Improved determination of the myelin water fraction in human brain using magnetic resonance imaging through Bayesian analysis of mcDESPOT

“…The first evaluation was performed for different f s values of 0.05, 0.1, 0.15, 0.2, 0.3 and 0.5 at SNR values of 500, 1000 and 2000, where the other underlying input parameters were fixed at T 2,s = 15 ms, T 2,l = 100 ms, T 1,s = 450 ms and T 1,l = 1800 ms. The second evaluation was performed by fixing the f s and SNR values to 0.15 and 500, respectively, and exploring several values of the remaining parameters, with T 2,s = (10, 15, 20, 25) ms, T 2,l = (70, 100, 110, 150) ms, T 1,s = (300, 400, 450, 550) ms and T 1,l = (1400, 1800, 2000, 2500) ms.…”

“…Results were obtained by fixing the f s and SNR values to 0.15 and 500, respectively, and varying the values of the remaining parameters. Values used were T 2,s = (10, 15, 20, 25) ms, T 2,l = (70, 100, 110, 150) ms, T 1,s = (350, 400, 450, 550) ms and T 1,l = (1400, 1800, 2000, 2500) ms.…”

Brain Myelin Water Fraction and Diffusion Tensor Imaging Atlases for 9‐10 Year‐Old Children

“…MWI has been used to investigate myelin in a wide range of diseases including multiple sclerosis, neuromyelitis optica spectrum disorder, schizophrenia, Niemann‐Pick disease, prenatal alcohol exposure, amyotrophic and primary lateral sclerosis, Krabbe disease, autism, stroke, and traumatic brain injury 34–47 . Rapid myelination has been demonstrated in early childhood and adolescence using multicomponent driven equilibrium single pulse observation of T 1 and T 2 (mcDESPOT), an alternate approach to MWI using a steady‐state acquisition 48–53 . Adult multi‐echo T 2 ‐based MWF atlases in both brain and cervical spinal cord have recently been published, 54,55 but, to our knowledge, no such atlases exist for children.…”

Rapid simultaneous high-resolution mapping of myelin water fraction and relaxation times in human brain using BMC-mcDESPOT

“…1–2) as well as more quantitative analysis (Fig. 3–5) showed higher dispersion in derived parameter estimates as compared to the literature (26, 28, 33, 44, 49, 67), especially using SRC. We attribute this to our use of relatively large parameter space bounds to avoid bias (33, 47, 51); as has previously been noted, this comes at the expense of increased dispersion (47, 51).…”

“…Finally, mcDESPOT makes use of conventional MR acquisition sequences, namely, fully balanced steady state free precession (bSSFP) and spoiled gradient recalled echo (SPGR), widely available on clinical MRI systems. Overall, then, mcDESPOT is particularly attractive for clinical investigations (26, 30, 45). …”