Association of gestational age with MRI-based biometrics of brain development in fetuses
- PDF / 1,595,591 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 1 Downloads / 152 Views
Open Access
RESEARCH ARTICLE
Association of gestational age with MRI‑based biometrics of brain development in fetuses Yuequan Shi1, Yunjing Xue2, Chunxia Chen1, Kaiwu Lin1 and Zuofu Zhou1*
Abstract Background: Reported date of last menstrual period and ultrasonography measurements are the most commonly used methods for determining gestational age in antenatal life. However, the mother cannot always determine the last menstrual period with certainty, and ultrasonography measurements are accurate only in the first trimester. We aimed to assess the ability of various biometric measurements on magnetic resonance imaging (MRI) in determining the accurate gestational age of an individual fetus in the second half of gestation. Methods: We used MRI to scan a total of 637 fetuses ranging in age from 22 to 40 gestational weeks. We evaluated 9 standard fetal 2D biometric parameters, and regression models were fitted to assess normal fetal brain development. A stepwise linear regression model was constructed to predict gestational age, and measurement accuracy was determined in a held-out, unseen test sample (n = 49). Results: A second-order polynomial regression model was found to be the best descriptor of biometric measures including brain bi-parietal diameter, head circumference, and fronto-occipital diameter in relation to normal fetal growth. Normal fetuses showed divergent growth patterns for the cerebrum and cerebellum, where the cerebrum undergoes rapid growth in the second trimester, while the cerebellum undergoes rapid growth in the third trimester. Moreover, a linear model based on biometrics of brain bi-parietal diameter, length of the corpus callosum, vermis area, transverse cerebellar diameter, and cerebellar area accurately predicted gestational age in the second and third trimesters (cross-validation R2 = 0.822, p 14 weeks’ gestation which was determined with consistent estimation both by last menstrual period and ultrasonography dating (termed ‘measured GA’ here); (2) no history of exposure to risk factors or drug abuse during pregnancy; (3) no abnormality in structural brain anatomy in MRI. Exclusion criteria were: (1) delivery complications, congenital malformations or maternal infection, chromosomal abnormality, inadequate MRI image quality; (2) claustrophobia or contraindications to MRI. An independent held-out cohort comprising of 65 (referred to as validation dataset) normal fetuses aged from 23–38 weeks of gestation with MRI examination conducted between November 2017 and January 2018 was included as a validation dataset. This study was approved by the Institutional Review Board of Fujian Maternity and Child Health Hospital, and written informed consent for participation was obtained from all participants. Image acquisition
Fetal MRI data were collected using a 1.5 T Signa (General Electric Medical Systems, Milwaukee, WI) wholebody MRI scanner with an 8-channel phased array body coil. The mother was positioned feet-first into the scanner without sedation, and was instructed to stay as relaxed
Data Loading...
