ID-Seg: an infant deep learning-based segmentation framework to improve limbic structure estimates.
Yun Wang, Fateme Sadat Haghpanah, Xuzhe Zhang, Katie Santamaria, Gabriela Koch da Costa Aguiar Alves, Elizabeth Bruno, Natalie Aw, Alexis Maddocks, Cristiane S Duarte, Catherine Monk, Andrew Laine, Jonathan Posner, program collaborators for Environmental influences on Child Health Outcomes
Author Information
Yun Wang: Department of Psychiatry and Behavioral Sciences, Duke University, Durham, NC, USA.
Fateme Sadat Haghpanah: Department of Computer Science, University Of Toronto, Toronto, ON, Canada.
Xuzhe Zhang: Department of Biomedical Engineering, Columbia University, New York, NY, USA.
Katie Santamaria: New York State Psychiatric Institute, New York, NY, USA.
Gabriela Koch da Costa Aguiar Alves: New York State Psychiatric Institute, New York, NY, USA.
Elizabeth Bruno: New York State Psychiatric Institute, New York, NY, USA.
Natalie Aw: New York State Psychiatric Institute, New York, NY, USA.
Alexis Maddocks: Department of Radiology, Columbia University, New York, NY, USA.
Cristiane S Duarte: New York State Psychiatric Institute, New York, NY, USA.
Catherine Monk: New York State Psychiatric Institute, New York, NY, USA.
Andrew Laine: Department of Biomedical Engineering, Columbia University, New York, NY, USA.
Jonathan Posner: Department of Psychiatry and Behavioral Sciences, Duke University, Durham, NC, USA. Jonathan.Posner@duke.edu.
Infant brain magnetic resonance imaging (MRI) is a promising approach for studying early neurodevelopment. However, segmenting small regions such as limbic structures is challenging due to their low inter-regional contrast and high curvature. MRI studies of the adult brain have successfully applied deep learning techniques to segment limbic structures, and similar deep learning models are being leveraged for infant studies. However, these deep learning-based infant MRI segmentation models have generally been derived from small datasets, and may suffer from generalization problems. Moreover, the accuracy of segmentations derived from these deep learning models relative to more standard Expectation-Maximization approaches has not been characterized. To address these challenges, we leveraged a large, public infant MRI dataset (n���=���473) and the transfer-learning technique to first pre-train a deep convolutional neural network model on two limbic structures: amygdala and hippocampus. Then we used a leave-one-out cross-validation strategy to fine-tune the pre-trained model and evaluated it separately on two independent datasets with manual labels. We term this new approach the Infant Deep learning SEGmentation Framework (ID-Seg). ID-Seg performed well on both datasets with a mean dice similarity score (DSC) of 0.87, a mean intra-class correlation (ICC) of 0.93, and a mean average surface distance (ASD) of 0.31 mm. Compared to the Developmental Human Connectome pipeline (dHCP) pipeline, ID-Seg significantly improved segmentation accuracy. In a third infant MRI dataset (n���=���50), we used ID-Seg and dHCP separately to estimate amygdala and hippocampus volumes and shapes. The estimates derived from ID-seg, relative to those from the dHCP, showed stronger associations with behavioral problems assessed in these infants at age 2. In sum, ID-Seg consistently performed well on two different datasets with an 0.87 DSC, however, multi-site testing and extension for brain regions beyond the amygdala and hippocampus are still needed.
