Haghpanah, F. S. | bioRxiv Science

bioRxiv2021

ID-Seg: An Accurate and Reliable Infant Deep learning Segmentation Framework for Limbic Structures

Early postnatal period brain magnetic resonance imaging (MRI) is becoming an important approach to measure the impact of prenatal exposures on neurodevelopment and to investigate early biomarkers for risk. Among brain structures, Limbic structures are particular of interest in psychiatric disorder-related research. However, despite the promise of infant neuroimaging and the success of initial infant MRI studies, assessing limbic regions structure and function remains a significant challenge due to low inter-regional intensity contrast and high curvature (e.g., hippocampus). In addition, the agreement between existing automatic techniques and manual segmentation remains either untested or insufficient, particularly for the amygdala and hippocampus. In this work, we developed an accurate (based on three segmentation evaluation metrics), reliable and efficient infant deep learning segmentation framework (ID-Seg) to address the aforementioned challenges. Specifically, we leveraged a large dataset of 473 infant MRI scans to train ID-Seg and rigorously evaluated ID-Segs performance on internal and external datasets with manual segmentations. Compared with a state-of-the-art segmentation pipeline, we demonstrated that ID-Seg significantly improved the segmentation accuracy of limbic structures (hippocampus and amygdala) in newborn infants. Moreover, in a medium-size dataset, we found that ID-Seg-derived morphometric measures yield strong brain-behavior associations. As such, our ID-Seg may improve our capacity and efficiency to measure MRI-based brain features relevant to neuropsychological development and ultimately advance the success of quantitative analyses on large-scale datasets.

bioengineering↗

bioRxiv2020

A transfer-learning approach for first-year developmental infant brain segmentation using deep neural networks

The months between birth and age 2 are increasingly recognized as a period critical for neuro-development, with potentially life-long implications for cognitive functioning. However, little is known about the growth trajectories of brain structure and function across this time period. This is in large part because of insufficient approaches to analyze infant MRI scans at different months, especially brain segmentation. Addressing technical gaps in infant brain segmentation would significantly improve our capacity to efficiently measure and identify relevant infant brain structures and connectivity, and their role in long-term development. In this paper, we propose a transfer-learning approach based on convolutional neural network (CNN)-based image segmentation architecture, QuickNAT, to segment brain structures for newborns and 6-month infants separately. We pre-trained QuickNAT on auxiliary labels from a large-scale dataset, fine-tuned on manual labels, and then cross-validated the models performance on two separate datasets. Compared to other commonly used methods, our transfer-learning approach showed superior segmentation performance on both newborns and 6-month infants. Moreover, we demonstrated improved hippocampus segmentation performance via our approach in preterm infants.

neuroscience↗