Search

Objectives: Brain segmentation of infant magnetic resonance (MR) images is vitally important in studying developmental mental health and disease. The infant brain undergoes many changes throughout the first years of postnatal life, making tissue segmentation difficult for most existing algorithms. Here, we introduce a deep neural network BIBSNet ( B aby and I nfant B rain S egmentation Neural Net work), an open-source, community-driven model that relies on data augmentation and a large sample size of manually annotated images to facilitate the production of robust and generalizable brain segmentations., Experimental Design: Included in model training and testing were MR brain images on 84 participants with an age range of 0-8 months (median postmenstrual ages of 13.57 months). Using manually annotated real and synthetic segmentation images, the model was trained using a 10-fold cross-validation procedure. Testing occurred on MRI data processed with the DCAN labs infant-ABCD-BIDS processing pipeline using segmentations produced from gold standard manual annotation, joint-label fusion (JLF), and BIBSNet to assess model performance., Principal Observations: Using group analyses, results suggest that cortical metrics produced using BIBSNet segmentations outperforms JLF segmentations. Additionally, when analyzing individual differences, BIBSNet segmentations perform even better., Conclusions: BIBSNet segmentation shows marked improvement over JLF segmentations across all age groups analyzed. The BIBSNet model is 600x faster compared to JLF and can be easily included in other processing pipelines.

Longitudinal studies of infants' brains are essential for research and clinical detection of neurodevelopmental disorders. However, for infant brain MRI scans, effective deep learning-based segmentation frameworks exist only within small age intervals due to the large image intensity and contrast changes that take place in the early postnatal stages of development. However, using different segmentation frameworks or models at different age intervals within the same longitudinal data set would cause segmentation inconsistencies and age-specific biases. Thus, an age-agnostic segmentation model for infants' brains is needed. In this paper, we present "Infant-SynthSeg", an extension of the contrast-agnostic SynthSeg segmentation framework applicable to MRI data of infants at ages within the first year of life. Our work mainly focuses on extending learning strategies related to synthetic data generation and augmentation, with the aim of creating a method that employs training data capturing features unique to infants' brains during this early-stage development. Comparison across different learning strategy settings, as well as a more-traditional contrast-aware deep learning model (nnU-net) are presented. Our experiments show that our trained Infant-SynthSeg models show consistently high segmentation performance on MRI scans of infant brains throughout the first year of life. Furthermore, as the model is trained on ground truth labels at different ages, even labels that are not present at certain ages (such as cerebellar white matter at 1 month) can be appropriately segmented via Infant-SynthSeg across the whole age range. Finally, while Infant-SynthSeg shows consistent segmentation performance across the first year of life, it is outperformed by age-specific deep learning models trained for a specific narrow age range.

Background: We aimed to systematically identify, standardise and disseminate individual-level dietary intake surveys from up to 207 countries for 54 foods, beverages and nutrients, including subnational intakes by age, sex, education and urban/rural residence, from 1980 to 2015., Methods: Between 2008-2011 and 2014-2020, the Global Dietary Database (GDD) project systematically searched for surveys assessing individual-level intake worldwide. We prioritised nationally or subnationally representative surveys using 24-hour recalls, Food-Frequency Questionnaires or short standardised questionnaires. Data were retrieved from websites or corresponding members as individual-level food group microdata or aggregate stratum-level data. Standardisation included quality assessment; data cleaning; categorising of foods and nutrients and their units; aggregation by demographic strata and energy adjustment., Results: We standardised and incorporated 1220 surveys into the final GDD 2017 database, together represented 188 countries and 99.0% of the world's population in 2015. 72.1% were nationally, 17.0% subnationally, and 10.9% community-level representative. 41.2% used Food-Frequency Questionnaires; 23.4%, 24-hour recalls; 15.8%, Demographic Health Survey questionnaires; 13.1%, biomarkers and 6.4%, household surveys. 73.9% of surveys included data on children; 52.2%, by urban and rural residence; and 30.2%, by education. Most surveys were in high-income countries, followed by sub-Saharan Africa and Asia. Most commonly ascertained foods were fruits (N=803 surveys), non-starchy vegetables (N=787) and sugar-sweetened beverages (N=440); and nutrients, sodium (N=343), energy (N=256), calcium (N=224) and fibre (N=200). Least available data were on iodine, vitamin A, plant protein, selenium, added sugar and animal protein., Conclusions: This systematic search, retrieval and standardised effort provides the most comprehensive empirical evidence on dietary intakes across and within countries worldwide., Competing Interests: Competing interests: VM reports a grant from the American Heart Association, outside the submitted work. PW reports research grants and contracts from the US Agency for International Development and UNICEF and personal fees from the Global Panel on Agriculture and Food Systems for Nutrition, outside the submitted work. RM reports grants from National Institute of Health, Nestle, Danone, personal fees from Bunge, Development Initiatives, outside the submitted work. DM reports research funding from the National Institutes of Health and the Bill & Melinda Gates Foundation; personal fees from GOED, Bunge, Indigo Agriculture, Motif FoodWorks, Amarin, Acasti Pharma, Cleveland Clinic Foundation, America’s Test Kitchen and Danone; scientific advisory board member for Brightseed, Day Two, Elysium Health, Filtricine, HumanCo and Tiny Organics; and chapter royalties from UpToDate, all outside the submitted work., (© Author(s) (or their employer(s)) 2021. Re-use permitted under CC BY. Published by BMJ.)