Learning Strategies for Contrast-agnostic Segmentation via SynthSeg for Infant MRI data

Ziyao Shang, Md Asadullah Turja, Eric Feczko, Audrey Houghton, Amanda Rueter, Lucille A Moore, Kathy Snider, Timothy Hendrickson, Paul Reiners, Sally Stoyell, Omid Kardan, Monica Rosenberg, Jed T Elison, Damien A Fair, Martin Andreas Styner

Keywords: MRI, infant MRI, brain development, segmentation

TL;DR: This is an application paper presenting minor, but crucially needed modifications to the existing SynthSeg segmentation framework to handle infant MRI data throughout the first year of life.

Abstract: 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 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 infant 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 (NN-Unet) 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.

Registration: I acknowledge that publication of this at MIDL and in the proceedings requires at least one of the authors to register and present the work during the conference.

Authorship: I confirm that I am the author of this work and that it has not been submitted to another publication before.

Paper Type: validation/application paper

Primary Subject Area: Segmentation

Secondary Subject Area: Segmentation

Confidentiality And Author Instructions: I read the call for papers and author instructions. I acknowledge that exceeding the page limit and/or altering the latex template can result in desk rejection.

Code And Data:

Code: https://github.com/ZiyaoShang/infant_SynthSeg
Data:
- The template data is located with the Code, the training data itself is generated from the template data and all training data is synthetic
- The testing data is currently not available, as the final touches on the corresponding MRI template database are being done. The template database is expected to be released publicly this spring (and thus should be publicly available by the time of MIDL 2022)