Low field pediatric brain magnetic resonance Image Segmentation and quality Assurance

Abstract

In early life, an accurate description of the structural changes in the developing human brain is crucial for understanding healthy development and identifying neurodevelopmental disorders. Existing adult brain deep learning (DL) based segmentation tools struggle with pediatric MRI due to poor gray/white matter differentiation and rapid growth. These challenges impair the ability of existing algorithms to accurately segment pediatric brain structures even with high-field (1.5T or 3T) MRI systems. In low and middle-income countries, high-field MRI systems are rare due to their high cost and maintenance requirements. To fill this gap, 0.064T Hyperfine SWOOP scanners are being tested in these settings by the UNITY Consortium, funded by the Gates Foundation. Despite lower image quality, low-field MRI offers portability, cost-effectiveness, and eliminates the need for sedation in children. To continue translating recent improvements in infant brain segmentation to underprivileged communities, we resume the Low-field pediatric brain magnetic resonance Image Segmentation and quality Assurance (LISA) challenge. This year, we will expand the LISA challenge tasks, along with their associated datasets. In the past, Task 1 involved evaluating objective, quantifiable quality assurance (QA) methods that rated the overall quality of low-field MRI to ensure the acquired MRI met specific accuracy and consistency standards. For LISA 2026, we will subdivide this into Task 1a, which will focus again on image artifact evaluation, and Task 1b, which will concentrate on improving image quality through artifact removal or reduction to recover data that would otherwise be substandard. Task 2 has centered around the automatic segmentation of subcortical structures. Previous iterations focused on individual structures, the hippocampi and basal ganglia; however, we are now expanding to a multi-structure segmentation task that encompasses all the deep gray matter structures that can be reliably segmented along with the corpus callosum, and ventricular systems. Specifically, the challenge now includes the hippocampi (critical for memory and cognitive functions), the caudate, putamen and globus pallidus (essential for motor control and executive function), the thalamus (vital for sensory and motor integration), the corpus callosum (the primary white matter pathway for interhemispheric communication), and the lateral ventricles (important markers of brain development and structural integrity).