Back

Normative trajectories of extra-axial cerebrospinal fluid during childhood and adolescence defined in a clinically-acquired MRI dataset

Mandal, A. S.; Dorfschmidt, L.; Schabdach, J. M.; Gardner, M.; Yerys, B. E.; Bethlehem, R. A. I.; Sotardi, S.; Henry, M. K.; Wood, J. N.; Chaiyachati, B. H.; Alexander-Bloch, A.; Seidlitz, J.

2024-09-26 neurology
10.1101/2024.09.24.24314251 medRxiv
Show abstract

BackgroundExtra-axial cerebrospinal fluid (eaCSF) refers to the CSF in the subarachnoid spaces that surrounds the brain parenchyma. Benign enlargement of the subarachnoid space (BESS), a condition marked by increased eaCSF thickness, has been associated with macrocephaly and may be associated with subdural collections. However, diagnosis of BESS is complicated by the lack of age-specific normative data which hinders rigorous investigation of its clinical associations. Growth charts of eaCSF could shed light on normal CSF dynamics while also providing a normative benchmark to assist the diagnosis of BESS and other associated conditions. MethodsWe accessed clinically-acquired T1w MRI scans from 1226 pediatric patients to form a clinical control cohort. Nine scans from subjects with a diagnosis of BESS from a board-certified pediatric neuroradiologist were also reviewed. SynthSeg was used to segment each T1w scan into various tissue types, including eaCSF. Growth charts of eaCSF were modeled using the clinical control cohort. The confirmed BESS cases were then benchmarked against these charts to test the performance of eaCSF growth charts. ResultseaCSF thickness varied nonlinearly with age, steadily decreasing from birth to two years, then trending upwards in early adolescence. Seven of the nine patients with a clinical diagnosis of BESS were above the 97.5th percentile for their age for at least one eaCSF measure. Centile scores were able to distinguish BESS cases from controls with an area under curve (AUC) greater than 0.95. DiscussioneaCSF thickness evolves in a dynamic pattern throughout childhood and adolescence. Patients with BESS can be differentiated from clinical controls using computational measurements of eaCSF thickness paired with normative modeling. Our findings demonstrate the feasibility of computational extraction of eaCSF with a potential point of clinical relevance, delineation of BESS diagnosis. Enhanced understanding of normative eaCSF is critical in further investigations its clinical associations.

Published in Radiology: Artificial Intelligence · not in our set (fewer than 10 published preprints to learn from) · training set

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.