Back

Aperiodic parameters of the fMRI power spectrum associate with preterm birth and neonatal age

Suuronen, I.; Luotonen, S.; Railo, H.; Airola, A.; Bano, W.; Merisaari, H.; Pulli, E. P.; Mariani Wigley, I. L. C.; Vartiainen, E.; Hashempour, N.; Karlsson, H.; Karlsson, L.; Kringelbach, M. L.; Batalle, D.; Tuulari, J. J.

2024-12-16 radiology and imaging
10.1101/2024.12.16.24318785 medRxiv
Show abstract

While strong associations of structural magnetic resonance imaging (sMRI) with preterm birth and post-menstrual age (PMA) have been reported, such associations for functional MRI (fMRI) have been considerably weaker. We studied the associations of the aperiodic parameters of neonatal fMRI Blood-oxygen-level-dependent (BOLD) signal power spectrum with preterm birth and PMA at scan using task-free fMRI data from the Developing Human Connectome Project (dHCP). First, we studied the associations of the aperiodic parameters of the BOLD signal from pre- and postcentral gyri with preterm birth and mapped the associations with PMA, postnatal age, and sex. Second, we used machine learning regression to predict PMA and postnatal age with 90 cortical and subcortical regions of interest (ROIs). We found clear differences between preterm and full-term groups, as well as males and females. Furthermore, our machine learning model predicted the age variables with relatively high accuracy (mean test R{superscript 2} = 0.20 - 0.41).

Published in Communications Biology (predicted rank #11) · training set

Matching journals

The top 4 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.