Individual characteristics outperform resting-state fMRI for the prediction of behavioral phenotypes
Standard
Individual characteristics outperform resting-state fMRI for the prediction of behavioral phenotypes. / Omidvarnia, Amir; Sasse, Leonard; Larabi, Daouia I; Raimondo, Federico; Hoffstaedter, Felix; Kasper, Jan; Dukart, Jürgen; Petersen, Marvin; Cheng, Bastian; Thomalla, Götz; Eickhoff, Simon B; Patil, Kaustubh R.
in: COMMUN BIOL, Jahrgang 7, Nr. 1, 26.06.2024, S. 771.Publikationen: SCORING: Beitrag in Fachzeitschrift/Zeitung › SCORING: Zeitschriftenaufsatz › Forschung › Begutachtung
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Individual characteristics outperform resting-state fMRI for the prediction of behavioral phenotypes
AU - Omidvarnia, Amir
AU - Sasse, Leonard
AU - Larabi, Daouia I
AU - Raimondo, Federico
AU - Hoffstaedter, Felix
AU - Kasper, Jan
AU - Dukart, Jürgen
AU - Petersen, Marvin
AU - Cheng, Bastian
AU - Thomalla, Götz
AU - Eickhoff, Simon B
AU - Patil, Kaustubh R
N1 - © 2024. The Author(s).
PY - 2024/6/26
Y1 - 2024/6/26
N2 - In this study, we aimed to compare imaging-based features of brain function, measured by resting-state fMRI (rsfMRI), with individual characteristics such as age, gender, and total intracranial volume to predict behavioral measures. We developed a machine learning framework based on rsfMRI features in a dataset of 20,000 healthy individuals from the UK Biobank, focusing on temporal complexity and functional connectivity measures. Our analysis across four behavioral phenotypes revealed that both temporal complexity and functional connectivity measures provide comparable predictive performance. However, individual characteristics consistently outperformed rsfMRI features in predictive accuracy, particularly in analyses involving smaller sample sizes. Integrating rsfMRI features with demographic data sometimes enhanced predictive outcomes. The efficacy of different predictive modeling techniques and the choice of brain parcellation atlas were also examined, showing no significant influence on the results. To summarize, while individual characteristics are superior to rsfMRI in predicting behavioral phenotypes, rsfMRI still conveys additional predictive value in the context of machine learning, such as investigating the role of specific brain regions in behavioral phenotypes.
AB - In this study, we aimed to compare imaging-based features of brain function, measured by resting-state fMRI (rsfMRI), with individual characteristics such as age, gender, and total intracranial volume to predict behavioral measures. We developed a machine learning framework based on rsfMRI features in a dataset of 20,000 healthy individuals from the UK Biobank, focusing on temporal complexity and functional connectivity measures. Our analysis across four behavioral phenotypes revealed that both temporal complexity and functional connectivity measures provide comparable predictive performance. However, individual characteristics consistently outperformed rsfMRI features in predictive accuracy, particularly in analyses involving smaller sample sizes. Integrating rsfMRI features with demographic data sometimes enhanced predictive outcomes. The efficacy of different predictive modeling techniques and the choice of brain parcellation atlas were also examined, showing no significant influence on the results. To summarize, while individual characteristics are superior to rsfMRI in predicting behavioral phenotypes, rsfMRI still conveys additional predictive value in the context of machine learning, such as investigating the role of specific brain regions in behavioral phenotypes.
KW - Humans
KW - Magnetic Resonance Imaging/methods
KW - Male
KW - Female
KW - Brain/diagnostic imaging
KW - Phenotype
KW - Middle Aged
KW - Adult
KW - Machine Learning
KW - Aged
KW - Behavior
KW - Rest/physiology
KW - Brain Mapping/methods
U2 - 10.1038/s42003-024-06438-5
DO - 10.1038/s42003-024-06438-5
M3 - SCORING: Journal article
C2 - 38926486
VL - 7
SP - 771
JO - COMMUN BIOL
JF - COMMUN BIOL
SN - 2399-3642
IS - 1
ER -