Can we predict sleep health based on brain features? A large-scale machine learning study DOI Creative Commons
Federico Raimondo, Hanwen Bi, Vera Komeyer

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 13, 2024

Abstract Objectives Normal sleep is crucial for brain health. Recent studies have reported robust associations between disturbance and various structural functional traits. However, the complex interplay health macro-scale organization remains inconclusive. In this study, we aimed to uncover links imaging features diverse health-related characteristics by means of Machine Learning (ML). Methods We used 28,088 participants from UK Biobank calculate 4677 neuroimaging markers. Then, employed them predict self-reported insomnia symptoms, duration, easiness getting up in morning, chronotype, daily nap, daytime sleepiness, snoring. built seven different linear nonlinear ML models each characteristic assess their predictability. Results performed an extensive analysis that involved more than 100,000 hours computing. observed relatively low performance predicting all (e.g., balanced accuracy ranging 0.50-0.59). Across models, best achieved was 0.59, using a Linear SVM morning. Conclusions The capability multimodal markers characteristics, even under optimization large population sample suggests relationship organization.

Language: Английский

Subtypes of Insomnia Disorder Identified by Cortical Morphometric Similarity Network DOI Creative Commons

Haobo Zhang,

Haonan Sun, Jiaqi Li

et al.

Human Brain Mapping, Journal Year: 2025, Volume and Issue: 46(1)

Published: Jan. 1, 2025

ABSTRACT Insomnia disorder (ID) is a highly heterogeneous psychiatric disease, and the use of neuroanatomical data to objectively define biological subtypes essential. We aimed examine ID by morphometric similarity network (MSN) association between MSN changes specific transcriptional expression patterns. recruited 144 IDs 124 healthy controls (HC). performed heterogeneity through discriminant analysis (HYDRA) identified within strength. Differences in HC were compared, clinical behavioral differences compared subtypes. In addition, we investigated brain gene different using partial least squares regression assess genetic commonalities disorders further functional enrichment analyses. Two distinct identified, each exhibiting HC. Furthermore, subtype 1 characterized objective short sleep, impaired cognitive function, some relationships with major depressive autism spectrum (ASD). contrast, 2 has normal sleep duration but subjectively reports poor only related ASD. The pathogenesis may be genes that regulate rhythms sleep–wake cycles. more due adverse emotion perception regulation. Overall, these findings provide insights into ID, elucidating structural molecular aspects relevant

Language: Английский

Citations

2
Sleep Health and White Matter Integrity in the UK Biobank DOI Creative Commons

Roxana Petri,

Florian Holub, Julian Schiel

et al.

Journal of Sleep Research, Journal Year: 2025, Volume and Issue: unknown

Published: March 12, 2025

ABSTRACT Many people experience impaired sleep health, yet knowledge about its neurobiological correlates is limited. As previous studies have found associations between white matter integrity and several traits, could be causally implicated in poor health. However, these were often limited by small sample sizes. In this study, we examine multiple indices of health 29,114 UK Biobank participants. Late chronotype, daytime sleepiness, insomnia symptoms and, most extensively, long duration independently associated with diffusion MRI markers reduced integrity. Previous findings showing an association decreased fractional anisotropy (FA) the anterior internal capsule not replicated. To our knowledge, current analysis first study to find microstructural assumptions concerning role for are challenged.

Language: Английский

Citations

0
Symptom network analysis of prefrontal seizures DOI Creative Commons
Christophe Gauld, Fabrice Bartoloméi, Jean‐Arthur Micoulaud‐Franchi

et al.

Epilepsia, Journal Year: 2025, Volume and Issue: unknown

Published: March 19, 2025

Abstract Objective Prefrontal seizures pose significant challenges in accurately identifying the complex interactions between clinical manifestations and brain electrophysiological activities. This proof‐of‐concept study aims to propose a new approach rigorously support electroclinical reasoning field of epilepsy. Methods We analyzed stereoelectroencephalographic data from 42 patients with drug‐resistant focal epilepsy, whose involved prefrontal cortex at seizure onset. Semiological activities features were scored by expert observers. performed symptom network analysis semiological feature hybrid analysis, coupling ictal Centrality measures used identify most influential networks. Results Our identified impairment consciousness as central network. In network, anterior cingulate area (here incorporating Brodmann [BA]‐32 and/or rostral part BA‐24) emerged activity feature. Significance By integrating into networks, offers an effective quantitative tool for examining relationships semiology correlates seizures. provides opportunity advance novel investigate intricacies correlations, sustaining development dynamic models, on different series epilepsies, larger cohorts, automatically extracted artificial intelligence, that better reflect temporal spatial complexities propagation

Language: Английский

Citations

0
Distinct Convergent Brain Alterations in Sleep Disorders and Sleep Deprivation DOI

Gerion Michael Reimann,

Alireza Hoseini, Mihrican Koçak

et al.

JAMA Psychiatry, Journal Year: 2025, Volume and Issue: unknown

Published: April 23, 2025

Importance Sleep disorders have different etiologies yet share some nocturnal and daytime symptoms, suggesting common neurobiological substrates; healthy individuals undergoing experimental sleep deprivation also report analogous symptoms. However, brain similarities differences between long-term short-term are unclear. Objective To investigate the shared specific neural correlates across deprivation. Data Sources PubMed, Web of Science, Embase, Scopus, BrainMap were searched up to January 2024 identify relevant structural functional neuroimaging articles. Study Selection Whole-brain articles reporting voxel-based group patients with control participants or total partial sleep-deprived well-rested included. Extraction Synthesis Significant coordinates comparisons, their contrast direction (eg, < controls), imaging modality extracted. For each article, 2 raters independently evaluated eligibility extracted data. Subsequently, several meta-analyses performed revised activation likelihood estimation algorithm using P .05 cluster-level familywise error correction. Main Outcomes Measures Transdiagnostic regional alterations identified among Their associated behavioral functions task-based task-free connectivity patterns explored independent datasets (BrainMap enhanced Nathan Kline Institute–Rockland Sample). Results A 231 (140 unique experiments, 3380 participants) retrieved. The analysis (n = 95 experiments) subgenual anterior cingulate cortex (176 voxels, z score 4.86), reward, reasoning, gustation, amygdala hippocampus (130 4.00), negative emotion processing, memory, olfaction. Both clusters had positive default mode network. right thalamus (153 5.21) emerged as a consistent alteration following 45 experiments). This cluster was thermoregulation, action, pain perception showed subcortical (pre)motor regions. Subanalyses regarding demonstrated that exhibited decreased activation, connectivity, and/or volume, while increased volume. Conclusions Relevance Distinct convergent abnormalities observed (probably reflecting symptoms)

Language: Английский

Citations

0
Neural correlates of insomnia with depression and anxiety from a neuroimaging perspective: A systematic review DOI
Peng Chen, Kai Wang, Jinyu Wang

et al.

Sleep Medicine Reviews, Journal Year: 2025, Volume and Issue: unknown, P. 102093 - 102093

Published: May 1, 2025

Language: Английский

Citations

0
The association between insomnia and cognitive decline: a scoping review DOI

Xiaotu Zhang,

Jiawei Yin,

Xuefeng Sun

et al.

Sleep Medicine, Journal Year: 2024, Volume and Issue: 124, P. 540 - 550

Published: Oct. 17, 2024

Language: Английский

Citations

2
Gene expression is associated with brain function of insomnia disorder, rather than brain structure DOI

Haobo Zhang,

Haonan Sun, Jiatao Li

et al.

Progress in Neuro-Psychopharmacology and Biological Psychiatry, Journal Year: 2024, Volume and Issue: unknown, P. 111209 - 111209

Published: Nov. 1, 2024

Language: Английский

Citations

1
Can we predict sleep health based on brain features? A large-scale machine learning study DOI Creative Commons
Federico Raimondo, Hanwen Bi, Vera Komeyer

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 13, 2024

Abstract Objectives Normal sleep is crucial for brain health. Recent studies have reported robust associations between disturbance and various structural functional traits. However, the complex interplay health macro-scale organization remains inconclusive. In this study, we aimed to uncover links imaging features diverse health-related characteristics by means of Machine Learning (ML). Methods We used 28,088 participants from UK Biobank calculate 4677 neuroimaging markers. Then, employed them predict self-reported insomnia symptoms, duration, easiness getting up in morning, chronotype, daily nap, daytime sleepiness, snoring. built seven different linear nonlinear ML models each characteristic assess their predictability. Results performed an extensive analysis that involved more than 100,000 hours computing. observed relatively low performance predicting all (e.g., balanced accuracy ranging 0.50-0.59). Across models, best achieved was 0.59, using a Linear SVM morning. Conclusions The capability multimodal markers characteristics, even under optimization large population sample suggests relationship organization.

Language: Английский

Citations

0