Multi-scale asynchronous correlation and 2D convolutional autoencoder for adolescent health risk prediction with limited fMRI data DOI Creative Commons
Di Gao,

Guanghao Yang,

Jiarun Shen

et al.

Frontiers in Computational Neuroscience, Journal Year: 2024, Volume and Issue: 18

Published: Oct. 15, 2024

Introduction Adolescence is a fundamental period of transformation, encompassing extensive physical, psychological, and behavioral changes. Effective health risk assessment during this stage crucial for timely intervention, yet traditional methodologies often fail to accurately predict mental risks due the intricacy neural dynamics scarcity quality-annotated fMRI datasets. Methods This study introduces an innovative deep learning-based framework in adolescents by employing combination two-dimensional convolutional autoencoder (2DCNN-AE) with multi-sequence learning multi-scale asynchronous correlation information extraction techniques. approach facilitates intricate analysis spatial temporal features within data, aiming enhance accuracy process. Results Upon examination using Adolescent Risk Behavior (AHRB) dataset, which includes scans from 174 individuals aged 17–22, proposed methodology exhibited significant improvement over conventional models. It attained precision 83.116%, recall 84.784%, F1-score 83.942%, surpassing standard benchmarks most pertinent evaluative measures. Discussion The results underscore superior performance understanding predicting health-related adolescents. underscores value advancing assessments, offering enhanced tool early detection potential intervention strategies sensitive developmental stage.

Language: Английский

Reliability and stability challenges in ABCD task fMRI data DOI Creative Commons

James T. Kennedy,

Michael P. Harms,

Ozlem Korucuoglu

et al.

NeuroImage, Journal Year: 2022, Volume and Issue: 252, P. 119046 - 119046

Published: March 1, 2022

Trait stability of measures is an essential requirement for individual differences research. Functional MRI has been increasingly used in studies that rely on the assumption trait stability, such as attempts to relate task related brain activation behavior and psychopathology. However, recent research using adult samples questioned task-fMRI measures, assessed by test-retest correlations. To date, little known about fMRI children. Here, we examined within-session reliability long-term provided adolescent cognitive development (ABCD) Study Release v4.0 individual's average regional activity, its tasks focused reward processing, response inhibition, working memory. We also evaluated effects factors potentially affecting stability. Reliability (quantified ratio non-scanner stable variance all variances) was poor virtually regions, with value 0.088 0.072 short term (within-session) (between-session) respectively, regions interest (ROIs) historically-recruited tasks. Only one or ROIs exceeded 'poor' cut-off 0.4, fact rarely 0.2 (only 4.9%). Motion had a pronounced effect estimated reliability/stability, lowest motion quartile participants having mean reliability/stability 2.5 times higher (albeit still 'poor') than highest quartile. Poor task-fMRI, particularly children, diminishes potential utility data due drastic reduction sizes and, consequently, statistical power detection brain-behavior associations. This issue urgently needs be addressed through optimization design, scanning parameters, acquisition protocols, preprocessing pipelines, denoising methods.

Language: Английский

Citations

77
From Classical Methods to Generative Models: Tackling the Unreliability of Neuroscientific Measures in Mental Health Research DOI
Nathaniel Haines, Holly Sullivan‐Toole, Thomas M. Olino

et al.

Biological Psychiatry Cognitive Neuroscience and Neuroimaging, Journal Year: 2023, Volume and Issue: 8(8), P. 822 - 831

Published: Jan. 11, 2023

Language: Английский

Citations

23
Annual Research Review: Puberty and the development of anhedonia – considering childhood adversity and inflammation DOI
Tina Gupta, Kristen Eckstrand, Erika E. Forbes

et al.

Journal of Child Psychology and Psychiatry, Journal Year: 2024, Volume and Issue: 65(4), P. 459 - 480

Published: Feb. 23, 2024

Anhedonia, or diminished pleasure and motivation, is a symptom of severe mental illness (e.g., depressive disorder, bipolar schizophrenia) that emerges during adolescence. Anhedonia pernicious related to social impairments, treatment resistance, suicide. As the mechanisms anhedonia are postulated include frontostriatal circuitry dopamine neuromodulatory system, development plasticity these systems vulnerable period adolescence, as well their sensitivity pubertal hormones, suggest maturation could play role in anhedonia. This review takes developmental perspective, considering possibility context adolescent development, with childhood adversity chronic inflammation influencing neural reward accelerate anhedonia's progression. Here, we relevant extant literature on components this model directions for future research.

Language: Английский

Citations

7
Peer Connectedness and Pre‐Existing Social Reward Processing Predicts U.S. Adolescent Girls’ Suicidal Ideation During COVID‐19 DOI Open Access
Emily Hutchinson, Stefanie Sequeira, Jennifer S. Silk

et al.

Journal of Research on Adolescence, Journal Year: 2021, Volume and Issue: 31(3), P. 703 - 716

Published: Aug. 26, 2021

There is major concern about the impact of COVID-19 pandemic on adolescent suicidal ideation (SI) and peer relationships. We investigated (1) rates SI (2) extent to which connectedness pre-existing neural activation social reward predicted during initial stay-at-home orders (April-May 2020) in a longitudinal sample girls (N = 93; M

Language: Английский

Citations

37
Longitudinal Stability of Mood-Related Resting-State Networks in Youth with Symptomatic Bipolar-I/II Disorder DOI Open Access
Danella Hafeman, Jamie Feldman,

Jessica Mak

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 13, 2025

ABSTRACT Bipolar disorder (BD) is characterized by temporal instability of mood and energy, but the neural correlates this are poorly understood. In previous cross-sectional studies, state in BD has been associated with differential functional connectivity (FC) amongst several subcortical regions ventromedial prefrontal cortex. Here, we assess whether longitudinal within mood-related network interest (NOI). Young people BD-I/II were scanned 4-6 times healthy controls (HC) 4 over 9 months. Following preprocessing 20-minute resting-state scans, assessed across-scan correlation FC, focusing on FC between previously state. Utilizing Bayesian models, relationship diagnostic group within-person, correlation, adjusting for motion, time-of-day, inter-scan interval; prediction intervals (PI) reported. a sample 16 youth (11 BD, 5 HC; 16.3-23.3 years old) 70 scans (50 20 HC), NOI stability was higher within-than between-person (0.70 vs. 0.54; p<.0001). (vs. HC) within-person scan-pairs showed lower (mean -0.109; 95% PI -0.181, -0.038), distinguishing HC excellent accuracy (AUC=0.95). more pronounced manic symptoms -0.012; -0.023, -0.0002) BD-II BD-I; mean -0.071; 90% -0.136, -0.007). Results persisted after accounting medications, comorbidity, sleep/arousal measures. Within pilot sample, less NOI. While preliminary, these results highlight possible role precision imaging approaches to elucidate mechanisms underlying BD.

Language: Английский

Citations

0
Relevance of Probabilistic Reversal Learning for Adolescent Drinking Trajectories DOI Creative Commons
Juliane H. Fröhner, Maria Waltmann, Andrea M.F. Reiter

et al.

Addiction Biology, Journal Year: 2025, Volume and Issue: 30(3)

Published: March 1, 2025

One of the many human capabilities acquired during adolescence is adaptivity in changing environments. In this longitudinal study, we investigated adaptivity, as measured by probabilistic reversal learning (PReL) tasks, N = 143 adolescents at ages 14, 16 and 18. Computational modelling functional magnetic resonance imaging were applied to identify neurocognitive processes underlying its development. Previous studies have demonstrated a correlation between heavy alcohol use impaired learning. Our hypothesis was that PReL negatively associated with current future impairs altering processes. Behaviourally, performance improved, which lower probability switching choices considered an adaptive process. Computationally, accounted for higher rates, enhanced sensitivity wins reduced losses older adolescents. Alcohol consumption increased but remained low level most participants. More risky drinking less medial frontal activity elicited reward prediction errors. These findings suggest may be more relevant maintenance or escalation than low-level drinking. Challenges potential solutions such reliability are discussed.

Language: Английский

Citations

0
Impact of clinical factors on reproducibility of dopamine transporter imaging: A 99mTc-TRODAT SPECT study with sufficient size DOI
Kai‐Chun Yang, Yuan-Hwa Chou, Bang‐Hung Yang

et al.

Progress in Neuro-Psychopharmacology and Biological Psychiatry, Journal Year: 2025, Volume and Issue: 138, P. 111370 - 111370

Published: April 1, 2025

Language: Английский

Citations

0
Impact of analytic decisions on test-retest reliability of individual and group estimates in functional magnetic resonance imaging: a multiverse analysis using the monetary incentive delay task DOI Creative Commons
Michael I. Demidenko, Jeanette A. Mumford, Russell A. Poldrack

et al.

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

Published: March 20, 2024

Empirical studies reporting low test-retest reliability of individual blood oxygen-level dependent (BOLD) signal estimates in functional magnetic resonance imaging (fMRI) data have resurrected interest among cognitive neuroscientists methods that may improve fMRI. Over the last decade, several reported modeling decisions, such as smoothing, motion correction and contrast selection, BOLD estimates. However, it remains an empirical question whether certain analytic decisions

Language: Английский

Citations

3
Anhedonia in adolescents at transdiagnostic familial risk for severe mental illness: Clustering by symptoms and mechanisms of association with behavior DOI Creative Commons
Tanu Gupta, Kristen Eckstrand,

Carly Lenniger

et al.

Journal of Affective Disorders, Journal Year: 2023, Volume and Issue: 347, P. 249 - 261

Published: Nov. 21, 2023

Language: Английский

Citations

8
Behavioral and Neural Trajectories of Risk Taking for Peer and Parent in Adolescence DOI
Seh‐Joo Kwon, Jessica Flannery, Caitlin C. Turpyn

et al.

Journal of Cognitive Neuroscience, Journal Year: 2023, Volume and Issue: 35(5), P. 802 - 815

Published: Jan. 1, 2023

Abstract One feature of adolescence is a rise in risk-taking behaviors, whereby the consequences adolescents' risky action often impact their immediate surrounding such as peers and parents (vicarious risk taking). Yet, little known about how vicarious taking develops, particularly depending on who affects type behavior. In 3-wave longitudinal fMRI study, 173 adolescents completed 1–3 years decision-making task where they took risks to win money for best friend parent (n with behavioral data ranges from 139–144 100–116 participants, respectively, per wave). Results this preregistered study suggest that did not differentially take adaptive (sensitivity expected value reward during taking) general (decision-making when values staying safe are equivalent) sixth ninth grade. At neural level, ROI analyses revealed no differences ventral striatum ventromedial pFC nor versus over time. Furthermore, exploratory whole-brain subthreshold between trajectories within regulatory regions social-cognitive taking. Our findings demonstrate brain implicated cognitive control processes may distinguish behaviors involving

Language: Английский

Citations

6