Data leakage inflates prediction performance in connectome-based machine learning models

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Feb. 28, 2024

Abstract Predictive modeling is a central technique in neuroimaging to identify brain-behavior relationships and test their generalizability unseen data. However, data leakage undermines the validity of predictive models by breaching separation between training Leakage always an incorrect practice but still pervasive machine learning. Understanding its effects on can inform how affects existing literature. Here, we investigate five forms leakage–involving feature selection, covariate correction, dependence subjects–on functional structural connectome-based learning across four datasets three phenotypes. via selection repeated subjects drastically inflates prediction performance, whereas other have minor effects. Furthermore, small exacerbate leakage. Overall, our results illustrate variable underscore importance avoiding improve reproducibility modeling.

Language: Английский

---

Functional magnetic resonance imaging in schizophrenia: current evidence, methodological advances, limitations and future directions

World Psychiatry, Journal Year: 2024, Volume and Issue: 23(1), P. 26 - 51

Published: Jan. 12, 2024

Functional neuroimaging emerged with great promise and has provided fundamental insights into the neurobiology of schizophrenia. However, it faced challenges criticisms, most notably a lack clinical translation. This paper provides comprehensive review critical summary literature on functional neuroimaging, in particular magnetic resonance imaging (fMRI), We begin by reviewing research fMRI biomarkers schizophrenia high risk phase through historical lens, moving from case-control regional brain activation to global connectivity advanced analytical approaches, more recent machine learning algorithms identify predictive features. Findings studies negative symptoms as well neurocognitive social cognitive deficits are then reviewed. neural markers these may represent promising treatment targets Next, we summarize related antipsychotic medication, psychotherapy psychosocial interventions, neurostimulation, including response resistance, therapeutic mechanisms, targeting. also utility data-driven approaches dissect heterogeneity schizophrenia, beyond comparisons, methodological considerations advances, consortia precision fMRI. Lastly, limitations future directions field discussed. Our suggests that, order for be clinically useful care patients should address potentially actionable decisions that routine treatment, such which prescribed or whether given patient is likely have persistent impairment. The potential influenced must weighed against cost accessibility factors. Future evaluations prognostic consider health economics analysis.

Language: Английский

---

Affording reusable data: recommendations for researchers from a data-intensive project

Scientific Data, Journal Year: 2025, Volume and Issue: 12(1)

Published: Feb. 12, 2025

Scientists are increasingly required by funding agencies, publishers and their institutions to produce publish data that Findable, Accessible, Interoperable Reusable (FAIR). This requires curatorial activities, which expensive in terms of both time effort. Based on our experience supporting a multidisciplinary research team, we provide recommendations direct the efforts researchers towards affordable ways achieve reasonable degree "FAIRness" for become reusable upon its publication. The accompanied concrete insights challenges faced when trying implement them an actual data-intensive reference project.

Language: Английский

---

The power of many brains: Catalyzing neuropsychiatric discovery through open neuroimaging data and large-scale collaboration

Science Bulletin, Journal Year: 2024, Volume and Issue: 69(10), P. 1536 - 1555

Published: March 6, 2024

Recent advances in open neuroimaging data are enhancing our comprehension of neuropsychiatric disorders. By pooling images from various cohorts, statistical power has increased, enabling the detection subtle abnormalities and robust associations, fostering new research methods. Global collaborations imaging have furthered knowledge neurobiological foundations brain disorders aided imaging-based prediction for more targeted treatment. Large-scale magnetic resonance initiatives driving innovation analytics supporting generalizable psychiatric studies. We also emphasize significant role big understanding neural mechanisms early identification precise treatment However, challenges such as harmonization across different sites, privacy protection, effective sharing must be addressed. With proper governance science practices, we conclude with a projection how large-scale resources could revolutionize diagnosis, selection, outcome prediction, contributing to optimal health.

Language: Английский

---

A practical guide to EEG hyperscanning in joint action research: from motivation to implementation

Social Cognitive and Affective Neuroscience, Journal Year: 2024, Volume and Issue: 19(1)

Published: Jan. 1, 2024

Developments in cognitive neuroscience have led to the emergence of hyperscanning, simultaneous measurement brain activity from multiple people. Hyperscanning is useful for investigating social cognition, including joint action, because its ability capture neural processes that occur within and between people as they coordinate actions toward a shared goal. Here, we provide practical guide researchers considering using hyperscanning study action seeking avoid frequently raised concerns skeptics. We focus specifically on Electroencephalography (EEG) which widely available optimally suited capturing fine-grained temporal dynamics coordination. Our guidelines cover questions are likely arise when planning project, ranging whether appropriate answering one's research considerations design, dependent variable selection, data analysis visualization. By following clear facilitate careful consideration theoretical implications design choices other methodological decisions, can mitigate interpretability issues maximize benefits paradigms.

Language: Английский

---

Quality over quantity: powering neuroimaging samples in psychiatry

Neuropsychopharmacology, Journal Year: 2024, Volume and Issue: 50(1), P. 58 - 66

Published: June 20, 2024

Language: Английский

---

Interpersonal neural synchrony and mental disorders: unlocking potential pathways for clinical interventions

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

Published: March 11, 2024

Introduction Interpersonal synchronization involves the alignment of behavioral, affective, physiological, and brain states during social interactions. It facilitates empathy, emotion regulation, prosocial commitment. Mental disorders characterized by interaction dysfunction, such as Autism Spectrum Disorder (ASD), Reactive Attachment (RAD), Social Anxiety (SAD), often exhibit atypical with others across multiple levels. With introduction “second-person” neuroscience perspective, our understanding interpersonal neural (INS) has improved, however, so far, it hardly impacted development novel therapeutic interventions. Methods To evaluate potential INS-based treatments for mental disorders, we performed two systematic literature searches identifying studies that directly target INS through neurofeedback (12 publications; 9 independent studies) or stimulation techniques (7 studies), following PRISMA guidelines. In addition, narratively review indirect manipulations biofeedback, hormonal We discuss ASD, RAD, SAD using a database search assess acceptability (4 neurostimulation in patients dysfunction. Results Although behavioral approaches, engaging eye contact cooperative actions, have been shown to be associated increased INS, little is known about long-term consequences Few proof-of-concept utilized techniques, like transcranial direct current neurofeedback, showing feasibility preliminary evidence interventions can boost synchrony connectedness. Yet, optimal protocols parameters are still undefined. For SAD, far no randomized controlled trial proven efficacy intervention although general methods seem well accepted these patient groups. Discussion Significant work remains translate into effective disorders. Future research should focus on mechanistic insights technological advancements, rigorous design standards. Furthermore, will key compare targeting those other modalities define dyads clinical

Language: Английский

---

Connectome-based fingerprinting: reproducibility, precision, and behavioral prediction

Neuropsychopharmacology, Journal Year: 2024, Volume and Issue: 50(1), P. 114 - 123

Published: Aug. 15, 2024

Language: Английский

---

Reliability of Mental Workload Index Assessed by EEG with Different Electrode Configurations and Signal Pre-Processing Pipelines

Sensors, Journal Year: 2023, Volume and Issue: 23(3), P. 1367 - 1367

Published: Jan. 26, 2023

Background and Objective: Mental workload (MWL) is a relevant construct involved in all cognitively demanding activities, its assessment an important goal many research fields. This paper aims at evaluating the reproducibility sensitivity of MWL from EEG signals considering effects different electrode configurations pre-processing pipelines (PPPs). Methods: Thirteen young healthy adults were enrolled asked to perform 45 min Simon’s task elicit cognitive demand. data collected using 32-channel system with (fronto-parietal; Fz Pz; Cz) analyzed PPPs, simplest bandpass filtering combination filtering, Artifact Subspace Reconstruction (ASR) Independent Component Analysis (ICA). The indexes estimation their changes assessed Intraclass Correlation Coefficient statistical analysis. Results: PPPs showed reliability ranging good very most (average consistency > 0.87 average absolute agreement 0.92). Larger fronto-parietal configurations, albeit being more affected by choice provide better detection if compared single-electrode configuration (18 vs. 10 statistically significant differences detected, respectively). Conclusions: complex have been proven ensure (>0.90) experimental conditions. In conclusion, we propose use least two-electrode (Fz Pz) including ICA algorithm (even ASR) mitigate artifacts obtain reliable sensitive during tasks.

Language: Английский

---

Genetic and brain similarity independently predict childhood anthropometrics and neighborhood socioeconomic conditions

Developmental Cognitive Neuroscience, Journal Year: 2024, Volume and Issue: 65, P. 101339 - 101339

Published: Jan. 4, 2024

Linking the developing brain with individual differences in clinical and demographic traits is challenging due to substantial interindividual heterogeneity of anatomy organization. Here we employ an integrative approach that parses both cortical thickness common genetic variants, assess their effects on a wide set childhood traits. The uses linear mixed model framework obtain unique each type similarity, as well covariance. We this sample 7760 unrelated children ABCD cohort baseline (mean age 9.9, 46.8% female). In general, associations between similarity were limited anthropometrics such height, weight, birth marker neighborhood socioeconomic conditions. Common variants explained significant proportions variance across nearly all included outcomes, although estimates somewhat lower than previous reports. No covariance was found. present findings highlight connection conditions brain, which appear be independent from population-based sample.

Language: Английский

---