How to establish robust brain–behavior relationships without thousands of individuals

Nature Neuroscience, Год журнала: 2022, Номер 25(7), С. 835 - 837

Опубликована: Июнь 16, 2022

Язык: Английский

---

Multivariate BWAS can be replicable with moderate sample sizes

Nature, Год журнала: 2023, Номер 615(7951), С. E4 - E7

Опубликована: Март 8, 2023

Язык: Английский

---

Interpretable machine learning for dementia: A systematic review

Alzheimer s & Dementia, Год журнала: 2023, Номер 19(5), С. 2135 - 2149

Опубликована: Фев. 3, 2023

Abstract Introduction Machine learning research into automated dementia diagnosis is becoming increasingly popular but so far has had limited clinical impact. A key challenge building robust and generalizable models that generate decisions can be reliably explained. Some are designed to inherently “interpretable,” whereas post hoc “explainability” methods used for other models. Methods Here we sought summarize the state‐of‐the‐art of interpretable machine dementia. Results We identified 92 studies using PubMed, Web Science, Scopus. Studies demonstrate promising classification performance vary in their validation procedures reporting standards rely heavily on data sets. Discussion Future work should incorporate clinicians validate explanation make conclusive inferences about dementia‐related disease pathology. Critically analyzing model explanations also requires an understanding interpretability itself. Patient‐specific required benefit practice.

Язык: Английский

---

Brain structure-function coupling provides signatures for task decoding and individual fingerprinting

NeuroImage, Год журнала: 2022, Номер 250, С. 118970 - 118970

Опубликована: Фев. 4, 2022

Brain signatures of functional activity have shown promising results in both decoding brain states, meaning distinguishing between different tasks, and fingerprinting, that is identifying individuals within a large group. Importantly, these do not account for the underlying anatomy on which function takes place. Structure-function coupling based graph signal processing (GSP) has recently revealed meaningful spatial gradient from unimodal to transmodal regions, average healthy subjects during resting-state. Here, we explore specificity structure-function distinct states (tasks) individual subjects. We used multimodal magnetic resonance imaging 100 unrelated Human Connectome Project rest seven tasks adopted support vector machine classification approach with various cross-validation settings. found measures allow accurate classifications task fingerprinting. In particular, key information fingerprinting more liberal portion signals, contributions strikingly localized fronto-parietal network. Moreover, signals showed strong correlation cognitive traits, assessed partial least square analysis, corroborating its relevance By introducing new perspective GSP-based filtering FC decomposition, show provides class cognition organization at tasks. Further, they provide insights clarifying role low high frequencies structural connectome, leading understanding where characterizing can be across connectome spectrum.

Язык: Английский

---

Improving power in functional magnetic resonance imaging by moving beyond cluster-level inference

Proceedings of the National Academy of Sciences, Год журнала: 2022, Номер 119(32)

Опубликована: Авг. 4, 2022

Inference in neuroimaging typically occurs at the level of focal brain areas or circuits. Yet, increasingly, well-powered studies paint a much richer picture broad-scale effects distributed throughout brain, suggesting that many reports may only reflect tip iceberg underlying effects. How versus perspectives influence inferences we make has not yet been comprehensively evaluated using real data. Here, compare sensitivity and specificity across procedures representing multiple levels inference an empirical benchmarking procedure resamples task-based connectomes from Human Connectome Project dataset (∼1,000 subjects, 7 tasks, 3 resampling group sizes, inferential procedures). Only (network whole brain) obtained traditional 80% statistical power to detect average effect, reflecting >20% more than (edge cluster) procedures. Power also increased substantially for false discovery rate- compared with familywise error rate-controlling The downsides are fairly limited; loss FDR was relatively modest gains power. Furthermore, methods introduce simple, fast, easy use, providing straightforward starting point researchers. This points promise sophisticated functional connectivity but related fields, including activation. Altogether, this work demonstrates shifting scale choosing control both immediately attainable can help remedy issues plaguing typical field.

Язык: Английский

---

The Burden of Reliability: How Measurement Noise Limits Brain-Behaviour Predictions

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2023, Номер unknown

Опубликована: Фев. 10, 2023

Abstract Major efforts in human neuroimaging strive to understand individual differences and find biomarkers for clinical applications by predicting behavioural phenotypes from brain imaging data. An essential prerequisite identifying generalizable replicable brain-behaviour prediction models is sufficient measurement reliability. However, the selection of targets predominantly guided scientific interest or data availability rather than reliability considerations. Here we demonstrate impact low phenotypic on out-of-sample performance. Using simulated empirical Human Connectome Projects, found that levels common across many can markedly limit ability link behaviour. Next, using 5000 subjects UK Biobank, show only highly reliable fully benefit increasing sample sizes hundreds thousands participants. Overall, our findings highlight importance brain–behaviour associations differences.

Язык: Английский

---

Relationship between prediction accuracy and feature importance reliability: An empirical and theoretical study

NeuroImage, Год журнала: 2023, Номер 274, С. 120115 - 120115

Опубликована: Апрель 23, 2023

There is significant interest in using neuroimaging data to predict behavior. The predictive models are often interpreted by the computation of feature importance, which quantifies relevance an imaging feature. Tian and Zalesky (2021) suggest that importance estimates exhibit low split-half reliability, as well a trade-off between prediction accuracy reliability across parcellation resolutions. However, it unclear whether universal. Here, we demonstrate that, with sufficient sample size, (operationalized Haufe-transformed weights) can achieve fair excellent reliability. With size 2600 participants, weights average intra-class correlation coefficients 0.75, 0.57 0.53 for cognitive, personality mental health measures respectively. much more reliable than original regression univariate FC-behavior correlations. Original not even participants. Intriguingly, strongly positively correlated phenotypes. Within particular behavioral domain, there no clear relationship performance models. Furthermore, show mathematically necessary, but sufficient, error. In case linear models, lower error related Therefore, higher might yield accuracy. Finally, discuss how our theoretical results relate features measures. Overall, current study provides empirical insights into

Язык: Английский

---

The challenges and prospects of brain-based prediction of behaviour

Nature Human Behaviour, Год журнала: 2023, Номер 7(8), С. 1255 - 1264

Опубликована: Июль 31, 2023

Язык: Английский

---

MRI economics: Balancing sample size and scan duration in brain wide association studies

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Фев. 18, 2024

Abstract A pervasive dilemma in neuroimaging is whether to prioritize sample size or scan time given fixed resources. Here, we systematically investigate this trade-off the context of brain-wide association studies (BWAS) using functional magnetic resonance imaging (fMRI). We find that total duration (sample × per participant) robustly explains individual-level phenotypic prediction accuracy via a logarithmic model, suggesting and are broadly interchangeable up 20-30 min data. However, returns diminish relative size, which explain with principled theoretical derivations. When accounting for overhead costs associated each participant (e.g., recruitment, non-imaging measures), many small-scale some large-scale BWAS might benefit from longer than typically assumed. These results generalize across domains, scanners, acquisition protocols, racial groups, mental disorders, age as well resting-state task-state connectivity. Overall, our study emphasizes importance time, ignored standard power calculations. Standard calculations maximize at expense can result sub-optimal accuracies inefficient use Our empirically informed reference available future design: WEB_APPLICATION_LINK

Язык: Английский

---

Functional brain networks are associated with both sex and gender in children

Science Advances, Год журнала: 2024, Номер 10(28)

Опубликована: Июль 12, 2024

Sex and gender are associated with human behavior throughout the life span across health disease, but whether they similar or distinct neural phenotypes is unknown. Here, we demonstrate that, in children, sex uniquely reflected intrinsic functional connectivity of brain. Somatomotor, visual, control, limbic networks preferentially sex, while network correlates more distributed cortex. These results suggest that irreducible to one another not only society also biology.

Язык: Английский

---