Trust me if you can: a survey on reliability and interpretability of machine learning approaches for drug sensitivity prediction in cancer.

With the ever-increasing number of artificial intelligence (AI) systems, mitigating risks associated with their use has become one of the most urgent scientific and societal issues. To this end, the European Union passed the EU AI Act, proposing solution strategies that can be summarized under the umbrella term trustworthiness. In anti-cancer drug sensitivity prediction, machine learning (ML) methods are developed for application in medical decision support systems, which require an extraordinary level of trustworthiness. This review offers an overview of the ML landscape of methods for anti-cancer drug sensitivity prediction, including a brief introduction to the four major ML realms (supervised, unsupervised, semi-supervised, and reinforcement learning). In particular, we address the question to what extent trustworthiness-related properties, more specifically, interpretability and reliability, have been incorporated into anti-cancer drug sensitivity prediction methods over the previous decade. In total, we analyzed 36 papers with approaches for anti-cancer drug sensitivity prediction. Our results indicate that the need for reliability has hardly been addressed so far. Interpretability, on the other hand, has often been considered for model development. However, the concept is rather used intuitively, lacking clear definitions. Thus, we propose an easily extensible taxonomy for interpretability, unifying all prevalent connotations explicitly or implicitly used within the field.

Dissociation of Reliability, Heritability, and Predictivity in Coarse- and Fine-Scale Functional Connectomes during Development.

The functional connectome supports information transmission through the brain at various spatial scales, from exchange between broad cortical regions to finer-scale, vertex-wise connections that underlie specific information processing mechanisms. In adults, while both the coarse- and fine-scale functional connectomes predict cognition, the fine scale can predict up to twice the variance as the coarse-scale functional connectome. Yet, past brain-wide association studies, particularly using large developmental samples, focus on the coarse connectome to understand the neural underpinnings of individual differences in cognition. Using a large cohort of children (age 9-10 years; n = 1,115 individuals; both sexes; 50% female, including 170 monozygotic and 219 dizygotic twin pairs and 337 unrelated individuals), we examine the reliability, heritability, and behavioral relevance of resting-state functional connectivity computed at different spatial scales. We use connectivity hyperalignment to improve access to reliable fine-scale (vertex-wise) connectivity information and compare the fine-scale connectome with the traditional parcel-wise (coarse scale) functional connectomes. Though individual differences in the fine-scale connectome are more reliable than those in the coarse-scale, they are less heritable. Further, the alignment and scale of connectomes influence their ability to predict behavior, whereby some cognitive traits are equally well predicted by both connectome scales, but other, less heritable cognitive traits are better predicted by the fine-scale connectome. Together, our findings suggest there are dissociable individual differences in information processing represented at different scales of the functional connectome which, in turn, have distinct implications for heritability and cognition.

Rescuing error control in crosslinking mass spectrometry.

Crosslinking mass spectrometry is a powerful tool to study protein-protein interactions under native or near-native conditions in complex mixtures. Through novel search controls, we show how biassing results towards likely correct proteins can subtly undermine error estimation of crosslinks, with significant consequences. Without adjustments to address this issue, we have misidentified an average of 260 interspecies protein-protein interactions across 16 analyses in which we synthetically mixed data of different species, misleadingly suggesting profound biological connections that do not exist. We also demonstrate how data analysis procedures can be tested and refined to restore the integrity of the decoy-false positive relationship, a crucial element for reliably identifying protein-protein interactions.

Measures of resting-state brain network segregation and integration vary in relation to data quantity: implications for within and between subject comparisons of functional brain network organization.

Measures of functional brain network segregation and integration vary with an individual's age, cognitive ability, and health status. Based on these relationships, these measures are frequently examined to study and quantify large-scale patterns of network organization in both basic and applied research settings. However, there is limited information on the stability and reliability of the network measures as applied to functional time-series; these measurement properties are critical to understand if the measures are to be used for individualized characterization of brain networks. We examine measurement reliability using several human datasets (Midnight Scan Club and Human Connectome Project [both Young Adult and Aging]). These datasets include participants with multiple scanning sessions, and collectively include individuals spanning a broad age range of the adult lifespan. The measurement and reliability of measures of resting-state network segregation and integration vary in relation to data quantity for a given participant's scan session; notably, both properties asymptote when estimated using adequate amounts of clean data. We demonstrate how this source of variability can systematically bias interpretation of differences and changes in brain network organization if appropriate safeguards are not included. These observations have important implications for cross-sectional, longitudinal, and interventional comparisons of functional brain network organization.

A radiomics-based brain network in T1 images: construction, attributes, and applications.

T1 image is a widely collected imaging sequence in various neuroimaging datasets, but it is rarely used to construct an individual-level brain network. In this study, a novel individualized radiomics-based structural similarity network was proposed from T1 images. In detail, it used voxel-based morphometry to obtain the preprocessed gray matter images, and radiomic features were then extracted on each region of interest in Brainnetome atlas, and an individualized radiomics-based structural similarity network was finally built using the correlational values of radiomic features between any pair of regions of interest. After that, the network characteristics of individualized radiomics-based structural similarity network were assessed, including graph theory attributes, test-retest reliability, and individual identification ability (fingerprinting). At last, two representative applications for individualized radiomics-based structural similarity network, namely mild cognitive impairment subtype discrimination and fluid intelligence prediction, were exemplified and compared with some other networks on large open-source datasets. The results revealed that the individualized radiomics-based structural similarity network displays remarkable network characteristics and exhibits advantageous performances in mild cognitive impairment subtype discrimination and fluid intelligence prediction. In summary, the individualized radiomics-based structural similarity network provides a distinctive, reliable, and informative individualized structural brain network, which can be combined with other networks such as resting-state functional connectivity for various phenotypic and clinical applications.

Reliability of the TMS-evoked potential in dorsolateral prefrontal cortex.

We currently lack a reliable method to probe cortical excitability noninvasively from the human dorsolateral prefrontal cortex (dlPFC). We recently found that the strength of early and local dlPFC transcranial magnetic stimulation (TMS)-evoked potentials (EL-TEPs) varied widely across dlPFC subregions. Despite these differences in response amplitude, reliability at each target is unknown. Here we quantified within-session reliability of dlPFC EL-TEPs after TMS to six left dlPFC subregions in 15 healthy subjects. We evaluated reliability (concordance correlation coefficient [CCC]) across targets, time windows, quantification methods, regions of interest, sensor- vs. source-space, and number of trials. On average, the medial target was most reliable (CCC = 0.78) and the most anterior target was least reliable (CCC = 0.24). However, all targets except the most anterior were reliable (CCC > 0.7) using at least one combination of the analytical parameters tested. Longer (20 to 60 ms) and later (30 to 60 ms) windows increased reliability compared to earlier and shorter windows. Reliable EL-TEPs (CCC up to 0.86) were observed using only 25 TMS trials at a medial dlPFC target. Overall, medial dlPFC targeting, wider windows, and peak-to-peak quantification improved reliability. With careful selection of target and analytic parameters, highly reliable EL-TEPs can be extracted from the dlPFC after only a small number of trials.

Test-retest reliability and time-of-day variations of perfusion imaging at rest and during a vigilance task.

Arterial spin-labeled perfusion and blood oxygenation level-dependent functional MRI are indispensable tools for noninvasive human brain imaging in clinical and cognitive neuroscience, yet concerns persist regarding the reliability and reproducibility of functional MRI findings. The circadian rhythm is known to play a significant role in physiological and psychological responses, leading to variability in brain function at different times of the day. Despite this, test-retest reliability of brain function across different times of the day remains poorly understood. This study examined the test-retest reliability of six repeated cerebral blood flow measurements using arterial spin-labeled perfusion imaging both at resting-state and during the psychomotor vigilance test, as well as task-induced cerebral blood flow changes in a cohort of 38 healthy participants over a full day. The results demonstrated excellent test-retest reliability for absolute cerebral blood flow measurements at rest and during the psychomotor vigilance test throughout the day. However, task-induced cerebral blood flow changes exhibited poor reliability across various brain regions and networks. Furthermore, reliability declined over longer time intervals within the day, particularly during nighttime scans compared to daytime scans. These findings highlight the superior reliability of absolute cerebral blood flow compared to task-induced cerebral blood flow changes and emphasize the importance of controlling time-of-day effects to enhance the reliability and reproducibility of future brain imaging studies.

On the reliability of individual economic rationality measurements.

A contemporary research agenda in behavioral economics and neuroeconomics aims to identify individual differences and (neuro)psychological correlates of rationality. This research has been widely received in important interdisciplinary and field outlets. However, the psychometric reliability of such measurements of rationality has been presumed without enough methodological scrutiny. Drawing from multiple original and published datasets (in total over 1,600 participants), we unequivocally show that contemporary measurements of rationality have moderate to poor reliability according to common standards. Further analyses of the variance components, as well as a allowing participants to revise previous choices, suggest that this is driven by low between-subject variance rather than high measurement error. As has been argued previously for other behavioral measurements, this poses a challenge to the predominant correlational research designs and the search for sociodemographic or neural predictors. While our results draw a sobering picture of the prospects of contemporary measurements of rationality, they are not necessarily surprising from a theoretical perspective, which we outline in our discussion.

Examining the generalizability of research findings from archival data.

This initiative examined systematically the extent to which a large set of archival research findings generalizes across contexts. We repeated the key analyses for 29 original strategic management effects in the same context (direct reproduction) as well as in 52 novel time periods and geographies; 45% of the reproductions returned results matching the original reports together with 55% of tests in different spans of years and 40% of tests in novel geographies. Some original findings were associated with multiple new tests. Reproducibility was the best predictor of generalizability-for the findings that proved directly reproducible, 84% emerged in other available time periods and 57% emerged in other geographies. Overall, only limited empirical evidence emerged for context sensitivity. In a forecasting survey, independent scientists were able to anticipate which effects would find support in tests in new samples.

Graphene-All-Around Cobalt Interconnect with a Back-End-of-Line Compatible Process.

The graphene-all-around (GAA) structure has been verified to grow directly at 380 °C using hot-wire chemical vapor deposition, within the thermal budget of the back end of the line (BEOL). The cobalt (Co) interconnects with the GAA structure have demonstrated a 10.8% increase in current density, a 27% reduction in resistance, and a 36 times longer electromigration lifetime. X-ray photoelectron spectroscopy and density functional theory calculations have revealed the presence of bonding between carbon and Co, which makes the Co atom more stable to resist external forces. The ability of graphene to act as a diffusion barrier in the GAA structure was confirmed through time-dependent dielectric breakdown measurement. The Co interconnect within the GAA structure exhibits enhanced electrical properties and reliability, which indicates compatibility applications as next-generation interconnect materials in CMOS BEOL.

Establishing the relationship between subjective perception and neural responses: Insights from correlation analysis and representational similarity analysis.

Exploring the relationship between sensory perception and brain responses holds important theoretical and clinical implications. However, commonly used methodologies like correlation analysis performed either intra- or inter- individually often yield inconsistent results across studies, limiting their generalizability. Representational similarity analysis (RSA), a method that assesses the perception-response relationship by calculating the correlation between behavioral and neural patterns, may offer a fresh perspective to reveal novel findings. Here, we delivered a series of graded sensory stimuli of four modalities (i.e., nociceptive somatosensory, non-nociceptive somatosensory, visual, and auditory) to/near the left or right hand of 107 healthy subjects and collected their single-trial perceptual ratings and electroencephalographic (EEG) responses. We examined the relationship between sensory perception and brain responses using within- and between-subject correlation analysis and RSA, and assessed their stability across different numbers of subjects and trials. We found that within-subject and between-subject correlations yielded distinct results: within-subject correlation revealed strong and reliable correlations between perceptual ratings and most brain responses, while between-subject correlation showed weak correlations that were vulnerable to the change of subject number. In addition to verifying the correlation results, RSA revealed some novel findings, i.e., correlations between behavioral and neural patterns were observed in some additional neural responses, such as "γ-ERS" in the visual modality. RSA results were sensitive to the trial number, but not to the subject number, suggesting that consistent results could be obtained for studies with relatively small sample sizes. In conclusion, our study provides a novel perspective on establishing the relationship between behavior and brain activity, emphasizing that RSA holds promise as a method for exploring this pattern relationship in future research.

Efficient functional localization of language regions in the brain.

Important recent advances in the cognitive neuroscience of language have been made using functional localizers to demarcate language-selective regions in individual brains. Although single-subject localizers offer insights that are unavailable in classic group analyses, they require additional scan time that imposes costs on investigators and participants. In particular, the unique practical challenges of scanning children and other special populations has led to less adoption of localizers for neuroimaging research with these theoretically and clinically important groups. Here, we examined how measurements of the spatial extent and functional response profiles of language regions are affected by the duration of an auditory language localizer. We compared how parametrically smaller amounts of data collected from one scanning session affected (i) consistency of group-level whole-brain parcellations, (ii) functional selectivity of subject-level activation in individually defined functional regions of interest (fROIs), (iii) sensitivity and specificity of subject-level whole-brain and fROI activation, and (iv) test-retest reliability of subject-level whole-brain and fROI activation. For many of these metrics, the localizer duration could be reduced by 50-75% while preserving the stability and reliability of both the spatial extent and functional response profiles of language areas. These results indicate that, for most measures relevant to cognitive neuroimaging studies, the brain's language network can be localized just as effectively with 3.5 min of scan time as it can with 12 min. Minimizing the time required to reliably localize the brain's language network allows more effective localizer use in situations where each minute of scan time is particularly precious.

Mapping morphological cortical networks with joint probability distributions from multiple morphological features.

Morphological features sourced from structural magnetic resonance imaging can be used to infer human brain connectivity. Although integrating different morphological features may theoretically be beneficial for obtaining more precise morphological connectivity networks (MCNs), the empirical evidence to support this supposition is scarce. Moreover, the incorporation of different morphological features remains an open question. In this study, we proposed a method to construct cortical MCNs based on multiple morphological features. Specifically, we adopted a multi-dimensional kernel density estimation algorithm to fit regional joint probability distributions (PDs) from different combinations of four morphological features, and estimated inter-regional similarity in the joint PDs via Jensen-Shannon divergence. We evaluated the method by comparing the resultant MCNs with those built based on different single morphological features in terms of topological organization, test-retest reliability, biological plausibility, and behavioral and cognitive relevance. We found that, compared to MCNs built based on different single morphological features, MCNs derived from multiple morphological features displayed less segregated, but more integrated network architecture and different hubs, had higher test-retest reliability, encompassed larger proportions of inter-hemispheric edges and edges between brain regions within the same cytoarchitectonic class, and explained more inter-individual variance in behavior and cognition. These findings were largely reproducible when different brain atlases were used for cortical parcellation. Further analysis of macaque MCNs revealed weak, but significant correlations with axonal connectivity from tract-tracing, independent of the number of morphological features. Altogether, this paper proposes a new method for integrating different morphological features, which will be beneficial for constructing MCNs.

Voxel-based texture similarity networks reveal individual variability and correlate with biological ontologies.

The human brain is organized as a complex, hierarchical network. However, the structural covariance patterns among brain regions and the underlying biological substrates of such covariance networks remain to be clarified. The present study proposed a novel individualized structural covariance network termed voxel-based texture similarity networks (vTSNs) based on 76 refined voxel-based textural features derived from structural magnetic resonance images. Validated in three independent longitudinal healthy cohorts (40, 23, and 60 healthy participants, respectively) with two common brain atlases, we found that the vTSN could robustly resolve inter-subject variability with high test-retest reliability. In contrast to the regional-based texture similarity networks (rTSNs) that calculate radiomic features based on region-of-interest information, vTSNs had higher inter- and intra-subject variability ratios and test-retest reliability in connectivity strength and network topological properties. Moreover, the Spearman correlation indicated a stronger association of the gene expression similarity network (GESN) with vTSNs than with rTSNs (vTSN: r = 0.600, rTSN: r = 0.433, z = 39.784, P < 0.001). Hierarchical clustering identified 3 vTSN subnets with differential association patterns with 13 coexpression modules, 16 neurotransmitters, 7 electrophysiology, 4 metabolism, and 2 large-scale structural and 4 functional organization maps. Moreover, these subnets had unique biological hierarchical organization from the subcortex-limbic system to the ventral neocortex and then to the dorsal neocortex. Based on 424 unrelated, qualified healthy subjects from the Human Connectome Project, we found that vTSNs could sensitively represent sex differences, especially for connections in the subcortex-limbic system and between the subcortex-limbic system and the ventral neocortex. Moreover, a multivariate variance component model revealed that vTSNs could explain a significant proportion of inter-subject behavioral variance in cognition (80.0 %) and motor functions (63.4 %). Finally, using 494 healthy adults (aged 19-80 years old) from the Southwest University Adult Lifespan Dataset, the Spearman correlation identified a significant association between aging and vTSN strength, especially within the subcortex-limbic system and between the subcortex-limbic system and the dorsal neocortex. In summary, our proposed vTSN is robust in uncovering individual variability and neurobiological brain processes, which can serve as biologically plausible measures for linking biological processes and human behavior.

A precision neuroscience approach to estimating reliability of neural responses during emotion processing: Implications for task-fMRI.

Recent work demonstrating low test-retest reliability of neural activation during fMRI tasks raises questions about the utility of task-based fMRI for the study of individual variation in brain function. Two possible sources of the instability in task-based BOLD signal over time are noise or measurement error in the instrument, and meaningful variation across time within-individuals in the construct itself-brain activation elicited during fMRI tasks. Examining the contribution of these two sources of test-retest unreliability in task-evoked brain activity has far-reaching implications for cognitive neuroscience. If test-retest reliability largely reflects measurement error, it suggests that task-based fMRI has little utility in the study of either inter- or intra-individual differences. On the other hand, if task-evoked BOLD signal varies meaningfully over time, it would suggest that this tool may yet be well suited to studying intraindividual variation. We parse these sources of variance in BOLD signal in response to emotional cues over time and within-individuals in a longitudinal sample with 10 monthly fMRI scans. Test-retest reliability was low, reflecting a lack of stability in between-person differences across scans. In contrast, within-person, within-session internal consistency of the BOLD signal was higher, and within-person fluctuations across sessions explained almost half the variance in voxel-level neural responses. Additionally, monthly fluctuations in neural response to emotional cues were associated with intraindividual variation in mood, sleep, and exposure to stressors. Rather than reflecting trait-like differences across people, neural responses to emotional cues may be more reflective of intraindividual variation over time. These patterns suggest that task-based fMRI may be able to contribute to the study of individual variation in brain function if more attention is given to within-individual variation approaches, psychometrics-beginning with improving reliability beyond the modest estimates observed here, and the validity of task fMRI beyond the suggestive associations reported here.

Test-retest reliability of the attention network test from the perspective of intrinsic network organization.

The attention network test (ANT), developed based on the triple-network taxonomy by Posner and colleagues, has been widely used to examine the efficacy of alerting, orienting and executive control in clinical and developmental neuroscience studies. Recent research suggests the imperfect reliability of the behavioural ANT and its variants. However, the classical ANT fMRI task's test-retest reliability has received little attention. Moreover, it remains ambiguous whether the attention-related intrinsic network components, especially the dorsal attention, ventral attention and frontoparietal network, manifest acceptable reliability. The present study approaches these issues by utilizing an openly available ANT fMRI dataset for participants with Parkinson's disease and healthy elderly. The reproducibility of group-level activations across sessions and participant groups and the test-retest reliability at the individual level were examined at the voxel, region and network levels. The intrinsic network was defined using the Yeo-Schaefer atlas. Our results reveal three critical facets: (1) the overlapping of the group-level contrast map between sessions and between participant groups was unsatisfactory; (2) the reliability of alerting, orienting and executive, defined as a contrast between conditions, was worse than estimates of specific conditions. (3) Dorsal attention, ventral attention, visual and somatomotor networks showed acceptable reliability for the congruent and incongruent conditions. Our results suggest that specific condition estimates might be used instead of the contrast map for individual or group-difference studies.

The effects of spatial leakage correction on the reliability of EEG-based functional connectivity networks.

Electroencephalography (EEG) functional connectivity (FC) estimates are confounded by the volume conduction problem. This effect can be greatly reduced by applying FC measures insensitive to instantaneous, zero-lag dependencies (corrected measures). However, numerous studies showed that FC measures sensitive to volume conduction (uncorrected measures) exhibit higher reliability and higher subject-level identifiability. We tested how source reconstruction contributed to the reliability difference of EEG FC measures on a large (n = 201) resting-state data set testing eight FC measures (including corrected and uncorrected measures). We showed that the high reliability of uncorrected FC measures in resting state partly stems from source reconstruction: idiosyncratic noise patterns define a baseline resting-state functional network that explains a significant portion of the reliability of uncorrected FC measures. This effect remained valid for template head model-based, as well as individual head model-based source reconstruction. Based on our findings we made suggestions how to best use spatial leakage corrected and uncorrected FC measures depending on the main goals of the study.

Reliability of dynamic causal modelling of resting-state magnetoencephalography.

This study assesses the reliability of resting-state dynamic causal modelling (DCM) of magnetoencephalography (MEG) under conductance-based canonical microcircuit models, in terms of both posterior parameter estimates and model evidence. We use resting-state MEG data from two sessions, acquired 2 weeks apart, from a cohort with high between-subject variance arising from Alzheimer's disease. Our focus is not on the effect of disease, but on the reliability of the methods (as within-subject between-session agreement), which is crucial for future studies of disease progression and drug intervention. To assess the reliability of first-level DCMs, we compare model evidence associated with the covariance among subject-specific free energies (i.e., the 'quality' of the models) with versus without interclass correlations. We then used parametric empirical Bayes (PEB) to investigate the differences between the inferred DCM parameter probability distributions at the between subject level. Specifically, we examined the evidence for or against parameter differences (i) within-subject, within-session, and between-epochs; (ii) within-subject between-session; and (iii) within-site between-subjects, accommodating the conditional dependency among parameter estimates. We show that for data acquired close in time, and under similar circumstances, more than 95% of inferred DCM parameters are unlikely to differ, speaking to mutual predictability over sessions. Using PEB, we show a reciprocal relationship between a conventional definition of 'reliability' and the conditional dependency among inferred model parameters. Our analyses confirm the reliability and reproducibility of the conductance-based DCMs for resting-state neurophysiological data. In this respect, the implicit generative modelling is suitable for interventional and longitudinal studies of neurological and psychiatric disorders.

Reliability of variability and complexity measures for task and task-free BOLD fMRI.

Brain activity continuously fluctuates over time, even if the brain is in controlled (e.g., experimentally induced) states. Recent years have seen an increasing interest in understanding the complexity of these temporal variations, for example with respect to developmental changes in brain function or between-person differences in healthy and clinical populations. However, the psychometric reliability of brain signal variability and complexity measures-which is an important precondition for robust individual differences as well as longitudinal research-is not yet sufficiently studied. We examined reliability (split-half correlations) and test-retest correlations for task-free (resting-state) BOLD fMRI as well as split-half correlations for seven functional task data sets from the Human Connectome Project to evaluate their reliability. We observed good to excellent split-half reliability for temporal variability measures derived from rest and task fMRI activation time series (standard deviation, mean absolute successive difference, mean squared successive difference), and moderate test-retest correlations for the same variability measures under rest conditions. Brain signal complexity estimates (several entropy and dimensionality measures) showed moderate to good reliabilities under both, rest and task activation conditions. We calculated the same measures also for time-resolved (dynamic) functional connectivity time series and observed moderate to good reliabilities for variability measures, but poor reliabilities for complexity measures derived from functional connectivity time series. Global (i.e., mean across cortical regions) measures tended to show higher reliability than region-specific variability or complexity estimates. Larger subcortical regions showed similar reliability as cortical regions, but small regions showed lower reliability, especially for complexity measures. Lastly, we also show that reliability scores are only minorly dependent on differences in scan length and replicate our results across different parcellation and denoising strategies. These results suggest that the variability and complexity of BOLD activation time series are robust measures well-suited for individual differences research. Temporal variability of global functional connectivity over time provides an important novel approach to robustly quantifying the dynamics of brain function. PRACTITIONER POINTS: Variability and complexity measures of BOLD activation show good split-half reliability and moderate test-retest reliability. Measures of variability of global functional connectivity over time can robustly quantify neural dynamics. Length of fMRI data has only a minor effect on reliability.

Spinal fMRI demonstrates segmental organisation of functionally connected networks in the cervical spinal cord: A test-retest reliability study.

Resting functional magnetic resonance imaging (fMRI) studies have identified intrinsic spinal cord activity, which forms organised motor (ventral) and sensory (dorsal) resting-state networks. However, to facilitate the use of spinal fMRI in, for example, clinical studies, it is crucial to first assess the reliability of the method, particularly given the unique anatomical, physiological, and methodological challenges associated with acquiring the data. Here, we characterise functional connectivity relationships in the cervical cord and assess their between-session test-retest reliability in 23 young healthy volunteers. Resting-state networks were estimated in two ways (1) by estimating seed-to-voxel connectivity maps and (2) by calculating seed-to-seed correlations. Seed regions corresponded to the four grey matter horns (ventral/dorsal and left/right) of C5-C8 segmental levels. Test-retest reliability was assessed using the intraclass correlation coefficient. Spatial overlap of clusters derived from seed-to-voxel analysis between sessions was examined using Dice coefficients. Following seed-to-voxel analysis, we observed distinct unilateral dorsal and ventral organisation of cervical spinal resting-state networks that was largely confined in the rostro-caudal extent to each spinal segmental level, with more sparse connections observed between segments. Additionally, strongest correlations were observed between within-segment ipsilateral dorsal-ventral connections, followed by within-segment dorso-dorsal and ventro-ventral connections. Test-retest reliability of these networks was mixed. Reliability was poor when assessed on a voxelwise level, with more promising indications of reliability when examining the average signal within clusters. Reliability of correlation strength between seeds was highly variable, with the highest reliability achieved in ipsilateral dorsal-ventral and dorso-dorsal/ventro-ventral connectivity. However, the spatial overlap of networks between sessions was excellent. We demonstrate that while test-retest reliability of cervical spinal resting-state networks is mixed, their spatial extent is similar across sessions, suggesting that these networks are characterised by a consistent spatial representation over time.