Imaging neuroscience (Cambridge, Mass.)最新文献

{"title":"Functional connectivity across the human subcortical auditory system using an autoregressive matrix-Gaussian copula graphical model approach with partial correlations.","authors":"Noirrit Kiran Chandra, Kevin R Sitek, Bharath Chandrasekaran, Abhra Sarkar","doi":"10.1162/imag_a_00258","DOIUrl":"10.1162/imag_a_00258","url":null,"abstract":"<p><p>The auditory system comprises multiple subcortical brain structures that process and refine incoming acoustic signals along the primary auditory pathway. Due to technical limitations of imaging small structures deep inside the brain, most of our knowledge of the subcortical auditory system is based on research in animal models using invasive methodologies. Advances in ultrahigh-field functional magnetic resonance imaging (fMRI) acquisition have enabled novel noninvasive investigations of the human auditory subcortex, including fundamental features of auditory representation such as tonotopy and periodotopy. However, functional connectivity across subcortical networks is still underexplored in humans, with ongoing development of related methods. Traditionally, functional connectivity is estimated from fMRI data with full correlation matrices. However, partial correlations reveal the relationship between two regions after removing the effects of all other regions, reflecting more direct connectivity. Partial correlation analysis is particularly promising in the ascending auditory system, where sensory information is passed in an obligatory manner, from nucleus to nucleus up the primary auditory pathway, providing redundant but also increasingly abstract representations of auditory stimuli. While most existing methods for learning conditional dependency structures based on partial correlations assume independently and identically Gaussian distributed data, fMRI data exhibit significant deviations from Gaussianity as well as high-temporal autocorrelation. In this paper, we developed an autoregressive matrix-Gaussian copula graphical model (ARMGCGM) approach to estimate the partial correlations and thereby infer the functional connectivity patterns within the auditory system while appropriately accounting for autocorrelations between successive fMRI scans. Our results show strong positive partial correlations between successive structures in the primary auditory pathway on each side (left and right), including between auditory midbrain and thalamus, and between primary and associative auditory cortex. These results are highly stable when splitting the data in halves according to the acquisition schemes and computing partial correlations separately for each half of the data, as well as across cross-validation folds. In contrast, full correlation-based analysis identified a rich network of interconnectivity that was not specific to adjacent nodes along the pathway. Overall, our results demonstrate that unique functional connectivity patterns along the auditory pathway are recoverable using novel connectivity approaches and that our connectivity methods are reliable across multiple acquisitions.</p>","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"2 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11485223/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142482357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of a 1H Brain MRS Benchmark Dataset to Deep Learning for Out-of-Voxel Artifacts","authors":"Aaron T. Gudmundson, Christopher W. Davies-Jenkins, İpek Özdemir, Saipavitra Murali-Manohar, Helge J. Zöllner, Yulu Song, Kathleen E. Hupfeld, Alfons Schnitzler, Georg Oeltzschner, Craig E. L. Stark, Richard A. E. Edden","doi":"10.1162/imag_a_00025","DOIUrl":"https://doi.org/10.1162/imag_a_00025","url":null,"abstract":"Abstract Neural networks are potentially valuable for many of the challenges associated with MRS data. The purpose of this manuscript is to describe the AGNOSTIC dataset, which contains 259,200 synthetic 1H MRS examples for training and testing neural networks. AGNOSTIC was created using 270 basis sets that were simulated across 18 field strengths and 15 echo times. The synthetic examples were produced to resemble in vivo brain data with combinations of metabolite, macromolecule, residual water signals, and noise. To demonstrate the utility, we apply AGNOSTIC to train two Convolutional Neural Networks (CNNs) to address out-of-voxel (OOV) echoes. A Detection Network was trained to identify the point-wise presence of OOV echoes, providing proof of concept for real-time detection. A Prediction Network was trained to reconstruct OOV echoes, allowing subtraction during post-processing. Complex OOV signals were mixed into 85% of synthetic examples to train two separate CNNs for the detection and prediction of OOV signals. AGNOSTIC is available through Dryad, and all Python 3 code is available through GitHub. The Detection network was shown to perform well, identifying 95% of OOV echoes. Traditional modeling of these detected OOV signals was evaluated and may prove to be an effective method during linear-combination modeling. The Prediction Network greatly reduces OOV echoes within FIDs and achieved a median log10 normed-MSE of—1.79, an improvement of almost two orders of magnitude.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"250 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135011810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Synchronous high-amplitude co-fluctuations of functional brain networks during movie-watching","authors":"Jacob C. Tanner, Joshua Faskowitz, Lisa Byrge, Daniel P. Kennedy, Olaf Sporns, Richard F. Betzel","doi":"10.1162/imag_a_00026","DOIUrl":"https://doi.org/10.1162/imag_a_00026","url":null,"abstract":"Abstract Recent studies have shown that functional connectivity can be decomposed into its exact frame-wise contributions, revealing short-lived, infrequent, and high-amplitude time points referred to as “events.” Events contribute disproportionately to the time-averaged connectivity pattern, improve identifiability and brain-behavior associations, and differences in their expression have been linked to endogenous hormonal fluctuations and autism. Here, we explore the characteristics of events while subjects watch movies. Using two independently-acquired imaging datasets in which participants passively watched movies, we find that events synchronize across individuals and based on the level of synchronization, can be categorized into three distinct classes: those that synchronize at the boundaries between movies, those that synchronize during movies, and those that do not synchronize at all. We find that boundary events, compared to the other categories, exhibit greater amplitude, distinct co-fluctuation patterns, and temporal propagation. We show that underlying boundary events1 is a specific mode of co-fluctuation involving the activation of control and salience systems alongside the deactivation of visual systems. Events that synchronize during the movie, on the other hand, display a pattern of co-fluctuation that is time-locked to the movie stimulus. Finally, we found that subjects’ time-varying brain networks are most similar to one another during these synchronous events.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"43 2-3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135111391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Combining Magnetoencephalography with Telemetric Streaming of Intracranial Recordings and Deep Brain Stimulation – a Feasibility Study","authors":"Mansoureh Fahimi Hnazaee, Matthias Sure, George O’Neill, Gaetano Leogrande, Alfons Schnitzler, Esther Florin, Vladimir Litvak","doi":"10.1162/imag_a_00029","DOIUrl":"https://doi.org/10.1162/imag_a_00029","url":null,"abstract":"Abstract The combination of subcortical Local Field Potential (LFP) recordings and stimulation with Magnetoencephalography (MEG) in Deep Brain Stimulation (DBS) patients enables the investigation of cortico-subcortical communication patterns and provides insights into DBS mechanisms. Until now, these recordings have been carried out in post-surgical patients with externalised leads. However, a new generation of telemetric stimulators makes it possible to record and stream LFP data in chronically implanted patients. Nevertheless, whether such streaming can be combined with MEG has not been tested. In the present study, we tested the most commonly implanted telemetric stimulator—Medtronic Percept PC with a phantom in three different MEG systems: two cryogenic scanners (CTF and MEGIN) and an experimental Optically Pumped Magnetometry (OPM)-based system. We found that when used in combination with the new SenSight segmented leads, Percept PC telemetric streaming only generates band-limited interference in the MEG at 123 Hz and harmonics. However, the “legacy streaming mode” used with older lead models generates multiple, dense artefact peaks in the physiological range of interest (below 50 Hz). The effect of stimulation on MEG critically depends on whether it is done in bipolar (between two contacts on the lead) or monopolar (between a lead contact and the stimulator case) mode. Monopolar DBS creates severe interference in the MEG as previously reported. However, we found that the OPM system is more resilient to this interference and could provide artefact-free measurements, at least for limited frequency ranges. A resting measurement in the MEGIN system from a Parkinson’s patient implanted with Percept PC and subthalamic SenSight leads revealed artefact patterns consistent with our phantom recordings. Moreover, analysis of LFP-MEG coherence in this patient showed oscillatory coherent networks consistent in their frequency and topography with those described in published group studies done with externalised leads. In conclusion, Percept PC telemetric streaming with SenSight leads is compatible with MEG. Furthermore, OPM sensors could provide additional new opportunities for studying DBS effects.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"406 16","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135112037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Short- and long-term effects of emotion up- and down-regulation","authors":"Kersten Diers, Anne Gärtner, Sabine Schönfeld, Denise Dörfel, Henrik Walter, Burkhard Brocke, Alexander Strobel","doi":"10.1162/imag_a_00028","DOIUrl":"https://doi.org/10.1162/imag_a_00028","url":null,"abstract":"Abstract It is an open question in cognitive emotion regulation research how emotion regulation unfolds over time, and whether the brain regions involved in down-regulation are also recruited during up-regulation of emotions. As a replication and extension of our preceding study, we conducted an fMRI study in young healthy adults on the neural basis of up- and down-regulation of negative and neutral pictures during the immediate stimulation phase as well as after short- and long-term delays (N=47 for immediate and short-term delays, a subset of N=30 for long-term delays). For this, we employed three experimental conditions—down-regulation (distance), maintenance (permit), and up-regulation (intensify)—for negative and neutral pictures, and investigated the neural responses during the stimulation and post-stimulation phase as well as during re-exposure after 10 min and after 1 week. We observed the following main results: first, we found greater activation in emotion-generating regions such as the amygdala in the permit vs. distance and the intensify vs. distance comparisons, but not in the intensify vs. permit comparison. Second, we observed greater activation in emotion-regulating regions such as the right inferior parietal and right superior / middle frontal cortex in the distance vs. permit and the distance vs. intensify contrasts, but not the permit vs. intensify contrast. Third, we found that the activation difference between distance and intensify within the amygdala reversed after the regulation period. Fourth, previous emotion regulation did not influence the activation during re-exposure, neither after 10 min nor after 1 week. Taken together, the results provide a partial replication of persistent effects observed in our preceding study, indicate different neural systems for up- and down-regulation, and demonstrate that a broader perspective on emotion regulation can be achieved by simultaneously considering different goals, directions, and strategies of emotion regulation in a single experiment.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"394 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135111847","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Family income is not significantly associated with T1w/T2w ratio in the Human Connectome Project in Development","authors":"David G. Weissman, Graham L. Baum, Ashley Sanders, Maya L. Rosen, Deanna M. Barch, Katie A. McLaughlin, Leah H. Somerville","doi":"10.1162/imag_a_00021","DOIUrl":"https://doi.org/10.1162/imag_a_00021","url":null,"abstract":"Abstract Growing evidence indicates that brain development varies as a function of family socioeconomic status (SES). Numerous studies have demonstrated that children from low-SES backgrounds have thinner cortex than children from higher-SES backgrounds. A recent study in a large developmental sample found widespread associations between lower SES and greater cortical T1w/T2w ratio—thought to be an indirect proxy for cortical myelin. We evaluated the association of family income with cortical T1w/T2w ratio as a function of age in the Human Connectome Project in Development sample of 989 youth aged 8-21 years. We observed no associations between family income and T1w/T2w ratio that were significant after corrections for multiple comparisons at the region, network, or whole-brain level. Region of practical equivalence (ROPE) analyses were also consistent with the absence of an association between family income and T1w/T2w ratio. We discuss potential methodological sources of inconsistency between this and the previous study examining the same question. While the question of whether family income may influence cortical myelin development remains, these null results may indicate that the association between SES and cortical myelin development may not be as strong as with other aspects of brain structure.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135324534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Striatal response to negative feedback in a stop signal task operates as a multi-value learning signal","authors":"Benjamin J. Smith, Megan Lipsett, Danielle Cosme, Victoria A. Braun, Anastasia M. Browning O’Hagan, Elliot T. Berkman","doi":"10.1162/imag_a_00024","DOIUrl":"https://doi.org/10.1162/imag_a_00024","url":null,"abstract":"Abstract Background and aim: We examined error-driven learning in fMRI activity of 217 subjects in a stop signal task to obtain a more robust characterization of the relation between behavioral measures of learning and corresponding neural learning signals than previously possible. Methods: The stop signal task is a two-alternative forced choice in which participants respond to an arrow by pressing a left or right button but must inhibit that response on 1 in 7 trials when cued by an auditory “stop signal.” We examined post-error learning by comparing brain activity (BOLD signal) and behavioral responses on trials preceded by successful (correct stop) vs. failed (failed stop) inhibition. Results: There was strong evidence of greater bilateral striatal activity in the period immediately following correct (vs. failed) stop trials (most evident in the putamen; peak MNI coordinates [-26 8 -2], 430 voxels, p &amp;lt; 0.001; [24 14 0], 527 voxels, p &amp;lt; 0.001). We measured median activity in the bilateral striatal cluster following every failed stop and correct stop trial and correlated it with learning signals for (a) probability and (b) latency of the stop signal. In a mixed-effects model predicting activity 5-10 s after the stop signal, both reaction time (RT) change (B = -0.05, t = 3.0, χ2 = 11.3, p &amp;lt; 0.001) and probability of stop trial change (B = 1.53, t = 6.0, χ2 = 43.0, p &amp;lt; 0.001) had significant within-subjects effects on median activity. In a similar mixed model predicting activity 1-5 s after the stop signal, only probability of stop trial change was predictive. Conclusions: A mixed-effects model indicates the striatal activity might be a learning signal that encodes reaction time change and the current expected probability of a stop trial occuring. This extends existing evidence that the striatum encodes a reward prediction error signal for learning within the stop signal task, and demonstrates for the first time that this signal seems to encode both change in stop signal probability and in stop signal delay.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134978923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Test-Retest Reliability of the Human Connectome: An OPM-MEG study","authors":"Lukas Rier, Sebastian Michelmann, Harrison Ritz, Vishal Shah, Ryan M. Hill, James Osborne, Cody Doyle, Niall Holmes, Richard Bowtell, Matthew J. Brookes, Kenneth A. Norman, Uri Hasson, Jonathan D. Cohen, Elena Boto","doi":"10.1162/imag_a_00020","DOIUrl":"https://doi.org/10.1162/imag_a_00020","url":null,"abstract":"Abstract Magnetoencephalography with optically pumped magnetometers (OPM-MEG) offers a new way to record electrophysiological brain function, with significant advantages over conventional MEG, including adaptability to head shape/size, free movement during scanning, increased signal amplitude, and no reliance on cryogenics. However, OPM-MEG remains in its infancy, with significant questions to be answered regarding the optimal system design. Here, we present an open-source dataset acquired using a newly constructed OPM-MEG system with a triaxial sensor design, 168 channels, OPM-optimised magnetic shielding, and active background field control. We measure the test-retest reliability of the human connectome, which was computed using amplitude envelope correlation to measure whole-brain (parcellated) functional connectivity, in 10 individuals while they watch a 600 s move clip. Our results show high repeatability between experimental runs at the group level, with a correlation coefficient of 0.81 in the θ, 0.93 in α, and 0.94 in β frequency ranges. At the individual subject level, we found marked differences between individuals, but high within-subject robustness (correlations of 0.56 ± 0.25, 0.72 ± 0.15, and 0.78 ± 0.13 in α, θ, and β respectively). These results compare well to previous findings using conventional MEG and show that OPM-MEG is a viable way to robustly characterise connectivity.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135273621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The impact of quality control on cortical morphometry comparisons in autism","authors":"Saashi A. Bedford, Alfredo Ortiz-Rosa, Jenna M. Schabdach, Manuela Costantino, Stephanie Tullo, Tom Piercy, Meng-Chuan Lai, Michael V. Lombardo, Adriana Di Martino, Gabriel A. Devenyi, M. Mallar Chakravarty, Aaron F. Alexander-Bloch, Jakob Seidlitz, Simon Baron-Cohen, Richard A.I. Bethlehem","doi":"10.1162/imag_a_00022","DOIUrl":"https://doi.org/10.1162/imag_a_00022","url":null,"abstract":"Abstract Structural magnetic resonance imaging (MRI) quality is known to impact and bias neuroanatomical estimates and downstream analysis, including case-control comparisons, and a growing body of work has demonstrated the importance of careful quality control (QC) and evaluated the impact of image and image-processing quality. However, the growing size of typical neuroimaging datasets presents an additional challenge to QC, which is typically extremely time and labour intensive. One of the most important aspects of MRI quality is the accuracy of processed outputs, which have been shown to impact estimated neurodevelopmental trajectories. Here, we evaluate whether the quality of surface reconstructions by FreeSurfer (one of the most widely used MRI processing pipelines) interacts with clinical and demographic factors. We present a tool, FSQC, that enables quick and efficient yet thorough assessment of outputs of the FreeSurfer processing pipeline. We validate our method against other existing QC metrics, including the automated FreeSurfer Euler number, two other manual ratings of raw image quality, and two popular automated QC methods. We show strikingly similar spatial patterns in the relationship between each QC measure and cortical thickness; relationships for cortical volume and surface area are largely consistent across metrics, though with some notable differences. We next demonstrate that thresholding by QC score attenuates but does not eliminate the impact of quality on cortical estimates. Finally, we explore different ways of controlling for quality when examining differences between autistic individuals and neurotypical controls in the Autism Brain Imaging Data Exchange (ABIDE) dataset, demonstrating that inadequate control for quality can alter results of case-control comparisons.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135324546","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated Motion Artifact Detection in Early Pediatric Diffusion MRI Using a Convolutional Neural Network","authors":"Jayse Merle Weaver, Marissa DiPiero, Patrik Goncalves Rodrigues, Hassan Cordash, Richard J. Davidson, Elizabeth M. Planalp, Douglas C. Dean III","doi":"10.1162/imag_a_00023","DOIUrl":"https://doi.org/10.1162/imag_a_00023","url":null,"abstract":"Abstract Diffusion MRI (dMRI) is a widely used method to investigate the microstructure of the brain. Quality control (QC) of dMRI data is an important processing step that is performed prior to analysis using models such as diffusion tensor imaging (DTI) or neurite orientation dispersion and density imaging (NODDI). When processing dMRI data from infants and young children, where intra-scan motion is common, the identification and removal of motion artifacts is of the utmost importance. Manual QC of dMRI data is (1) time-consuming due to the large number of diffusion directions, (2) expensive, and (3) prone to subjective errors and observer variability. Prior techniques for automated dMRI QC have mostly been limited to adults or school-age children. Here, we propose a deep learning-based motion artifact detection tool for dMRI data acquired from infants and toddlers. The proposed framework uses a simple three-dimensional convolutional neural network (3DCNN) trained and tested on an early pediatric dataset of 2,276 dMRI volumes from 121 exams acquired at 1 month and 24 months of age. An average classification accuracy of 95% was achieved following four-fold cross-validation. A second dataset with different acquisition parameters and ages ranging from 2-36 months (consisting of 2,349 dMRI volumes from 26 exams) was used to test network generalizability, achieving 98% classification accuracy. Finally, to demonstrate the importance of motion artifact volume removal in a dMRI processing pipeline, the dMRI data were fit to the DTI and NODDI models and the parameter maps were compared with and without motion artifact removal.","PeriodicalId":73341,"journal":{"name":"Imaging neuroscience (Cambridge, Mass.)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135407944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}