Changes in indicators of cerebral metabolic stress following treatment with voxelotor in children and adolescents with sickle cell anemia.

Voxelotor is a small molecule that reduces the polymerization of sickle hemoglobin by increasing its affinity for oxygen. In patients with sickle cell anemia, it has been postulated that increasing hemoglobin-oxygen affinity could limit oxygen offloading from hemoglobin, causing an increase in cerebral metabolic stress. To investigate this hypothetical concern, we used multimodal brain imaging to define the effects of voxelotor on cerebral blood flow and oxygen extraction. We followed four patients for 2-5 months during and/or after voxelotor therapy. This study showed no observable increase in cerebral blood flow or oxygen extraction fraction during treatment.

Functional MRI Assessment of Brain Activity Patterns Associated with Reading in Medulloblastoma Survivors.

Medulloblastoma, a malignant brain tumor primarily affecting children, poses significant challenges to patients and clinicians due to its complex treatment and potential long-term cognitive consequences. While recent advancements in treatment have significantly improved survival rates, survivors often face cognitive impairments, particularly in reading, impacting their quality of life. According to the double deficit theory, reading impairments are caused by deficits in one or both of two independent reading-related functions: phonological awareness and rapid visual naming. This longitudinal study investigates neurofunctional changes related to reading in medulloblastoma survivors in comparison to controls using functional MRI acquired during rapid automatized naming tasks over three annual visits. Support vector machine classification of functional MRI data reveals a progressive divergence in brain activity patterns between medulloblastoma survivors and healthy controls over time, suggesting delayed effects of cancer treatment on brain function. Alterations in brain regions involved in visual processing and orthographic recognition during rapid naming tasks imply disruptions in the ventral visual pathway associated with normal orthographic processing. These alterations are correlated with performance in tasks involving sound awareness, reading fluency, and word attack. These findings underscore the dynamic nature of post-treatment neurofunctional alterations and the importance of early identification and intervention to address cognitive deficits in survivors.

Real-Time fMRI Neurofeedback Training of Selective Attention in Older Adults.

BACKGROUND: Selective attention declines with age, due to age-related functional changes in dorsal anterior cingulate cortex (dACC). Real-time functional magnetic resonance imaging (rtfMRI) neurofeedback has been used in young adults to train volitional control of brain activity, including in dACC. METHODS: For the first time, this study used rtfMRI neurofeedback to train 19 young and 27 older adults in volitional up- or down-regulation of bilateral dACC during a selective attention task. RESULTS: Older participants in the up-regulation condition (experimental group) showed greater reward points and dACC BOLD signal across training sessions, reflective of neurofeedback training success; and faster reaction time and better response accuracy, suggesting behavioral benefits on selective attention. These effects were not observed for older participants in the down-regulation condition (inverse condition control group), supporting specificity of volitional dACC up-regulation training in older adults. These effects were, unexpectedly, also not observed for young participants in the up-regulation condition (age control group), perhaps due to a lack of motivation to continue the training. CONCLUSIONS: These findings provide promising first evidence of functional plasticity in dACC in late life via rtfMRI neurofeedback up-regulation training, enhancing selective attention, and demonstrate proof of concept of rtfMRI neurofeedback training in cognitive aging.

Motor Intentions Decoded from fMRI Signals.

Motor intention is a high-level brain function related to planning for movement. Although studies have shown that motor intentions can be decoded from brain signals before movement execution, it is unclear whether intentions relating to mental imagery of movement can be decoded. Here, we investigated whether differences in spatial and temporal patterns of brain activation were elicited by intentions to perform different types of motor imagery and whether the patterns could be used by a multivariate pattern classifier to detect such differential intentions. The results showed that it is possible to decode intentions before the onset of different types of motor imagery from functional MR signals obtained from fronto-parietal brain regions, such as the premotor cortex and posterior parietal cortex, while controlling for eye movements and for muscular activity of the hands. These results highlight the critical role played by the aforementioned brain regions in covert motor intentions. Moreover, they have substantial implications for rehabilitating patients with motor disabilities.

Self-Regulation of the Posterior-Frontal Brain Activity with Real-Time fMRI Neurofeedback to Influence Perceptual Discrimination.

The Global Neuronal Workspace (GNW) hypothesis states that the visual percept is available to conscious awareness only if recurrent long-distance interactions among distributed brain regions activate neural circuitry extending from the posterior areas to prefrontal regions above a certain excitation threshold. To directly test this hypothesis, we trained 14 human participants to increase blood oxygenation level-dependent (BOLD) signals with real-time functional magnetic resonance imaging (rtfMRI)-based neurofeedback simultaneously in four specific regions of the occipital, temporal, insular and prefrontal parts of the brain. Specifically, we hypothesized that the up-regulation of the mean BOLD activity in the posterior-frontal brain regions lowers the perceptual threshold for visual stimuli, while down-regulation raises the threshold. Our results showed that participants could perform up-regulation (Wilcoxon test, session 1: p = 0.022; session 4: p = 0.041) of the posterior-frontal brain activity, but not down-regulation. Furthermore, the up-regulation training led to a significant reduction in the visual perceptual threshold, but no substantial change in perceptual threshold was observed after the down-regulation training. These findings show that the up-regulation of the posterior-frontal regions improves the perceptual discrimination of the stimuli. However, further questions as to whether the posterior-frontal regions can be down-regulated at all, and whether down-regulation raises the perceptual threshold, remain unanswered.

Hydroxyurea maintains working memory function in pediatric sickle cell disease.

Sickle cell disease (SCD) decreases the oxygen-carrying capacity of red blood cells. Children with SCD have reduced/restricted cerebral blood flow, resulting in neurocognitive deficits. Hydroxyurea is the standard treatment for SCD; however, whether hydroxyurea influences such effects is unclear. A key area of SCD-associated neurocognitive impairment is working memory, which is implicated in other cognitive and academic skills. The neural correlates of working memory can be tested using n-back tasks. We analyzed functional magnetic resonance imaging (fMRI) data of patients with SCD (20 hydroxyurea-treated patients and 11 controls, aged 7-18 years) while they performed n-back tasks. Blood-oxygenation level-dependent (BOLD) signals were assessed during working memory processing at 2 time points: before hydroxyurea treatment and ~1 year after treatment was initiated. Neurocognitive measures were also assessed at both time points. Our results suggested that working memory was stable in the treated group. We observed a treatment-by-time interaction in the right cuneus and angular gyrus for the 2- >0-back contrast. Searchlight-pattern classification of the 2 time points of the 2-back tasks identified greater changes in the pattern and magnitude of BOLD signals, especially in the posterior regions of the brain, in the control group than in the treated group. In the control group at 1-year follow-up, 2-back BOLD signals increased across time points in several clusters (e.g., right inferior temporal lobe, right angular gyrus). We hypothesize that these changes resulted from increased cognitive effort during working memory processing in the absence of hydroxyurea. In the treated group, 0- to 2-back BOLD signals in the right angular gyrus and left cuneus increased continuously with increasing working memory load, potentially related to a broader dynamic range in response to task difficulty and cognitive effort. These findings suggest that hydroxyurea treatment helps maintain working memory function in SCD.

Comparison of LED- and LASER-based fNIRS technologies to record the human peri­spinal cord neurovascular response.

Recently, functional Near-Infrared Spectroscopy (fNIRS) was applied to obtain, non-invasively, the human peri­spinal Neuro-Vascular Response (NVR) under a non-noxious electrical stimulation of a peripheral nerve. This method allowed the measurements of changes in the concentration of oxyhemoglobin (O2Hb) and deoxyhemoglobin (HHb) from the peri­spinal vascular network. However, there is a lack of clarity about the potential differences in perispinal NVR recorded by the different fNIRS technologies currently available. In this work, the two main noninvasive fNIRS technologies were compared, i.e., LED and LASER-based. The recording of the human peri­spinal NVR induced by non-noxious electrical stimulation of a peripheral nerve was recorded simultaneously at C7 and T10 vertebral levels. The amplitude, rise time, and full width at half maximum duration of the perispinal NVRs were characterized in healthy volunteers and compared between both systems. The main difference was that the LED-based system shows about one order of magnitude higher values of amplitude than the LASER-based system. No statistical differences were found for rise time and for duration parameters (at thoracic level). The comparison of point-to-point wave patterns did not show significant differences between both systems. In conclusion, the peri­spinal NRV response obtained by different fNIRS technologies was reproducible, and only the amplitude showed differences, probably due to the power of the system which should be considered when assessing the human peri­spinal vascular network.

Imaging Assessment of Radiation Therapy-Related Normal Tissue Injury in Children: A PENTEC Visionary Statement.

The Pediatric Normal Tissue Effects in the Clinic (PENTEC) consortium has made significant contributions to understanding and mitigating the adverse effects of childhood cancer therapy. This review addresses the role of diagnostic imaging in detecting, screening, and comprehending radiation therapy-related late effects in children, drawing insights from individual organ-specific PENTEC reports. We further explore how the development of imaging biomarkers for key organ systems, alongside technical advancements and translational imaging approaches, may enhance the systematic application of imaging evaluations in childhood cancer survivors. Moreover, the review critically examines knowledge gaps and identifies technical and practical limitations of existing imaging modalities in the pediatric population. Addressing these challenges may expand access to, minimize the risk of, and optimize the real-world application of, new imaging techniques. The PENTEC team envisions this document as a roadmap for the future development of imaging strategies in childhood cancer survivors, with the overarching goal of improving long-term health outcomes and quality of life for this vulnerable population.

Secondary cerebro-cerebellar and intra-cerebellar dysfunction in cerebellar mutism syndrome.

BACKGROUND: Cerebellar mutism syndrome (CMS) is characterized by deficits of speech, movement, and affect that can occur following tumor removal from the posterior fossa. The role of cerebro-cerebellar tract injuries in the etiology of CMS remains unclear, with recent studies suggesting that cerebro-cerebellar dysfunction may be related to chronic, rather than transient, symptomatology. METHODS: We measured functional connectivity between the cerebellar cortex and functional nodes throughout the brain using fMRI acquired after tumor removal but prior to adjuvant therapy in a cohort of 70 patients diagnosed with medulloblastoma. Surgical lesions were mapped to the infratentorial anatomy, and connectivity with cerebral cortex was tested for statistical dependence on extent of cerebellar outflow pathway injury. RESULTS: CMS diagnosis was associated with an increase in connectivity between the right cerebellar and left cerebral hemisphere, maximally between cerebellum and ventromedial prefrontal cortex (VM-PFC). Connectivity dependence on cerebellar outflow was significant for some speech nodes but not for VM-PFC, suggesting altered input to the cerebellum. Connectivity between posterior regions of cerebellar cortex and ipsilateral dentate nuclei was abnormal in CMS participants, maximally within the right cerebellar hemisphere. CONCLUSIONS: The functional abnormalities we identified are notably upstream of where causal surgical injury is thought to occur, indicating a secondary phenomenon. The VM-PFC is involved in several functions that may be relevant to the symptomatology of CMS, including emotional control and motor learning. We hypothesize that these abnormalities may reflect maladaptive learning within the cerebellum consequent to disordered motor and limbic function by the periaqueductal gray and other critical midbrain targets.

Clinical Potential of Transcranial Focused Ultrasound for Neurorehabilitation in Pediatric Cancer Survivors.

Cancer survivors are at a high risk for treatment-related late effects, particularly neurocognitive impairment in the attention and executive function domains. These can be compounded in pediatric populations still undergoing neural development, which has increased interest in survivorship studies and neurorehabilitation approaches to mitigate these effects. Cognitive training regimens have shown promise as a therapeutic intervention for improving cognitive function. Therapist-guided and computerized training programs with adaptive paradigms have been successfully implemented in pediatric populations, with positive outcomes on attention and working memory. Another interventional approach is neuromodulation to alter plasticity. Transcranial electrical stimulation can modulate cortical surface activity, and cranial nerve stimulation alters autonomic activity in afferent brainstem pathways. However, they are more systemic in nature and have diffuse spatial targeting. Transcranial focused ultrasound (tFUS) modulation overcomes these limitations with high spatial specificity and the ability to target deeper brain regions. In this review, we discuss the efficacy of tFUS for modulating specific brain regions and its potential utility to augment cognitive training programs as a complementary intervention.

Functional MRI Neurofeedback Outperforms Cognitive Behavioral Therapy for Reducing Tinnitus Distress: A Prospective Randomized Clinical Trial.

Background Cognitive behavioral therapy (CBT) is the current standard treatment for chronic severe tinnitus; however, preliminary evidence suggests that real-time functional MRI (fMRI) neurofeedback therapy may be more effective. Purpose To compare the efficacy of real-time fMRI neurofeedback against CBT for reducing chronic tinnitus distress. Materials and Methods In this prospective controlled trial, participants with chronic severe tinnitus were randomized from December 2017 to December 2021 to receive either CBT (CBT group) for 10 weekly group sessions or real-time fMRI neurofeedback (fMRI group) individually during 15 weekly sessions. Change in the Tinnitus Handicap Inventory (THI) score (range, 0-100) from baseline to 6 or 12 months was assessed. Secondary outcomes included four quality-of-life questionnaires (Beck Depression Inventory, Pittsburgh Sleep Quality Index, State-Trait Anxiety Inventory, and World Health Organization Disability Assessment Schedule). Questionnaire scores between treatment groups and between time points were assessed using repeated measures analysis of variance and the nonparametric Wilcoxon signed rank test. Results The fMRI group included 21 participants (mean age, 49 years ± 11.4 [SD]; 16 male participants) and the CBT group included 22 participants (mean age, 53.6 years ± 8.8; 16 male participants). The fMRI group showed a greater reduction in THI scores compared with the CBT group at both 6 months (mean score change, -28.21 points ± 18.66 vs -12.09 points ± 18.86; P = .005) and 12 months (mean score change, -30 points ± 25.44 vs -4 points ± 17.2; P = .01). Compared with baseline, the fMRI group showed improved sleep (mean score, 8.62 points ± 4.59 vs 7.25 points ± 3.61; P = .006) and trait anxiety (mean score, 44 points ± 11.5 vs 39.84 points ± 10.5; P = .02) at 1 month and improved depression (mean score, 13.71 points ± 9.27 vs 6.53 points ± 5.17; P = .01) and general functioning (mean score, 24.91 points ± 17.05 vs 13.06 points ± 10.1; P = .01) at 6 months. No difference in these metrics over time was observed for the CBT group (P value range, .14 to >.99). Conclusion Real-time fMRI neurofeedback therapy led to a greater reduction in tinnitus distress than the current standard treatment of CBT. ClinicalTrials.gov registration no.: NCT05737888; Swiss Ethics registration no.: BASEC2017-00813 © RSNA, 2024 Supplemental material is available for this article.

Odor cueing of declarative memories during sleep enhances coordinated spindles and slow oscillations.

Long-term memories are formed by repeated reactivation of newly encoded information during sleep. This process can be enhanced by using memory-associated reminder cues like sounds and odors. While auditory cueing has been researched extensively, few electrophysiological studies have exploited the various benefits of olfactory cueing. We used high-density electroencephalography in an odor-cueing paradigm that was designed to isolate the neural responses specific to the cueing of declarative memories. We show widespread cueing-induced increases in the duration and rate of sleep spindles. Higher spindle rates were most prominent over centro-parietal areas and largely overlapping with a concurrent increase in the amplitude of slow oscillations (SOs). Interestingly, greater SO amplitudes were linked to a higher likelihood of coupling a spindle and coupled spindles expressed during cueing were more numerous in particular around SO up states. We thus identify temporally and spatially coordinated enhancements of sleep spindles and slow oscillations as a candidate mechanism behind cueing-induced memory processing. Our results further demonstrate the feasibility of studying neural activity patterns linked to such processing using olfactory cueing during sleep.

Hydroxyurea maintains working memory function in pediatric sickle cell disease.

Pediatric patients with sickle cell disease (SCD) have decreased oxygen-carrying capacity in the blood and reduced or restricted cerebral blood flow resulting in neurocognitive deficits and cerebral infarcts. The standard treatment for children with SCD is hydroxyurea; however, the treatment-related neurocognitive effects are unclear. A key area of impairment in SCD is working memory, which is implicated in other cognitive and academic skills. N-back tasks are commonly used to investigate neural correlates of working memory. We analyzed functional magnetic resonance imaging (fMRI) of patients with SCD while they performed n-back tasks by assessing the blood-oxygenation level-dependent (BOLD) signals during working memory processing. Twenty hydroxyurea-treated and 11 control pediatric patients with SCD (7-18 years old) performed 0-, 1-, and 2-back tasks at 2 time points, once before hydroxyurea treatment (baseline) and ~1 year after treatment (follow-up). Neurocognitive measures (e.g., verbal comprehension, processing speed, full-scale intelligence quotient, etc.) were assessed at both time points. Although no significant changes in behavior performance of n-back tasks and neurocognitive measures were observed in the treated group, we observed a treatment-by-time interaction in the right cuneus and angular gyrus for the 2- > 0-back contrast. Through searchlight-pattern classifications in the treated and control groups to identify changes in brain activation between time points during the 2-back task, we found more brain areas, especially the posterior region, with changes in the pattern and magnitude of BOLD signals in the control group compared to the treated group. In the control group, increases in 2-back BOLD signals were observed in the right crus I cerebellum, right inferior parietal lobe, right inferior temporal lobe, right angular gyrus, left cuneus and left middle frontal gyrus at 1-year follow-up. Moreover, BOLD signals elevated as the working memory load increased from 0- to 1-back but did not increase further from 1- to 2-back in the right inferior temporal lobe, right angular gyrus, and right superior frontal gyrus. These observations may result from increased cognitive effort during working memory processing with no hydroxyurea treatment. In contrast, we found fewer changes in the pattern and magnitude of BOLD signals across time points in the treated group. Furthermore, BOLD signals in the left crus I cerebellum, right angular gyrus, left cuneus and right superior frontal gyrus of the treated group increased continuously with increasing working memory load from 0- to 2-back, potentially related to a broader dynamic range in response to task difficulty and cognitive effort. Collectively, these findings suggest that hydroxyurea treatment helped maintain working memory function in SCD.

Self-regulation learning as active inference: dynamic causal modeling of an fMRI neurofeedback task.

Introduction: Learning to self-regulate brain activity by neurofeedback has been shown to lead to changes in the brain and behavior, with beneficial clinical and non-clinical outcomes. Neurofeedback uses a brain-computer interface to guide participants to change some feature of their brain activity. However, the neural mechanism of self-regulation learning remains unclear, with only 50% of the participants succeeding in achieving it. To bridge this knowledge gap, our study delves into the neural mechanisms of self-regulation learning via neurofeedback and investigates the brain processes associated with successful brain self-regulation. Methods: We study the neural underpinnings of self-regulation learning by employing dynamical causal modeling (DCM) in conjunction with real-time functional MRI data. The study involved a cohort of 18 participants undergoing neurofeedback training targeting the supplementary motor area. A critical focus was the comparison between top-down hierarchical connectivity models proposed by Active Inference and alternative bottom-up connectivity models like reinforcement learning. Results: Our analysis revealed a crucial distinction in brain connectivity patterns between successful and non-successful learners. Particularly, successful learners evinced a significantly stronger top-down effective connectivity towards the target area implicated in self-regulation. This heightened top-down network engagement closely resembles the patterns observed in goal-oriented and cognitive control studies, shedding light on the intricate cognitive processes intertwined with self-regulation learning. Discussion: The findings from our investigation underscore the significance of cognitive mechanisms in the process of self-regulation learning through neurofeedback. The observed stronger top-down effective connectivity in successful learners indicates the involvement of hierarchical cognitive control, which aligns with the tenets of Active Inference. This study contributes to a deeper understanding of the neural dynamics behind successful self-regulation learning and provides insights into the potential cognitive architecture underpinning this process.

The hemodynamic initial-dip consists of both volumetric and oxymetric changes reflecting localized spiking activity.

The initial-dip is a transient decrease frequently observed in functional neuroimaging signals, immediately after stimulus onset, believed to originate from a rise in deoxy-hemoglobin (HbR) caused by local neural activity. It has been shown to be more spatially specific than the hemodynamic response, and is believed to represent focal neuronal activity. However, despite being observed in various neuroimaging modalities (such as fMRI, fNIRS, etc), its origins are disputed, and its precise neuronal correlates are unknown. Here we show that the initial-dip is dominated by a decrease in total-hemoglobin (HbT). We also find a biphasic response in deoxy-Hb (HbR), with an early decrease and later rebound. Both the HbT-dip and HbR-rebound were strongly correlated to highly localized spiking activity. However, HbT decreases were always large enough to counter the spiking-induced increase in HbR. We find that the HbT-dip counters spiking induced HbR increases, imposing an upper-limit to HbR concentration in the capillaries. Building on our results, we explore the possibility of active venule dilation (purging) as a possible mechanism for the HbT dip.

Real-Time fMRI Neurofeedback Training as a Neurorehabilitation Approach on Depressive Disorders: A Systematic Review of Randomized Control Trials.

Real-time functional magnetic resonance imaging neurofeedback (rt-fMRI-nf) training is an emerging intervention for neurorehabilitation. However, its translation into clinical use on participants with clinical depression is unclear, the effect estimates from randomized control trials and the certainty of the supporting evidence on the effect estimates are unknown. As the number of studies on neurofeedback increases every year, and better quality evidence becomes available, we evaluate the evidence of all randomized control trials available on the clinical application of rt-fMRI-nf training on participants with clinical depression. We performed electronic searches in Pubmed, Embase, CENTRAL, rtFIN database, Epistemonikos, trial registers, reference lists, other systematic reviews, conference abstracts, and cross-citation in Google Scholar. Reviewers independently selected studies, extracted data and evaluated the risk of bias. The certainty of the evidence was judged using the GRADE framework. This review complies with PRISMA guidelines and was submitted to PROSPERO registration. We found 435 results. After the selection process, we included 11 reports corresponding to four RCTs. The effect of rt-fMRI-nf on improving the severity of clinical depression scores demonstrated a tendency to favor the intervention; however, the general effect was not significant. At end of treatment, SMD (standardized mean difference): -0.32 (95% CI -0.73 to 0.10). At follow-up, SMD: -0.33 (95% CI -0.91, 1.25). All the studies showed changes in BOLD fMRI activation after training; however, only one study confirmed regulation success during a transfer run. Whole-brain analyses suggests that rt-fMRI nf may alter activity patterns in brain networks. More studies are needed to evaluate quality of life, acceptability, adverse effects, cognitive tasks, and physiology measures. We conclude that the current evidence on the effect of rt-fMRI-nf training for decision-making outcomes in patients with clinical depression is still based on low certainty of the evidence.

A real-time fMRI neurofeedback system for the clinical alleviation of depression with a subject-independent classification of brain states: A proof of principle study.

Most clinical neurofeedback studies based on functional magnetic resonance imaging use the patient's own neural activity as feedback. The objective of this study was to create a subject-independent brain state classifier as part of a real-time fMRI neurofeedback (rt-fMRI NF) system that can guide patients with depression in achieving a healthy brain state, and then to examine subsequent clinical changes. In a first step, a brain classifier based on a support vector machine (SVM) was trained from the neural information of happy autobiographical imagery and motor imagery blocks received from a healthy female participant during an MRI session. In the second step, 7 right-handed female patients with mild or moderate depressive symptoms were trained to match their own neural activity with the neural activity corresponding to the "happiness emotional brain state" of the healthy participant. The training (4 training sessions over 2 weeks) was carried out using the rt-fMRI NF system guided by the brain-state classifier we had created. Thus, the informative voxels previously obtained in the first step, using SVM classification and Effect Mapping, were used to classify the Blood-Oxygen-Level Dependent (BOLD) activity of the patients and converted into real-time visual feedback during the neurofeedback training runs. Improvements in the classifier accuracy toward the end of the training were observed in all the patients [Session 4-1 Median = 6.563%; Range = 4.10-27.34; Wilcoxon Test (0), 2-tailed p = 0.031]. Clinical improvement also was observed in a blind standardized clinical evaluation [HDRS CE2-1 Median = 7; Range 2 to 15; Wilcoxon Test (0), 2-tailed p = 0.016], and in self-report assessments [BDI-II CE2-1 Median = 8; Range 1-15; Wilcoxon Test (0), 2-tailed p = 0.031]. In addition, the clinical improvement was still present 10 days after the intervention [BDI-II CE3-2_Median = 0; Range -1 to 2; Wilcoxon Test (0), 2-tailed p = 0.50/ HDRS CE3-2 Median = 0; Range -1 to 2; Wilcoxon Test (0), 2-tailed p = 0.625]. Although the number of participants needs to be increased and a control group included to confirm these findings, the results suggest a novel option for neural modulation and clinical alleviation in depression using noninvasive stimulation technologies.

Sex-Based Differences in Functional Brain Activity During Working Memory in Survivors of Pediatric Acute Lymphoblastic Leukemia.

BACKGROUND: Long-term survivors of pediatric acute lymphoblastic leukemia are at elevated risk for neurocognitive deficits and corresponding brain dysfunction. This study examined sex-based differences in functional neuroimaging outcomes in acute lymphoblastic leukemia survivors treated with chemotherapy alone. METHODS: Functional magnetic resonance imaging (fMRI) and neurocognitive testing were obtained in 123 survivors (46% male; median [min-max] age = 14.2 years [8.3-26.5 years]; time since diagnosis = 7.7 years [5.1-12.5 years]) treated on the St. Jude Total XV treatment protocol. Participants performed the n-back working memory task in a 3 T scanner. Functional neuroimaging data were processed (realigned, slice time corrected, normalized, smoothed) and analyzed using statistical parametric mapping with contrasts for 1-back and 2-back conditions, which reflect varying degrees of working memory and task load. Group-level fMRI contrasts were stratified by sex and adjusted for age and methotrexate exposure. Statistical tests were 2-sided (P < .05 statistical significance threshold). RESULTS: Relative to males, female survivors exhibited less activation (ie, reduced blood oxygen dependent-level signals) in the right parietal operculum, supramarginal gyrus and inferior occipital gyrus, and bilateral superior frontal medial gyrus during increased working memory load (family-wise error-corrected P = .004 to .008, adjusting for age and methotrexate dose). Female survivors were slower to correctly respond to the 2-back condition than males (P < .05), though there were no differences in overall accuracy. Performance accuracy was negatively correlated with fMRI activity in female survivors (Pearson's r = -0.39 to -0.29, P = .001 to .02), but not in males. CONCLUSIONS: These results suggest the working memory network is more impaired in female survivors than male survivors, which may contribute to ongoing functional deficits.

Explainable artificial intelligence model to predict brain states from fNIRS signals.

Objective: Most Deep Learning (DL) methods for the classification of functional Near-Infrared Spectroscopy (fNIRS) signals do so without explaining which features contribute to the classification of a task or imagery. An explainable artificial intelligence (xAI) system that can decompose the Deep Learning mode's output onto the input variables for fNIRS signals is described here. Approach: We propose an xAI-fNIRS system that consists of a classification module and an explanation module. The classification module consists of two separately trained sliding window-based classifiers, namely, (i) 1-D Convolutional Neural Network (CNN); and (ii) Long Short-Term Memory (LSTM). The explanation module uses SHAP (SHapley Additive exPlanations) to explain the CNN model's output in terms of the model's input. Main results: We observed that the classification module was able to classify two types of datasets: (a) Motor task (MT), acquired from three subjects; and (b) Motor imagery (MI), acquired from 29 subjects, with an accuracy of over 96% for both CNN and LSTM models. The explanation module was able to identify the channels contributing the most to the classification of MI or MT and therefore identify the channel locations and whether they correspond to oxy- or deoxy-hemoglobin levels in those locations. Significance: The xAI-fNIRS system can distinguish between the brain states related to overt and covert motor imagery from fNIRS signals with high classification accuracy and is able to explain the signal features that discriminate between the brain states of interest.

Machine learning to investigate superficial white matter integrity in early multiple sclerosis.

BACKGROUND AND PURPOSE: This study aims todetermine the sensitivity of superficial white matter (SWM) integrity as a metric to distinguish early multiple sclerosis (MS) patients from healthy controls (HC). METHODS: Fractional anisotropy and mean diffusivity (MD) values from SWM bundles across the cortex and major deep white matter (DWM) tracts were extracted from 29 early MS patients and 31 age- and sex-matched HC. Thickness of 68 cortical regions and resting-state functional-connectivity (RSFC) among them were calculated. The distribution of structural and functional metrics between groups were compared using Wilcoxon rank-sum test. Utilizing a machine learning method (adaptive boosting), 6 models were built based on: 1-SWM, 2-DWM, 3-SWM and DWM, 4-cortical thickness, or 5-RSFC measures. In model 6, all features from previous models were incorporated. The models were trained with nested 5-folds cross-validation. Area under the receiver operating characteristic curve (AUCroc ) values were calculated to evaluate classification performance of each model. Permutation tests were used to compare the AUCroc values. RESULTS: Patients had higher MD in SWM bundles including insula, inferior frontal, orbitofrontal, superior and medial temporal, and pre- and post-central cortices (p < .05). No group differences were found for any other MRI metric. The model incorporating SWM and DWM features provided the best classification (AUCroc = 0.75). The SWM model provided higher AUCroc (0.74), compared to DWM (0.63), cortical thickness (0.67), RSFC (0.63), and all-features (0.68) models (p < .001 for all). CONCLUSION: Our results reveal a non-random pattern of SWM abnormalities at early stages of MS even before pronounced structural and functional alterations emerge.