Jian Li (Andrew)

Jian Li (Andrew)

Instructor

       

Google Scholar Page

  doi/url       pdf       open-source code / software       bib       video abstract

† co-first authors;     ‡ co-senior authors.

Pre-prints

2024

  • M. K. Cambareri, L. Lewis, A. Horn, J. Li‡, B. L. Edlow‡,
    “Subcortical hubs of brain nentworks sustaining human consciousness”,
    bioRxiv, 2024.      
    Abstract

    Neuromodulation of subcortical network hubs by pharmacologic, electrical, or ultrasonic stimulation is a promising therapeutic strategy for patients with disorders of consciousness (DoC). However, optimal subcortical targets for therapeutic stimulation are not well established. Here, we leveraged 7 Tesla resting-state functional MRI (rs-fMRI) data from 168 healthy subjects from the Human Connectome Project to map the subcortical connectivity of six canonical cortical networks that modulate higher-order cognition and function: the default mode, executive control, salience, dorsal attention, visual, and somatomotor networks. Based on spatiotemporally overlapped networks generated by the Nadam-Accelerated SCAlable and Robust (NASCAR) tensor decomposition method, our goal was to identify subcortical hubs that are functionally connected to multiple cortical networks. We found that the ventral tegmental area (VTA) in the midbrain and the central lateral (CL) and parafascicular (Pf) nuclei of the thalamus – regions that have historically been targeted by neuromodulatory therapies to restore consciousness – are subcortical hubs widely connected to multiple cortical networks. Further, we identified a subcortical hub in the pontomesencephalic tegmentum that overlapped with multiple reticular and extrareticular arousal nuclei and that encompassed a well-established "hot spot" for coma-causing brainstem lesions. Multiple hubs within the brainstem arousal nuclei and thalamic intralaminar nuclei were functionally connected to both the DMN and SN, emphasizing the importance of these cortical networks in integrative subcortico-cortical signaling. Additional subcortical connectivity hubs were observed within the caudate head, putamen, amygdala, hippocampus, and bed nucleus of the stria terminalis, regions classically associated with modulation of cognition, behavior, and sensorimotor function. Collectively, these results suggest that multiple subcortical hubs in the brainstem tegmentum, thalamus, basal ganglia, and medial temporal lobe modulate cortical function in the human brain. Our findings strengthen the evidence for targeting subcortical hubs in the ventral tegmental area, central lateral thalamus, and pontomesencephalic tegmentum to restore consciousness in patients with DoC. We release all subcortical connectivity maps to support ongoing efforts at therapeutic neuromodulation of consciousness.

  • Warren, M. Raguz̆, H. Friedrich, Schaper, J. Tasserie, S. B. Snider, J. Li, Chua, K. Butenko, M. Friedrich, R. Jha, J. E. Iglesias, P. W. Carney, M. D. Fox, A. D. Boes, B. L. Edlow, A. Horn, D. Chudy‡, J. D. Rolston‡,
    “A human brain network linked to restoration of consciousness after deep brain stimulation”,
    medRxiv, 2024.      
    Abstract

    Subcortical arousal systems are known to play a key role in controlling sustained changes in attention and conscious awareness. Recent studies indicate that these systems have a major influence on short-term dynamic modulation of visual attention, but their role across sensory modalities is not fully understood. In this study, we investigated shared subcortical arousal systems across sensory modalities during transient changes in attention using block and event-related fMRI paradigms. We analyzed massive publicly available fMRI datasets collected while 1,561 participants performed visual, auditory, tactile, and taste perception tasks. Our analyses revealed a shared circuit of subcortical arousal systems exhibiting early transient increases in activity in midbrain reticular formation and central thalamus across perceptual modalities, as well as less consistent increases in pons, hypothalamus, basal forebrain, and basal ganglia. Identifying these networks is critical for understanding mechanisms of normal attention and consciousness and may help facilitate subcortical targeting for therapeutic neuromodulation.

  • A. Khalaf, E. Lopez, J. Li, A. Horn, B. L. Edlow, H. Blumenfeld,
    “Shared subcortical arousal systems across sensory modalities during transient modulation of attention”,
    bioRxiv, 2024.      
    Abstract

    Subcortical arousal systems are known to play a key role in controlling sustained changes in attention and conscious awareness. Recent studies indicate that these systems have a major influence on short-term dynamic modulation of visual attention, but their role across sensory modalities is not fully understood. In this study, we investigated shared subcortical arousal systems across sensory modalities during transient changes in attention using block and event-related fMRI paradigms. We analyzed massive publicly available fMRI datasets collected while 1,561 participants performed visual, auditory, tactile, and taste perception tasks. Our analyses revealed a shared circuit of subcortical arousal systems exhibiting early transient increases in activity in midbrain reticular formation and central thalamus across perceptual modalities, as well as less consistent increases in pons, hypothalamus, basal forebrain, and basal ganglia. Identifying these networks is critical for understanding mechanisms of normal attention and consciousness and may help facilitate subcortical targeting for therapeutic neuromodulation.

  • H. J. Freeman, A. S. Atalay, J. Li, E. Sobczak, S. B. Snider, H. Carrington, E. Selmanovic, A. Pruyser, L. Bura, D. Sheppard, D. Hunt, A. C. Seifert, Y. G. Bodien, J. M. Hoffman, C. L. MacDonald, K. Dams-O’Connor‡, B. L. Edlow‡,
    “Longitudinal lesion expansion in chronic traumatic brain injury”,
    medRxiv, 2024.      
    Abstract

    Traumatic brain injury (TBI) is a risk factor for neurodegeneration and cognitive decline, yet the underlying pathophysiologic mechanisms are incompletely understood. This gap in knowledge is in part related to the lack of analytic methods to account for cortical lesions in prior neuroimaging studies. The objective of this study was to develop a lesion detection tool and apply it to an investigation of longitudinal changes in brain structure among individuals with chronic TBI. We identified 24 individuals with chronic moderate-to-severe TBI enrolled in the Late Effects of TBI (LETBI) study who had cortical lesions detected by T1-weighted MRI at two time points. Initial MRI scans were performed more than 1-year post-injury and follow-up scans were performed 3.1 (IQR=1.7) years later. We leveraged FreeSurfer parcellations of T1-weighted MRI volumes and a recently developed super-resolution technique, SynthSR, to identify cortical lesions in this longitudinal dataset. Trained raters received the data in a randomized order and manually corrected the automated lesion segmentation, yielding a final lesion mask for each scan at each timepoint. Lesion volume significantly increased between the two time points with a median volume change of 3.2 (IQR=5.9) mL (p<0.001), and the increases significantly exceeded the possible variance in lesion volume changes due to manual tracing errors (p < 0.001). Lesion volume significantly expanded longitudinally in 23 of 24 subjects, with all FDR corrected p-values ≤ 0.02. Inter-scan duration was not associated with the magnitude of lesion growth. We also demonstrated that the semi-automated tool showed a high level of accuracy compared to “ground truth” manual lesion segmentation. Semi-automated lesion segmentation is feasible in TBI studies and creates opportunities to elucidate mechanisms of post-traumatic neurodegeneration.

  • Y. Liu, J. Li, J. L. Wisnowski‡, R. M. Leahy‡,
    “Graph learning for cortical parcellation from tensor decompositions of resting-state fMRI”,
    bioRxiv, 2024.      
    Abstract

    Cortical parcellation has long been a cornerstone in the field of neuroscience, enabling the cerebral cortex to be partitioned into distinct, non-overlapping regions that facilitate the interpretation and comparison of complex neuroscientific data. In recent years, these parcellations have frequently been based on the use of resting-state fMRI (rsfMRI) data. In parallel, methods such as independent components analysis have long been used to identify large-scale functional networks with significant spatial overlap between networks. Despite the fact that both forms of decomposition make use of the same spontaneous brain activity measured with rsfMRI, a gap persists in establishing a clear relationship between disjoint cortical parcellations and brain-wide networks. To address this, we introduce a novel parcellation framework that integrates NASCAR, a three-dimensional tensor decomposition method that identifies a series of functional brain networks, with state-of-the-art graph representation learning to produce cortical parcellations that represent near-homogeneous functional regions that are consistent with these brain networks. Further, through the use of the tensor decomposition, we avoid the limitations of traditional approaches that assume statistical independence or orthogonality in defining the underlying networks. Our findings demonstrate that these parcellations are comparable or superior to established atlases in terms of homogeneity of the functional connectivity across parcels, task contrast alignment, and architectonic map alignment. Our methodological pipeline is highly automated, allowing for rapid adaptation to new datasets and the generation of custom parcellations in just minutes, a significant advancement over methods that require extensive manual input. We describe this integrated approach, which we refer to as Untamed, as a tool for use in the fields of cognitive and clinical neuroscientific research. Parcellations created from the Human Connectome Project dataset using Untamed, along with the code to generate atlases with custom parcel numbers, are publicly available at https://untamed-atlas.github.io.

Journal Publications

2024

  • J. Li, G. Tuckute, E. Fedorenko, B. L. Edlow, A. V. Dalca‡, B. Fischl‡,
    “JOSA: Joint surface-based registration and atlas construction of brain geometry and function”,
    Medical Image Analysis, , p. 103292, 2024.      
    Abstract

    Surface-based cortical registration is an important topic in medical image analysis and facilitates many downstream applications. Current approaches for cortical registration are mainly driven by geometric features, such as sulcal depth and curvature, and often assume that registration of folding patterns leads to alignment of brain function. However, functional variability of anatomically corresponding areas across subjects has been widely reported, particularly in higher-order cognitive areas. In this work, we present JOSA, a novel cortical registration framework that jointly models the mismatch between geometry and function while simultaneously learning an unbiased population-specific atlas. Using a semi-supervised training strategy, JOSA achieves superior registration performance in both geometry and function to the state-of-the-art methods but without requiring functional data at inference. This learning framework can be extended to any auxiliary data to guide spherical registration that is available during training but is difficult or impossible to obtain during inference, such as parcellations, architectonic identity, transcriptomic information, and molecular profiles. By recognizing the mismatch between geometry and function, JOSA provides new insights into the future development of registration methods using joint analysis of brain structure and function.

  • J. R. Stieger, P. Pinheiro-Chagas, Y. Fang, J. Li, Z. Lusk, C. M. Perry, D. Contreras, Q. Chen, J. Hugenard, B. L. Edlow, A. Wagner, V. Buch, J. Parvizi,
    “Cross-regional coordination of activity in the human brain during autobiographical self-referential processing”,
    Proceedings of the National Academy of Sciences, vol. 121, no. 32, p. e2316021121, 2024.    
    Abstract

    We recorded simultaneously from the human thalamus, hippocampus, posteromedial cortex, and orbitofrontal cortex during two autobiographical memory experiments and mapped their profile of activity and connectivity with millisecond resolution. Our findings provide significant insight into fast spatiotemporal dynamics of activity across distant structures of the brain as individuals are engaged in self-referential memory processing. For the human brain to operate, populations of neurons across anatomical structures must coordinate their activity within milliseconds. To date, our understanding of such interactions has remained limited. We recorded directly from the hippocampus (HPC), posteromedial cortex (PMC), ventromedial/orbital prefrontal cortex (OFC), and the anterior nuclei of the thalamus (ANT) during two experiments of autobiographical memory processing that are known from decades of neuroimaging work to coactivate these regions. In 31 patients implanted with intracranial electrodes, we found that the presentation of memory retrieval cues elicited a significant increase of low frequency (LF<6 Hz) activity followed by cross-regional phase coherence of this LF activity before select populations of neurons within each of the four regions increased high-frequency (HF>70 Hz) activity. The power of HF activity was modulated by memory content, and its onset followed a specific temporal order of ANT→HPC/PMC→OFC. Further, we probed cross-regional causal effective interactions with repeated electrical pulses and found that HPC stimulations cause the greatest increase in LF-phase coherence across all regions, whereas the stimulation of any region caused the greatest LF-phase coherence between that particular region and ANT. These observations support the role of the ANT in gating, and the HPC in synchronizing, the activity of cortical midline structures when humans retrieve self-relevant memories of their past. Our findings offer a fresh perspective, with high temporal fidelity, about the dynamic signaling and underlying causal connections among distant regions when the brain is actively involved in retrieving self-referential memories from the past.

  • B. L. Edlow, M. Olchanyi, H. J. Freeman, J. Li, C. Maffei, S. B. Snider, L. Zöllei, J. E. Iglesias, J. Augustinack, Y. G. Bodien, R. L. Haynes, D. N. Greve, B. R. Diamond, A. Stevens, J. T. Giacino, C. Destrieux, A. Kouwe, E. N. Brown, F. R. D., B. Fischl, H. C. Kinney,
    “Multimodal MRI reveals brainstem connections that sustain wakefulness in human consciousness”,
    Science Translational Medicine, vol. 16, no. 745, p. eadj4303, 2024.      
    Abstract

    Consciousness is composed of arousal (i.e., wakefulness) and awareness. Substantial progress has been made in mapping the cortical networks that underlie awareness in the human brain, but knowledge about the subcortical networks that sustain arousal in humans is incomplete. Here, we aimed to map the connectivity of a proposed subcortical arousal network that sustains wakefulness in the human brain, analogous to the cortical default mode network (DMN) that has been shown to contribute to awareness. We integrated data from ex vivo diffusion magnetic resonance imaging (MRI) of three human brains, obtained at autopsy from neurologically normal individuals, with immunohistochemical staining of subcortical brain sections. We identified nodes of the proposed default ascending arousal network (dAAN) in the brainstem, hypothalamus, thalamus, and basal forebrain. Deterministic and probabilistic tractography analyses of the ex vivo diffusion MRI data revealed projection, association, and commissural pathways linking dAAN nodes with one another and with DMN nodes. Complementary analyses of in vivo 7-tesla resting-state functional MRI data from the Human Connectome Project identified the dopaminergic ventral tegmental area in the midbrain as a widely connected hub node at the nexus of the subcortical arousal and cortical awareness networks. Our network-based autopsy methods and connectivity data provide a putative neuroanatomic architecture for the integration of arousal and awareness in human consciousness.

  • H. Shaker, J. Li, M. Kobayashi, O. Grinenko, I. Najm, J. Bulacio, R. Leahy, P. Chauvel,
    “Is high-frequency activity at seizure onset inhibitory: An stereoelectroencephalographic study of motor cortex seizures”,
    Annals of Neurology, 2024.      
    Abstract

    Objective: In the era of stereoelectroencephalography (SEEG), many studies have been devoted to understanding the role of interictal high-frequency oscillations. High-frequency activity (HFA) at seizure onset has been identified as a marker of epileptogenic zone. We address the physiological significance of ictal HFAs and their relation to clinical semiology. Methods: We retrospectively identified patients with pure focal primary motor epilepsy. We selected only patients in whom SEEG electrodes were optimally placed in the motor cortex as confirmed by electrical stimulation. Based on these narrow inclusion criteria, we extensively studied 5 patients (3 males and 2 females, mean age = 22.4 years) using time-frequency analysis and time correlation with motor signs onset. Results: A total of 157 analyzable seizures were recorded in 5 subjects. The first 2 subjects had tonic or clonic semiology with rare secondary generalization. Subject 3 had atonic onset followed by clonic hand/arm flexion. Subject 4 had clusters of tonic and atonic facial movements. Subject 5 had upper extremity tonic movements. The median frequency of the fast activity extracted from the Epileptogenic Zone Fingerprint pipeline in the first 4 subjects was 76 Hz (interquartile range = 21.9 Hz). Positive motor signs did not occur concomitantly with high gamma activity developing in the motor cortex. Motor signs began at the end of HFAs. Interpretation: This study supports the hypothesis of an inhibitory effect of ictal HFAs. The frequency range in the gamma band was associated with the direction of the clinical output effect. Changes from inhibitory to excitatory effect occurred when discharge frequency dropped to low gamma or beta.

2023

  • T. Aung†, O. Grinenko†, J. Li, J. Mosher, P. Chauvel, J. Martinez-Gonzalez,
    “Stereo-electroencephalography guided laser ablation in neocortical epilepsy: Electrophysiological correlations and outcome”,
    Epilepsia, 2023.      
    Abstract

    Objective: We aim to study the correlation between seizure outcomes in patients with drug-resistant epilepsy (DRE) who underwent laser interstitial thermal therapy (LITT) and stereo-electroencephalographic (SEEG) electrophysiologic patterns with respect to the extent of laser ablation. Methods: We retrospectively analyzed 16 consecutive DRE patients who underwent LITT. A seizure onset zone (SOZ) was obtained from multidisciplinary patient management conferences and again was confirmed independently by two epileptologists based on conventional analysis. SOZ were retrospectively divided into localized, lobar and multilobar, and nonlocalized onset types. A posteriori predicted epileptogenic zone (PEZ) was identified using the previously developed “EZ fingerprint” pipeline. The completeness of the SOZ and PEZ ablation was compared and correlated with the duration of seizure freedom (SF). Results: Out of 16 patients, 11 had a posteriori-identified PEZ. Three patients underwent complete ablation of SOZ with curative intent, and the other 13 with palliative intent. Out of three patients with complete ablation of the SOZ, two had concordant PEZ and SOZ and achieved 40- and 46-month SF without seizure recurrence. The remaining patient, without any PEZ identified, had seizure recurrence within one month. Six out of 13 patients with partial ablation of the SOZ and PEZ achieved a mean SF of 19.8 months (range: 1 to 44) with subsequent seizure recurrence. The remaining seven patients had partial ablation of the SOZ without the PEZ identified or ablation outside the PEZ with seizure recurrence within one to two months, except one patient who had 40 months SF after ablation of periventricular heterotopia. Significance: Only complete ablation of the well-restricted SOZ concordant with the PEZ was associated with long-term SF, whereas partial ablation of the PEZ might lead to SF with eventual seizure recurrence. Failure to identify PEZ and ablation limited to the SOZ often led to one to two months of SF.

  • B. L. Edlow, M. Fecchio, Y. G. Bodien, A. Comanducci, M. Rosanova, S. Casarotto, M. J. Young, J. Li, D. D. Dougherty, C. Koch, G. Tononi, M. Massimini, M. Boly,
    “Measuring consciousness in the intensive care unit”,
    Neurocritical Care, , no. 38, pp. 584–590, 2023.      
    Abstract

    Early reemergence of consciousness predicts long-term functional recovery for patients with severe brain injuries. However, tools to reliably detect consciousness in the intensive care unit (ICU) are lacking. Transcranial magnetic stimulation-electroencephalography has the potential to detect consciousness in the ICU, predict recovery, and prevent premature withdrawal of life-sustaining therapy.

  • J. Li, Y. Liu, J. L. Wisnowski‡, R. M. Leahy‡,
    “Identification of overlapping and interacting networks reveals intrinsic spatiotemporal organization of the human brain”,
    NeuroImage, vol. 270, p. 119944, 2023.        
    Abstract  

    The human brain is a complex network that exhibits dynamic fluctuations in activity across space and time. Depending on the analysis method, canonical brain networks identified from resting-state fMRI (rs-fMRI) are typically constrained to be either orthogonal or statistically independent in their spatial and/or temporal domains. We avoid imposing these potentially unnatural constraints through the combination of a temporal synchronization process ("BrainSync") and a three-way tensor decomposition method ("NASCAR") to jointly analyze rs-fMRI data from multiple subjects. The resulting set of interacting networks comprises minimally constrained spatiotemporal distributions, each representing one component of functionally coherent activity across the brain. We show that these networks can be clustered into six distinct functional categories and naturally form a representative functional network atlas for a healthy population. This functional network atlas could help explore group and individual differences in neurocognitive function, as we demonstrate in the context of ADHD and IQ prediction.

2022

  • L. Benschop, G. Vanhollebeke, J. Li, M. Vanderhasselt, R. M. Leahy, C. Baeken,
    “Reduced subgenual cingulate–dorsolateral prefrontal connectivity as an electrophysiological marker for depression”,
    Scientific Report, vol. 12, no. 1, p. 16903, 2022.      
    Abstract

    Major Depressive Disorder (MDD) is a widespread mental illness that causes considerable suffering, and neuroimaging studies are trying to reduce this burden by developing biomarkers that can facilitate detection. Prior fMRI- and neurostimulation studies suggest that aberrant subgenual Anterior Cingulate (sgACC)—dorsolateral Prefrontal Cortex (DLPFC) functional connectivity is consistently present within MDD. Combining the need for reliable depression markers with the electroencephalogram’s (EEG) high clinical utility, we investigated whether aberrant EEG sgACC–DLPFC functional connectivity could serve as a marker for depression. Source-space Amplitude Envelope Correlations (AEC) of 20 MDD patients and 20 matched controls were contrasted using non-parametric permutation tests. In addition, extracted AEC values were used to (a) correlate with characteristics of depression and (b) train a Support Vector Machine (SVM) to determine sgACC–DLPFC connectivity’s discriminative power. FDR-thresholded statistical maps showed reduced sgACC–DLPFC AEC connectivity in MDD patients relative to controls. This diminished AEC connectivity is located in the beta-1 (13–17 Hz) band and is associated with patients’ lifetime number of depressive episodes. Using extracted sgACC–DLPFC AEC values, the SVM achieved a classification accuracy of 84.6% (80% sensitivity and 89.5% specificity) indicating that EEG sgACC–DLPFC connectivity has promise as a biomarker for MDD.

  • R. Setton†, L. Mwilambwe-Tshilobo†, M. Girn, A. W. Lockrow, G. Baracchini, A. J. Lowe, B. N. Cassidy, J. Li, R. M. Leahy, T. Ge, D. S. Margulies, B. Misic, B. C. Bernhardt, W. D. Stevens, F. De Brigard, P. Kundu, G. R. Turner, R. N. Spreng,
    “Age differences in the functional architecture of the human brain”,
    Cerebral Cortex, vol. 33, no. 1, pp. 114–134, 2022.      
    Abstract

    The intrinsic functional organization of the brain changes into older adulthood. Age differences are observed at multiple spatial scales, from global reductions in modularity and segregation of distributed brain systems, to network-specific patterns of dedifferentiation. Whether dedifferentiation reflects an inevitable, global shift in brain function with age, circumscribed, experience-dependent changes, or both, is uncertain. We employed a multimethod strategy to interrogate dedifferentiation at multiple spatial scales. Multi-echo (ME) resting-state fMRI was collected in younger (n=181) and older (n=120) healthy adults. Cortical parcellation sensitive to individual variation was implemented for precision functional mapping of each participant while preserving group-level parcel and network labels. ME-fMRI processing and gradient mapping identified global and macroscale network differences. Multivariate functional connectivity methods tested for microscale, edge-level differences. Older adults had lower BOLD signal dimensionality, consistent with global network dedifferentiation. Gradients were largely age-invariant. Edge-level analyses revealed discrete, network-specific dedifferentiation patterns in older adults. Visual and somatosensory regions were more integrated within the functional connectome; default and frontoparietal control network regions showed greater connectivity; and the dorsal attention network was more integrated with heteromodal regions. These findings highlight the importance of multiscale, multimethod approaches to characterize the architecture of functional brain aging.

2021

  • H. Akrami, A. A. Joshi, J. Li, S. Aydore, R. M. Leahy,
    “A robust variational autoencoder using beta divergence”,
    Knowledge-Based Systems, vol. 238, p. 107886, 2021.    
    Abstract

    The presence of outliers can severely degrade learned representations and performance of deep learning methods and hence disproportionately affect the training process, leading to incorrect conclusions about the data. For example, anomaly detection using deep generative models is typically only possible when similar anomalies (or outliers) are not present in the training data. Here we focus on variational autoencoders (VAEs). While the VAE is a popular framework for anomaly detection tasks, we observe that the VAE is unable to detect outliers when the training data contains anomalies that have the same distribution as those in test data. In this paper we focus on robustness to outliers in training data in VAE settings using concepts from robust statistics. We propose a variational lower bound that leads to a robust VAE model that has the same computational complexity as the standard VAE and contains a single automatically-adjusted tuning parameter to control the degree of robustness. We present mathematical formulations for robust variational autoencoders (RVAEs) for Bernoulli, Gaussian and categorical variables. The RVAE model is based on beta-divergence rather than the standard Kullback–Leibler (KL) divergence. We demonstrate the performance of our proposed beta-divergence based autoencoder for a variety of image and categorical datasets showing improved robustness to outliers both qualitatively and quantitatively. We also illustrate the use of our robust VAE for detection of lesions in brain images, formulated as an anomaly detection task. Finally, we suggest a method to tune the hyperparameter of RVAE which makes our model completely unsupervised.

  • J. Li, W. H. Curley, B. Guerin, D. D. Dougherty, A. V. Dalca, B. Fischl, A. Horn, B. L. Edlow,
    “Mapping the subcortical connectivity of the human default mode network”,
    NeuroImage, vol. 245, p. 118758, 2021.        
    Abstract  

    The default mode network (DMN) mediates self-awareness and introspection, core components of human consciousness. Therapies to restore consciousness in patients with severe brain injuries have historically targeted subcortical sites in the brainstem, thalamus, hypothalamus, basal forebrain, and basal ganglia, with the goal of reactivating cortical DMN nodes. However, the subcortical connectivity of the DMN has not been fully mapped and optimal subcortical targets for therapeutic neuromodulation of consciousness have not been identified. In this work, we created a comprehensive map of DMN subcortical connectivity by combining high-resolution functional and structural datasets with advanced signal processing methods. We analyzed 7 Tesla resting-state functional MRI (rs-fMRI) data from 168 healthy volunteers acquired in the Human Connectome Project. The rs-fMRI blood-oxygen-level-dependent (BOLD) data were temporally synchronized across subjects using the BrainSync algorithm. Cortical and subcortical DMN nodes were jointly analyzed and identified at the group level by applying a novel Nadam-Accelerated SCAlable and Robust (NASCAR) tensor decomposition method to the synchronized dataset. The subcortical connectivity map was then overlaid on a 7 Tesla 100 micron ex vivo MRI dataset for neuroanatomic analysis using automated segmentation of nuclei within the brainstem, thalamus, hypothalamus, basal forebrain, and basal ganglia. We further compared the NASCAR subcortical connectivity map with its counterpart generated from canonical seed-based correlation analyses. The NASCAR method revealed that BOLD signal in the central lateral nucleus of the thalamus and ventral tegmental area of the midbrain is strongly correlated with that of the DMN. In an exploratory analysis, additional subcortical sites in the median and dorsal raphe, lateral hypothalamus, and caudate nuclei were correlated with the cortical DMN. We also found that the putamen and globus pallidus are negatively correlated (i.e., anti-correlated) with the DMN, providing rs-fMRI evidence for the mesocircuit hypothesis of human consciousness, whereby a striatopallidal feedback system modulates anterior forebrain function via disinhibition of the central thalamus. Seed-based analyses yielded similar subcortical DMN connectivity, but the NASCAR result showed stronger contrast and better spatial alignment with dopamine immunostaining data. The DMN subcortical connectivity map identified here advances understanding of the subcortical regions that contribute to human consciousness and can be used to inform the selection of therapeutic targets in clinical trials for patients with disorders of consciousness.

  • J. Li, J. L. Wisnowski, A. A. Joshi, R. M. Leahy,
    “Robust brain network identification from multi-subject asynchronous fMRI data”,
    NeuroImage, vol. 227, p. 117615, 2021.        
    Abstract

    We describe a novel method for robust identification of common brain networks and their corresponding temporal dynamics across subjects from asynchronous functional MRI (fMRI) using tensor decomposition. We first temporally align asynchronous fMRI data using the orthogonal BrainSync transform, allowing us to study common brain networks across sessions and subjects. We then map the synchronized fMRI data into a 3D tensor (vertices × time × subject/session). Finally, we apply Nesterov-accelerated adaptive moment estimation (Nadam) within a scalable and robust sequential Canonical Polyadic (CP) decomposition framework to identify a low rank tensor approximation to the data. As a result of CP tensor decomposition, we successfully identified twelve known brain networks with their corresponding temporal dynamics from 40 subjects using the Human Connectome Project's language task fMRI data without any prior information regarding the specific task designs. Seven of these networks show distinct subjects’ responses to the language task with differing temporal dynamics; two show sub-components of the default mode network that exhibit deactivation during the tasks; the remaining three components reflect non-task-related activities. We compare results to those found using group independent component analysis (ICA) and canonical ICA. Bootstrap analysis demonstrates increased robustness of networks found using the CP tensor approach relative to ICA-based methods.

2020

  • J. Li, S. Choi, A. A. Joshi, J. L. Wisnowski, R. M. Leahy,
    “Temporal non-local means filtering for studies of intrinsic brain connectivity from individual resting fMRI”,
    Medical Image Analysis, vol. 61, p. 101635, 2020.        
    Abstract

    Characterizing functional brain connectivity using resting functional magnetic resonance imaging (fMRI) is challenging due to the relatively small Blood-Oxygen-Level Dependent contrast and low signal-to-noise ratio. Denoising using surface-based Laplace-Beltrami (LB) or volumetric Gaussian filtering tends to blur boundaries between different functional areas. To overcome this issue, a time-based Non-Local Means (tNLM) filtering method was previously developed to denoise fMRI data while preserving spatial structure. The kernel and parameters that define the tNLM filter need to be optimized for each application. Here we present a novel Global PDF-based tNLM filtering (GPDF) algorithm that uses a data-driven kernel function based on a Bayes factor to optimize filtering for spatial delineation of functional connectivity in resting fMRI data. We demonstrate its performance relative to Gaussian spatial filtering and the original tNLM filtering via simulations. We also compare the effects of GPDF filtering against LB filtering using individual in-vivo resting fMRI datasets. Our results show that LB filtering tends to blur signals across boundaries between adjacent functional regions. In contrast, GPDF filtering enables improved noise reduction without blurring adjacent functional regions. These results indicate that GPDF may be a useful preprocessing tool for analyses of brain connectivity and network topology in individual fMRI recordings.

  • K. N. Taylor, A. A. Joshi, J. Li, J. A. Gonzalez-Martinez, X. Wang, R. M. Leahy, D. R. Nair, J. C. Mosher,
    “The FAST graph: A novel framework for the anatomically-guided visualization and analysis of cortico-cortical evoked potentials”,
    Epilepsy Research, vol. 161, p. 106264, 2020.    
    Abstract

    Background: Intracerebral electroencephalography (iEEG) using stereoelectroencephalography (SEEG) methodology for epilepsy surgery gives rise to complex data sets. The neurophysiological data obtained during the in-patient period includes categorization of the evoked potentials resulting from direct electrical cortical stimulation such as cortico-cortical evoked potentials (CCEPs). These potentials are recorded by hundreds of contacts, making these waveforms difficult to quickly interpret over such high-density arrays that are organized in three dimensional fashion. New Method: The challenge in analyzing CCEPs data arises not just from the density of the array, but also from the stimulation of a number of different intracerebral sites. A systematic methodology for visualization and analysis of these evoked data is lacking. We describe the process of incorporating anatomical information into the visualizations, which are then compared to more traditional plotting techniques to highlight the usefulness of the new framework. Results: We describe here an innovative framework for sorting, registering, labeling, ordering, and quantifying the functional CCEPs data, using the anatomical labelling of the brain, to provide an informative visualization and summary statistics which we call the "FAST graph" (Functional-Anatomical STacked area graphs). The fast graph analysis is used to depict the significant CCEPs responses in patient with focal epilepsy. Conclusions: The novel plotting approach shown here allows us to visualize high-density stimulation data in a single summary plot for subsequent detailed analyses. Improving the visual presentation of complex data sets aides in enhancing the clinical utility of the data.

  • J. Li, O. Grinenko, J. C. Mosher, J. Gonzalez-Martinez, R. M. Leahy, P. Chauvel,
    “Learning to define an electrical biomarker of the epileptogenic zone”,
    Human Brain Mapping, vol. 41, no. 2, pp. 429–441, 2020.        
    Abstract  

    The role of fast activity as a potential biomarker in localization of the epileptogenic zone (EZ) remains controversial due to recently reported unsatisfactory performance. We recently identified a "fingerprint" of the EZ as a time-frequency pattern that is defined by a combination of preictal spike(s), fast oscillatory activity, and concurrent suppression of lower frequencies. Here we examine the generalizability of the fingerprint in application to an independent series of patients (11 seizure-free and 13 non-seizure-free after surgery) and show that the fingerprint can also be identified in seizures with lower frequency (such as beta) oscillatory activity. In the seizure-free group, only 5 of 47 identified EZ contacts were outside the resection. In contrast, in the non-seizure-free group, 104 of 142 identified EZ contacts were outside the resection. We integrated the fingerprint prediction with the subject's MR images, thus providing individualized anatomical estimates of the EZ. We show that these fingerprint-based estimates in seizure-free patients are almost always inside the resection. On the other hand, for a large fraction of the nonseizure-free patients the estimated EZ was not well localized and was partially or completely outside the resection, which may explain surgical failure in such cases. We also show that when mapping fast activity alone onto MR images, the EZ was often over-estimated, indicating a reduced discriminative ability for fast activity relative to the full fingerprint for localization of the EZ.

2019

  • S. Choi, S. H. O’Neil, A. A. Joshi, J. Li, A. M. Bush, T. D. Coates, R. M. Leahy, J. C. Wood,
    “Anemia predicts lower white matter volume and cognitive performance in sickle and non-sickle cell anemia syndrome”,
    American Journal of Hematology, vol. 94, no. 10, pp. 1055–1065, 2019.      
    Abstract

    Severe chronic anemia is an independent predictor of overt stroke, white matter damage, and cognitive dysfunction in the elderly. Severe anemia also predisposes to white matter strokes in young children, independent of the anemia subtype. We previously demonstrated symmetrically decreased white matter (WM) volumes in patients with sickle cell disease (SCD). In the current study, we investigated whether patients with non-sickle anemia also have lower WM volumes and cognitive dysfunction. Magnetic Resonance Imaging was performed on 52 clinically asymptomatic SCD patients (age=21.4±7.7; F=27, M=25; hemoglobin=9.6±1.6 g/dL), 26 non-sickle anemic patients (age=23.9±7.9; F=14, M=12; hemoglobin=10.8±2.5 g/dL) and 40 control subjects (age=27.7±11.3; F=28, M=12; hemoglobin=13.4±1.3 g/dL). Voxel-wise changes in WM brain volumes were compared to hemoglobin levels to identify brain regions that are vulnerable to anemia. White matter volume was diffusely lower in deep, watershed areas proportionally to anemia severity. After controlling for age, sex, and hemoglobin level, brain volumes were independent of disease. WM volume loss was associated with lower Full Scale Intelligence Quotient (FSIQ; P=.0048; r2=.18) and an abnormal burden of silent cerebral infarctions (P=.029) in males, but not in females. Hemoglobin count and cognitive measures were similar between subjects with and without white-matter hyperintensities. The spatial distribution of volume loss suggests chronic hypoxic cerebrovascular injury, despite compensatory hyperemia. Neurocognitive consequences of WM volume changes and silent cerebral infarction were strongly sexually dimorphic. Understanding the possible neurological consequences of chronic anemia may help inform our current clinical practices.k

  • J. Li, S. E. Luczak, I. G. Rosen,
    “Comparing a distributed parameter model-based system identification technique with more conventional methods for inverse problems”,
    Journal of Inverse and Ill-posed Problems, vol. 27, no. 5, pp. 703–717, 2019.      
    Abstract

    Three methods for the estimation of blood or breath alcohol concentration (BAC/BrAC) from biosensor measured transdermal alcohol concentration (TAC) are evaluated and compared. Specifically, we consider a system identification/quasi-blind deconvolution scheme based on a distributed parameter model with unbounded input and output for ethanol transport in the skin and compare it to two more conventional system identification and filtering/deconvolution techniques for ill-posed inverse problems, one based on frequency domain methods and the other on a time series approach using an ARMA input/output model. Our basis for comparison are five statistical measures of interest to alcohol researchers and clinicians: peak BAC/BrAC, time of peak BAC/BrAC, the ascending and descending slopes of the BAC/BrAC curve, and the area underneath the BAC/BrAC curve.

  • J. Li, J. P. Haldar, J. C. Mosher, D. R. Nair, J. Gonzalez-Martinez, R. M. Leahy,
    “Scalable and robust tensor decomposition of spontaneous stereotactic EEG data”,
    IEEE Transactions on Biomedical Engineering, vol. 66, no. 6, pp. 1549–1558, 2019.        
    Abstract

    Objective: Identification of networks from resting brain signals is an important step in understanding the dynamics of spontaneous brain activity. We approach this problem using a tensor-based model. Methods: We develope a rank-recursive scalable and robust sequential canonical polyadic decomposition (SRSCPD) framework to decompose a tensor into several rank-1 components. Robustness and scalability are achieved using a warm start for each rank based on the results from the previous rank. Results: In simulations we show that SRSCPD consistently outperforms the multistart alternating least square (ALS) algorithm over a range of ranks and signal-to-noise ratios (SNRs), with lower computation cost. When applying SRSCPD to resting in-vivo stereotactic EEG (SEEG) data from two subjects with epilepsy, we found components corresponding to default mode and motor networks in both subjects. These components were also highly consistent within subject between two sessions recorded several hours apart. Similar components were not obtained using the conventional ALS algorithm. Conclusion: Consistent brain networks and their dynamic behaviors were identified from resting SEEG data using SRSCPD. Significance: SRSCPD is scalable to large datasets and therefore a promising tool for identification of brain networks in long recordings from single subjects.

  • E. Cvetkovska, W. A. Martins, J. Gonzalez-Martinez, K. Taylor, J. Li, O. Grinenko, J. C. Mosher, R. M. Leahy, P. Chauvel, D. Nair,
    “Heterotopia or overlaying cortex: What about in-between?”,
    Epilepsy and Behavior Case Reports, vol. 11, pp. 4–9, 2019.      
    Abstract

    We describe a patient with unilateral periventricular nodular heterotopia (PNH) and drug-resistant epilepsy, whose SEEG revealed that seizures were arising from the PNH, with the almost simultaneous involvement of heterotopic neurons ("micronodules") scattered within the white matter, and subsequently the overlying cortex. Laser ablation of heterotopic nodules and the adjacent white matter rendered the patient seizure free. This case elucidates that “micronodules” scattered in white matter between heterotopic nodules and overlying cortex might be another contributor in complex epileptogenicity of heterotopia. Detecting patient-specific targets in the epileptic network of heterotopia creates the possibility to disrupt the pathological circuit by minimally invasive procedures.

2018

  • A. A. Joshi, M. Chong, J. Li, S. Choi, R. M. Leahy,
    “Are you thinking what I’m thinking? Synchronization of resting fMRI time-series across subjects”,
    NeuroImage, vol. 172, pp. 740–752, 2018.        
    Abstract

    We describe BrainSync, an orthogonal transform that allows direct comparison of resting fMRI (rfMRI) time-series across subjects. For this purpose, we exploit the geometry of the rfMRI signal space to propose a novel orthogonal transformation that synchronizes rfMRI time-series across sessions and subjects. When synchronized, rfMRI signals become approximately equal at homologous locations across subjects. The method is based on the observation that rfMRI data exhibit similar connectivity patterns across subjects, as reflected in the pairwise correlations between different brain regions. We show that if the data for two subjects have similar correlation patterns then their time courses can be approximately synchronized by an orthogonal transformation. This transform is unique, invertible, efficient to compute, and preserves the connectivity structure of the original data for all subjects. Analogously to image registration, where we spatially align structural brain images, this temporal synchronization of brain signals across a population, or within-subject across sessions, facilitates cross-sectional and longitudinal studies of rfMRI data. The utility of the BrainSync transform is illustrated through demonstrative simulations and applications including quantification of rfMRI variability across subjects and sessions, cortical functional parcellation across a population, timing recovery in task fMRI data, comparison of task and resting state data, and an application to complex naturalistic stimuli for annotation prediction.

  • O. Grinenko†, J. Li†, J. C. Mosher, I. Z. Wang, J. C. Bulacio, J. Gonzalez-Martinez, D. Nair, I. Najm, R. M. Leahy, P. Chauvel,
    “A fingerprint of the epileptogenic zone in human epilepsies”,
    Brain, vol. 141, no. 1, pp. 117–131, 2018.        
    Abstract  

    Defining a bio-electrical marker for the brain area responsible for initiating a seizure remains an unsolved problem. Fast gamma activity has been identified as the most specific marker for seizure onset, but conflicting results have been reported. In this study, we describe an alternative marker, based on an objective description of interictal to ictal transition, with the aim of identifying a time-frequency pattern or 'fingerprint' that can differentiate the epileptogenic zone from areas of propagation. Seventeen patients who underwent stereoelectroencephalography were included in the study. Each had seizure onset characterized by sustained gamma activity and were seizure-free after tailored resection or laser ablation. We postulated that the epileptogenic zone was always located inside the resection region based on seizure freedom following surgery. To characterize the ictal frequency pattern, we applied the Morlet wavelet transform to data from each pair of adjacent intracerebral electrode contacts. Based on a visual assessment of the time-frequency plots, we hypothesized that a specific time-frequency pattern in the epileptogenic zone should include a combination of (i) sharp transients or spikes; preceding (ii) multiband fast activity concurrent; with (iii) suppression of lower frequencies. To test this hypothesis, we developed software that automatically extracted each of these features from the time-frequency data. We then used a support vector machine to classify each contact-pair as being within epileptogenic zone or not, based on these features. Our machine learning system identified this pattern in 15 of 17 patients. The total number of identified contacts across all patients was 64, with 58 localized inside the resected area. Subsequent quantitative analysis showed strong correlation between maximum frequency of fast activity and suppression inside the resection but not outside. We did not observe significant discrimination power using only the maximum frequency or the timing of fast activity to differentiate contacts either between resected and non-resected regions or between contacts identified as epileptogenic versus non-epileptogenic. Instead of identifying a single frequency or a single timing trait, we observed the more complex pattern described above that distinguishes the epileptogenic zone. This pattern encompasses interictal to ictal transition and may extend until seizure end. Its time-frequency characteristics can be explained in light of recent models emphasizing the role of fast inhibitory interneurons acting on pyramidal cells as a prominent mechanism in seizure triggering. The pattern clearly differentiates the epileptogenic zone from areas of propagation and, as such, represents an epileptogenic zone 'fingerprint'.

Conference Proceedings

2023

  • J. Li, G. Tuckute, E. Fedorenko, B. L. Edlow, B. Fischl‡, A. V. Dalca‡,
    “Joint cortical registration of geometry and function via semi-supervised learning”,
    Medical Imaging with Deep Learning, Nashville, TN, Jul. 2023, pp. 862–876.      
    Abstract

    Brain surface-based image registration, an important component of brain image analysis, establishes spatial correspondence between cortical surfaces. Existing iterative and learning-based approaches focus on accurate registration of folding patterns of the cerebral cortex, and assume that geometry predicts function and thus functional areas will also be well aligned. However, structure/functional variability of anatomically corresponding areas across subjects has been widely reported. In this work, we introduce a learning-based cortical registration framework, JOSA, which jointly aligns folding patterns and functional maps while simultaneously learning an optimal atlas. We demonstrate that JOSA can substantially improve registration performance in both anatomical and functional domains over existing methods. By employing a semi-supervised training strategy, the proposed framework obviates the need for functional data during inference, enabling its use in broad neuroscientific domains where functional data may not be observed.

2021

  • A. A. Joshi, S. Choi, J. Li, H. Akrami, R. M. Leahy,
    “A pairwise approach for fMRI group studies using BrainSync transform”,
    Proc. SPIE Medical Imaging 2021: Image Processing, San Diego, CA, Mar. 2021.      

2020

  • J. Li, A. A. Joshi, R. M. Leahy,
    “A network-based approach to study of ADHD using tensor decomposition of resting fMRI data”,
    IEEE 17th International Symposium on Biomedical Imaging, Iowa City, IA, Apr. 2020, pp. 1–5.      
    Abstract  

    Identifying changes in functional connectivity in Attention Deficit Hyperactivity Disorder (ADHD) using functional magnetic resonance imaging (fMRI) can help us understand the neural substrates of this brain disorder. Many studies of ADHD using resting state fMRI (rs-fMRI) data have been conducted in the past decade with either manually crafted features that do not yield satisfactory performance, or automatically learned features that often lack interpretability. In this work, we present a tensor-based approach to identify brain networks and extract features from rs-fMRI data. Results show the identified networks are interpretable and consistent with our current understanding of ADHD conditions. The extracted features are not only predictive of ADHD score but also discriminative for classification of ADHD subjects from typically developed children.

  • H. Akrami, A. A. Joshi, J. Li, S. Aydöre, R. M. Leahy,
    “Brain lesion detection using a robust variational autoencoder and transfer learning”,
    IEEE 17th International Symposium on Biomedical Imaging, Iowa City, IA, Apr. 2020, pp. 786–790.      
    Abstract

    Automated brain lesion detection from multi-spectral MR images can assist clinicians by improving sensitivity as well as specificity. Supervised machine learning methods have been successful in lesion detection. However, these methods usually rely on a large number of manually delineated images for specific imaging protocols and parameters and often do not generalize well to other imaging parameters and demographics. Most recently, unsupervised models such as autoencoders have become attractive for lesion detection since they do not need access to manually delineated lesions. Despite the success of unsupervised models, using pre-trained models on an unseen dataset is still a challenge. This difficulty is because the new dataset may use different imaging parameters, demographics, and different pre-processing techniques. Additionally, using a clinical dataset that has anomalies and outliers can make unsupervised learning challenging since the outliers can unduly affect the performance of the learned models. These two difficulties make unsupervised lesion detection a particularly challenging task. The method proposed in this work addresses these issues using a two-prong strategy: (1) we use a robust variational autoencoder model that is based on robust statistics, specifically the beta-divergence that can be trained with data that has outliers; (2) we use a transfer learning method for learning models across datasets with different characteristics. Our results on MRI datasets demonstrate that we can improve the accuracy of lesion detection by adapting robust statistical models and transfer learning for a variational autoencoder model.

2019

  • A. A. Joshi, H. Akrami, J. Li, R. M. Leahy,
    “A matched filter decomposition of fMRI into resting and task components”,
    22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, Shenzhen, China, Sept. 2019, pp. 673–681.      
    Abstract

    The human brain exhibits dynamic interactions among brain regions when responding to stimuli and executing tasks, which can be recorded using functional magnetic resonance imaging (fMRI). Functional MRI signals collected in response to specific tasks consist of a combination of task-related and spontaneous (task-independent) activity. By exploiting the highly structured spatiotemporal patterns of resting state networks, this paper presents a matched-filter approach to decomposing fMRI signals into task and resting-state components. To perform the decomposition, we first use a temporal alignment procedure that is a windowed version of the brainsync transform to synchronize a resting template to the brain's response to tasks. The resulting 'matched filter' removes the components of the fMRI signal that can be described by resting connectivity, leaving the portion of brain activity directly related to tasks. We present a closed-form expression for the windowed synchronization transform that is used by the matched filter. We demonstrate performance of this procedure in application to motor task and language task fMRI data. We show qualitatively and quantitatively that by removing the resting activity, we are able to identify task activated regions in the brain more clearly. Additionally, we show improved prediction accuracy in multivariate pattern analysis when using the matched filtered fMRI data.

  • J. Li, J. L. Wisnowski, A. A. Joshi, R. M. Leahy,
    “Brain network identification in asynchronous task fMRI data using robust and scalable tensor decomposition”,
    Proc. SPIE Medical Imaging 2019: Image Processing, San Diego, CA, Mar. 2019, pp. 164–172.        
    Abstract

    The goal of this work is to robustly identify common brain networks and their corresponding temporal dynamics across subjects in asynchronous task functional MRI (tfMRI) signals. We approached this problem using a robust and scalable tensor decomposition method combined with the BrainSync algorithm. We first used BrainSync algorithm to temporally align asynchronous tfMRI data, allowing us to study common brain networks across subjects. We mapped the synchronized tfMRI data into a 3D tensor (vertices × time × session) and performed a greedy canonical polyadic (CP) decomposition, reducing the rank to 20 in order to improve the signal-to-noise ratio (SNR). We incorporated the Nesterovaccelerated adaptive moment estimation into our previously developed scalable and robust sequential CP decomposition (SRSCPD) framework and applied this improved version of SRSCPD to the rank-reduced tensor to identify dynamic brain networks. We successfully identified 9 brain networks with their corresponding temporal dynamics from 40 subjects using Human Connectome Project tfMRI data without using any prior information with regard to the task designs. Three of these show the subjects’ responses to cues at the beginning of each task block (fronto-parietal attentional control network, visual network and executive control network); one corresponds to the default mode network that exhibits deactivation during the tasks; four show motors networks (left hand, right hand, tongue, and both feet) where the temporal dynamics are strongly correlated to the task designs, and the remaining component reflects physiological noise (respiration).

  • H. Akrami, A. A. Joshi, J. Li, R. M. Leahy,
    “Group-wise alignment of resting fMRI in space and time”,
    Proc. SPIE Medical Imaging 2019: Image Processing, San Diego, CA, Mar. 2019, pp. 737–744.      
    Abstract

    Spontaneous brain activity is an important biomarker for various neurological and psychological conditions and can be measured using resting functional Magnetic Resonance Imaging (rfMRI). Since brain activity during resting is spontaneous, it is not possible to directly compare rfMRI time-courses across subjects. Moreover, the spatial configuration of functionally specialized brain regions can vary across subjects throughout the cortex limiting our ability to make precise spatial comparisons. We describe a new approach to jointly align and synchronize fMRI data in space and time, across a group of subjects. We build on previously described methods for inter-subject spatial “Hyper-Alignment” and temporal synchronization through the “BrainSync” transform. We first describe BrainSync Alignment (BSA), a group-based extension of the pair-wise BrainSync transform, that jointly synchronizes resting or task fMRI data across time for multiple subjects. We then explore the combination of BSA with Response Hyper-Alignment (RHA) and compare with Connectivity Hyper-Alignment (CHA), an alternative approach to spatial alignment based on resting fMRI. The result of applying RHA and BSA is both to produce improved functional spatial correspondence across a group of subjects, and to align their time-series so that, even for spontaneous resting data, we see highly correlated temporal dynamics at homologous locations across the group. These spatiotemporally aligned data can then be used as an atlas in future applications. We validate these transfer functions by applying them to z-score maps of an independent dataset and calculating inter-subject correlation. The results show that RHA can be calculated from rfMRI and have comparable output with CHA by leveraging BSA. Moreover, through calculation and application to task fMRI-based spatial transformations on an independent dataset, we show that the combination of RHA and BSA produces improved spatial functional alignment significantly relative to either RHA or CHA alone.

  • A. A. Joshi, J. Li, H. Akrami, R. M. Leahy,
    “Predicting cognitive scores from resting fMRI data and geometric features of the brain”,
    Proc. SPIE Medical Imaging 2019: Image Processing, San Diego, CA, Mar. 2019, pp. 619–625.      
    Abstract

    Anatomical T1 weighted Magnetic Resonance Imaging (MRI) and functional magnetic resonance imaging collected during resting (rfMRI) are promising markers that offer insight into structure and function of the human brain. The objective of this work is to explore the use of a deep learning neural network to predict cognitive performance scores and ADHD indices in a group of ADHD and control subjects. First, we processed the rfMRI and MRI data of subjects using the BrainSuite fMRI Processing (BFP) pipeline to perform anatomical and functional preprocessing. This produces for each subject fMRI and geometric (anatomical) features represented in a standardized grayordinate system. The geometric and functional cortical data corresponding to the two hemispheres were then transformed to 128x128 multichannel images and input to a convolutional component of the neural network. Subcortical data were presented in a standard vector form and input to a standard input layer of the network. The neural network was implemented in Python using the Keras library with a TensorFlow backend. Training was performed on 168 images with 90 images used for testing. We observed significant correlation between predicted and actual values of the indices tested: Performance IQ: 0.47; Verbal IQ: 0.41, ADHD: 0.57. Comparing these values to those from network trained on functional-only and structural-only data, we saw that rfMRI is more informative than MRI, but the two modalities are highly complementary in terms of predicting these indices.

2018

  • A. A. Joshi, J. Li, M. Chong, H. Akrami, R. M. Leahy,
    “rfDemons: resting fMRI-based cortical surface registration using the BrainSync transform”,
    21st International Conference on Medical Image Computing and Computer-Assisted Intervention, Granada, Spain, Sept. 2018, pp. 198–205.    
    Abstract

    Cross subject functional studies of cerebral cortex require cortical registration that aligns functional brain regions. While cortical folding patterns are approximate indicators of the underlying cytoarchitecture, coregistration based on these features alone does not accurately align functional regions in cerebral cortex. This paper presents a method for cortical surface registration (rfDemons) based on resting fMRI (rfMRI) data that uses curvature-based anatomical registration as an initialization. In contrast to existing techniques that use connectivity-based features derived from rfMRI, the proposed method uses 'synchronized' resting rfMRI time series directly. The synchronization of rfMRI data is performed using the BrainSync transform which applies an orthogonal transform to the rfMRI time series to temporally align them across subjects. The rfDemons method was applied to rfMRI from the Human Connectome Project and evaluated using task fMRI data to explore the impact of cortical registration performed using resting fMRI data on functional alignment of the cerebral cortex.

  • J. Li, S. Choi, A. A. Joshi, J. L. Wisnowski, R. M. Leahy,
    “Global PDF-based temporal non-local means filtering reveals individual differences in brain connectivity”,
    IEEE 15th International Symposium on Biomedical Imaging, Washington D.C., Apr. 2018, pp. 15–19.        
    Abstract

    Characterizing functional brain connectivity using resting fMRI is challenging due to the relatively small BOLD signal contrast and low SNR. Gaussian filtering tends to undermine the individual differences detected by analysis of BOLD signal by smoothing signals across boundaries of different functional areas. Temporal non-local means (tNLM) filtering denoises fMRI data while preserving spatial structures but the kernel and parameters for tNLM filter need to be chosen carefully in order to achieve optimal results. Global PDF-based tNLM filtering (GPDF) is a new, data-dependent optimized kernel function for tNLM filtering which enables us to perform global filtering with improved noise reduction effects without blurring adjacent functional regions.

2017

  • J. Li, J. C. Mosher, D. R. Nair, J. Gonzalez-Martinez, R. M. Leahy,
    “Robust tensor decomposition of resting brain networks in stereotactic EEG”,
    IEEE 51st Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, Oct. 2017, pp. 1544–1548.        
    Abstract

    Stereotactically implanted Electro-Encephalography (SEEG) in patients with epilepsy provides a unique insight into spontaneous human brain activity. Exploring dynamic functional connectivity in spontaneous SEEG signals provides a rich framework for studying brain networks. Tensor decomposition is a powerful tool for decoding dynamic networks, capturing the intrinsic interactions between multiple dimensions with less restrictive constraints than traditional 2D matrix decomposition methods such as PCA and ICA. Tensor decomposition, however, is seldom used for decoding large resting brain datasets due to its high computational complexity and poor robustness. In this paper, we describe a Scalable and Robust Sequential Canonical Polyadic Decomposition (SRSCPD) framework that can sequentially and robustly identify tensor models of successively higher rank. We demonstrate that SRSCPD is not only more robust than the popular Alternating Least Square (ALS) algorithm, but can also be extended to large-scale problems.

  • J. Li, R. M. Leahy,
    “Parameter selection for optimized non-local means filtering of task fMRI”,
    IEEE 14th International Symposium on Biomedical Imaging, Melbourne, Australia, Apr. 2017, pp. 476–480.      
    Abstract

    Non-local means (NLM) filtering of fMRI can reduce noise while preserving spatial structure. We have developed a variant called temporal-NLM (tNLM) which uses similarity in time-series between voxels as the basis for computing the weights in the filter. Using tNLM, dynamic fMRI data can be denoised while spatial boundaries between functionally distinct areas in the brain tend to be preserved. The degree of smoothing in tNLM is determined by a parameter h. Here we describe a procedure for selection of h to optimize our ability to differentiate functionally discrete brain regions. We demonstrate the method in application to optimized filtering of task fMRI data.

2012

  • A. Kuruvilla, J. Li, P. H. Yeomans, P. Quelhas, N. Shaikh, A. Hoberman, J. Kovačević,
    “Otitis media vocabulary and grammar”,
    IEEE 19th International Conference on Image Processing, Orlando, FL, Oct. 2012, pp. 2845–2848.      
    Abstract

    We propose an automated algorithm for classifying diagnostic categories of otitis media (middle ear inflammation); acute otitis media, otitis media with effusion and no effusion. Acute otitis media represents a bacterial superinfection of the middle ear fluid and otitis media with effusion a sterile effusion that tends to subside spontaneously. Diagnosing children with acute otitis media is hard, leading to overprescription of antibiotics that are beneficial only for children with acute otitis media, prompting a need for an accurate and automated algorithm. To that end, we design a feature set understood by both otoscopists and engineers based on the actual visual cues used by otoscopists; we term this otitis media vocabulary. We also design a process to combine the vocabulary terms based on the decision process used by otoscopists; we term this otitis media grammar. The algorithm achieves 84% classification accuracy, in the range or outperforming clinicians who did not receive special training, as well as state-of-the-art classifiers.

Thesis/Dissertation

2019: Doctor of Philosophy in Electrical Engineering

  • J. Li,
    “Functional connectivity analysis and network identification in the human brain”,
    University of Southern California, Los Angeles, CA, May 2019.    
    Abstract

    Brain connectivity is modeled as a complex, segregative and integrative network of connections between different brain regions. Studying functional brain connectivity can offer us an effective way to examine how different brain networks relate to human behaviors as well as how those networks may be altered in neurological diseases. However, measuring functional connectivity poses a variety of mathematical, signal processing and neuroscience challenges. First, a good high-level representation of the data is often required in order to obtain an accurate estimation of the functional connectivity, because most of the typically-used linear measures are not capable of capturing the true highly non-linear brain interactions. Second, the temporal stationarity of the time series assumed by most of the studies may not be realistic due to the dynamic nature of the brain. Hence, how to reliably estimate the spatial and temporal dynamics of functional connectivity simultaneously is a key challenge to us. Moreover, signals collected via almost all neuroimaging techniques are heavily corrupted with noise. The inherent low signal-to-noise ratio prevents us from obtaining a robust estimation of functional connectivity. In this work, we present and validate several novel approaches and methods to address some of the challenges in functional connectivity estimation and brain network identification problems. To address the high-level data representation issue, we defined a bio-electrical marker that can differentiate the epileptogenic zone from areas of propagation in patients with epilepsy. We discovered a specific ictal time-frequency pattern, referred as the “fingerprint”, in the epileptogenic zone which contains a combination of sharp spikes preceding multi-band fast activity concurrent with suppression of lower frequencies. We developed a novel machine learning system that automatically extracts each of these features, classifies electrode contacts as being within the epileptogenic zone or outside the epileptogenic zone and generates individualized epileptogenic zone predictions for each patient based on their anatomical magnetic resonance images. To address the dynamic brain network identification issue, we developed a rank-recursive scalable and robust sequential canonical polyadic decomposition framework that allows us to robustly discover brain networks which can overlap in both space and time in large-scale datasets. The robustness and scalability were achieved by using lower-rank solutions as the warm start to higher-rank decompositions. This scalable and robust sequential canonical polyadic decomposition framework is flexible in the sense that it is not only applicable to wavelet-transformed electroencephalography data but also to multi-subject asynchronous functional magnetic resonance imaging data if the data is temporally aligned across subjects using the BrainSync algorithm. To address the noise corruption issue, we described an optimization-based method that provides a means of systematically selecting the parameter for the temporal non-local means filtering. We further developed global PDF-based temporal non-local means, a novel data-driven optimized kernel function based on Bayes factor for the temporal non-local means filtering, which allows us to perform global filtering with improved noise reduction effects but without blurring adjacent functional regions. Applications of these proposed methods are illustrated using a variety of simulated as well as in-vivo clinical data.

2017: Master of Science in Statistics

  • J. Li,
    “Distributed parameter model based system identification and input determination in the estimation of blood alcohol concentration from transdermal alcohol biosensor data”,
    University of Southern California, Los Angeles, CA, Jul. 2017.    
    Abstract

    Methods for the estimation of blood or breath alcohol concentration (BAC/BrAC) from biosensor measured transdermal alcohol concentration (TAC) are developed, evaluated and compared. Specifically, a scheme based on a distributed parameter model with unbounded input and output for ethanol transport in the skin is compared to more conventional filtering/deconvolution techniques, one based on frequency domain methods, and the other on a time series approach using an autoregressive moving average (ARMA) input/output model. Our basis for comparison are five statistics of interest to alcohol researchers and clinicians: peak BAC/BrAC, time of peak BAC/BrAC, the ascending and descending slopes of the BAC/BrAC curve and area underneath the BAC/BrAC curve. It can be shown that the ARMA-based method yields the best estimation of the peak while the distributed parameter model produces the best estimation of the time of the peak. The Fourier-based method has the least variance out of the three and is computationally very efficient.

2011: Bachelor of Engineering in Electronic and Information Engineering

  • J. Li,
    “Spatiotemporal correlation-based near-duplicate video detection”,
    Beijing University of Technology, Beijing, China, Jun. 2011.  
    Abstract

    视频拷贝检测是目前多媒体处理领域的前沿研究热点，在海量视频信息检索和版权保护等方面有着重要的应用价值。视频拷贝检测的实质在于判断不同的视频片段是否具有相同的内容，从而实现对特定视频内容的搜索、检测和跟踪。在视频拷贝检测技术领域，当前国内外的研究重点是寻找各种复杂的特征提取方法来提高拷贝检测的准确性。然而在实际应用中，最需要解决的问题是在大规模数据下，在保持检测的准确性、鲁棒性的同时，如何显著提高拷贝检测的速度。本论文提出了一种基于时空相关性的视频拷贝检测技术。该技术利用视频时间和空间的相关性，直接在压缩域进行视频拷贝检测，在不解压或者少解压的情况下，在保证检测准确性的同时，大大提高检测的速度。该方法的实现过程如下: 首先，从 MPEG-2 压缩码流中提取亮度、颜色、纹理、运动 以及显著图等信息，然后利用这些信息对待检测的两段视频进行粗略的镜头分割，使之成为视频段落，然后对压缩域中提取的各种特征信息并进行统计，依据某一准则进行比对，最后综合各种特征的对比结果，给出两段视频相似程度。实验结果表明，本文提出的基于时空相关性的视频拷贝检测算法能够在保证检测准确率的同时，有效地降低处理复杂度，提升检测效率，并对不同分辨率、不同质量、不同内容的视频均具有较强的鲁棒性。

Others

All other publications including non-peer-reviewed conference abstracts and technical reports can be found in this archive page