Clinical Focus

  • Neurology
  • Neurology with Special Qualifications in Child Neurology

Academic Appointments

Honors & Awards

  • Chief Resident, Neurology (2007-2008)
  • K12 NSADA Award, NIH/NINDS (2008-2011)
  • R. S. Fisher award for Teaching, Stanford Department of Neurology (2008)

Professional Education

  • Board Certification: American Osteopathic Board of Neurology and Psychiatry, Neurology with Special Qualifications in Child Neurology (2009)
  • Medical Education: Washington University School Of Medicine Registrar (2004) MO
  • Fellowship: Stanford University Radiology Residency (2011) CA
  • Board Certification: American Board of Psychiatry and Neurology, Epilepsy (2016)
  • Internship: University of California at San Francisco School of Medicine (2005) CA
  • Residency: Stanford University Medical Center (2009) CA
  • Epilepsy Fellowship, Stanford University Medical Center, Pediatric Epilepsy (2011)

Current Research and Scholarly Interests

My chief clinical focus is in pediatric epilepsy, especially how epilepsy affects learning and development. For my research, I background in neural development and computational neuroscience towards developing better learning algorithms and applying the latest techniques in machine learning for better diagnosis and treatment of disease.

2020-21 Courses

All Publications

  • Towards trustworthy seizure onset detection using workflow notes. NPJ digital medicine Saab, K., Tang, S., Taha, M., Lee-Messer, C., Re, C., Rubin, D. L. 2024; 7 (1): 42


    A major barrier to deploying healthcare AI is trustworthiness. One form of trustworthiness is a model's robustness across subgroups: while models may exhibit expert-level performance on aggregate metrics, they often rely on non-causal features, leading to errors in hidden subgroups. To take a step closer towards trustworthy seizure onset detection from EEG, we propose to leverage annotations that are produced by healthcare personnel in routine clinical workflows-which we refer to as workflow notes-that include multiple event descriptions beyond seizures. Using workflow notes, we first show that by scaling training data to 68,920 EEG hours, seizure onset detection performance significantly improves by 12.3 AUROC (Area Under the Receiver Operating Characteristic) points compared to relying on smaller training sets with gold-standard labels. Second, we reveal that our binary seizure onset detection model underperforms on clinically relevant subgroups (e.g., up to a margin of 6.5 AUROC points between pediatrics and adults), while having significantly higher FPRs (False Positive Rates) on EEG clips showing non-epileptiform abnormalities (+19 FPR points). To improve model robustness to hidden subgroups, we train a multilabel model that classifies 26 attributes other than seizures (e.g., spikes and movement artifacts) and significantly improve overall performance (+5.9 AUROC points) while greatly improving performance among subgroups (up to +8.3 AUROC points) and decreasing false positives on non-epileptiform abnormalities (by 8 FPR points). Finally, we find that our multilabel model improves clinical utility (false positives per 24 EEG hours) by a factor of 2*.

    View details for DOI 10.1038/s41746-024-01008-9

    View details for PubMedID 38383884

  • ATCON: Attention Consistency for Vision Models Mirzazadeh, A., Dubost, F., Pike, M., Maniar, K., Zuo, M., Lee-Messer, C., Rubin, D., IEEE IEEE COMPUTER SOC. 2023: 1880-1889
  • Editorial: Enlarged perivascular spaces: etiology and significance. Frontiers in neuroscience Cavallari, M., Dubost, F., Guttmann, C. R., Lee-Messer, C. W. 2023; 17: 1321691

    View details for DOI 10.3389/fnins.2023.1321691

    View details for PubMedID 38161800

  • Semi-Supervised Learning for Sparsely-Labeled Sequential Data: Application to Healthcare Video Processing Dubost, F., Hong, E., Tang, S., Bhaskhar, N., Lee-Messer, C., Rubin, D., IEEE IEEE COMPUTER SOC. 2023: 1890-1899
  • Connectivity increases during spikes and spike-free periods in self-limited epilepsy with centrotemporal spikes. Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology Goad, B. S., Lee-Messer, C., He, Z., Porter, B. E., Baumer, F. M. 2022


    OBJECTIVE: To understand the impact of interictal spikes on brain connectivity in patients with Self-Limited Epilepsy with Centrotemporal Spikes (SeLECTS).METHODS: Electroencephalograms from 56 consecutive SeLECTS patients were segmented into periods with and without spikes. Connectivity between electrodes was calculated using the weighted phase lag index. To determine if there are chronic alterations in connectivity in SeLECTS, we compared spike-free connectivity to connectivity in 65 matched controls. To understand the acute impact of spikes, we compared connectivity immediately before, during, and after spikes versus baseline, spike-free connectivity. We explored whether behavioral state, spike laterality, or antiseizure medications affected connectivity.RESULTS: Children with SeLECTS had markedly higher connectivity than controls during sleep but not wakefulness, with greatest difference in the right hemisphere. During spikes, connectivity increased globally; before and after spikes, left frontal and bicentral connectivity increased. Right hemisphere connectivity increased more during right-sided than left-sided spikes; left hemisphere connectivity was equally affected by right and left spikes.CONCLUSIONS: SeLECTS patient have persistent increased connectivity during sleep; connectivity is further elevated during the spike and perispike periods.SIGNIFICANCE: Testing whether increased connectivity impacts cognition or seizure susceptibility in SeLECTS and more severe epilepsies could help determine if spikes should be treated.

    View details for DOI 10.1016/j.clinph.2022.09.015

    View details for PubMedID 36307364

  • Editorial: Machine Learning in Neuroimaging. Frontiers in neurology Federau, C., Scalzo, F., Lee-Messer, C. W., Zaharchuk, G. 1800; 12: 778765

    View details for DOI 10.3389/fneur.2021.778765

    View details for PubMedID 34975734

  • Cross-Modal Data Programming Enables Rapid Medical Machine Learning. Patterns (New York, N.Y.) Dunnmon, J. A., Ratner, A. J., Saab, K., Khandwala, N., Markert, M., Sagreiya, H., Goldman, R., Lee-Messer, C., Lungren, M. P., Rubin, D. L., Re, C. 2020; 1 (2)


    A major bottleneck in developing clinically impactful machine learning models is a lack of labeled training data for model supervision. Thus, medical researchers increasingly turn to weaker, noisier sources of supervision, such as leveraging extractions from unstructured text reports to supervise image classification. A key challenge in weak supervision is combining sources of information that may differ in quality and have correlated errors. Recently, a statistical theory of weak supervision called data programming has shown promise in addressing this challenge. Data programming now underpins many deployed machine-learning systems in the technology industry, even for critical applications. We propose a new technique for applying data programming to the problem of cross-modal weak supervision in medicine, wherein weak labels derived from an auxiliary modality (e.g., text) are used to train models over a different target modality (e.g., images). We evaluate our approach on diverse clinical tasks via direct comparison to institution-scale, hand-labeled datasets. We find that our supervision technique increases model performance by up to 6 points area under the receiver operating characteristic curve (ROC-AUC) over baseline methods by improving both coverage and quality of the weak labels. Our approach yields models that on average perform within 1.75 points ROC-AUC of those supervised with physician-years of hand labeling and outperform those supervised with physician-months of hand labeling by 10.25 points ROC-AUC, while using only person-days of developer time and clinician work-a time saving of 96%. Our results suggest that modern weak supervision techniques such as data programming may enable more rapid development and deployment of clinically useful machine-learning models.

    View details for DOI 10.1016/j.patter.2020.100019

    View details for PubMedID 32776018

  • Weak supervision as an efficient approach for automated seizure detection in electroencephalography. Digital Medicine Saab, K., Dunnmon, J., Ré, C., Rubin, D., Lee-Messer, C. 2020; 3: 12
  • iEEG-BIDS, extending the Brain Imaging Data Structure specification to human intracranial electrophysiology. Scientific data Holdgraf, C., Appelhoff, S., Bickel, S., Bouchard, K., D'Ambrosio, S., David, O., Devinsky, O., Dichter, B., Flinker, A., Foster, B. L., Gorgolewski, K. J., Groen, I., Groppe, D., Gunduz, A., Hamilton, L., Honey, C. J., Jas, M., Knight, R., Lachaux, J., Lau, J. C., Lee-Messer, C., Lundstrom, B. N., Miller, K. J., Ojemann, J. G., Oostenveld, R., Petridou, N., Piantoni, G., Pigorini, A., Pouratian, N., Ramsey, N. F., Stolk, A., Swann, N. C., Tadel, F., Voytek, B., Wandell, B. A., Winawer, J., Whitaker, K., Zehl, L., Hermes, D. 2019; 6 (1): 102

    View details for DOI 10.1038/s41597-019-0105-7

    View details for PubMedID 31239438

  • Optimal recording duration of ambulatory EEG (aEEG). Epilepsy research Kuo, J., Lee-Messer, C., Le, S. 2018; 149: 9–12

    View details for DOI 10.1016/j.eplepsyres.2018.07.025

    View details for PubMedID 30399521

  • Clinical Transcriptome Sequencing Confirms Activation of a Cryptic Splice Site in Suspected <it><bold>SYNGAP1</it></bold>-Related Disorder MOLECULAR SYNDROMOLOGY Brimble, E., Lee-Messer, C., Nagy, P. L., Propst, J., Ruzhnikov, M. Z. 2018; 9 (6): 295–99

    View details for DOI 10.1159/000492706

    View details for Web of Science ID 000456045700004

  • Optogenetic Stimulation of Neural Grafts Enhances Neurotransmission and Downregulates the Inflammatory Response in Experimental Stroke Model. Cell transplantation Daadi, M. M., Klausner, J. Q., Bajar, B., Goshen, I., Lee-Messer, C., Lee, S. Y., Winge, M. C., Ramakrishnan, C., Lo, M., Sun, G., Deisseroth, K., Steinberg, G. K. 2016; 25 (7): 1371-1380


    Compelling evidence suggests that transplantation of neural stem cells (NSCs) from multiple sources ameliorates motor deficits after stroke. However, it is currently unknown to what extent the electrophysiological activity of grafted NSC progeny participates in the improvement of motor deficits and whether excitatory phenotypes of the grafted cells are beneficial or deleterious to sensorimotor performances. To address this question, we used optogenetic tools to drive the excitatory outputs of the grafted NSCs and assess the impact on local circuitry and sensorimotor performance. We genetically engineered NSCs to express the Channelrhodopsin-2 (ChR2), a light-gated cation channel that evokes neuronal depolarization and initiation of action potentials with precise temporal control to light stimulation. To test the function of these cells in a stroke model, rats were subjected to an ischemic stroke and grafted with ChR2-NSCs. The grafted NSCs identified with a human-specific nuclear marker survived in the peri-infarct tissue and coexpressed the ChR2 transgene with the neuronal markers TuJ1 and NeuN. Gene expression analysis in stimulated versus vehicle-treated animals showed a differential upregulation of transcripts involved in neurotransmission, neuronal differentiation, regeneration, axonal guidance, and synaptic plasticity. Interestingly, genes involved in the inflammatory response were significantly downregulated. Behavioral analysis demonstrated that chronic optogenetic stimulation of the ChR2-NSCs enhanced forelimb use on the stroke-affected side and motor activity in an open field test. Together these data suggest that excitatory stimulation of grafted NSCs elicits beneficial effects in experimental stroke model through cell replacement and non-cell replacement, anti-inflammatory/neurotrophic effects.

    View details for DOI 10.3727/096368915X688533

    View details for PubMedID 26132738

  • Prolonged neuropsychiatric effects following management of chloroquine intoxication with psychotropic polypharmacy. Clinical case reports Maxwell, N. M., Nevin, R. L., Stahl, S., Block, J., Shugarts, S., Wu, A. H., Dominy, S., Solano-Blanco, M. A., Kappelman-Culver, S., Lee-Messer, C., Maldonado, J., Maxwell, A. J. 2015; 3 (6): 379-387


    Susceptibility to quinoline antimalarial intoxication may reflect individual genetic and drug-induced variation in neuropharmacokinetics. In this report, we describe a case of chloroquine intoxication that appeared to be prolonged by subsequent use of multiple psychotropic medications. This case highlights important new considerations for the management of quinoline antimalarial intoxication.

    View details for DOI 10.1002/ccr3.238

    View details for PubMedID 26185633

    View details for PubMedCentralID PMC4498847

  • Prolonged neuropsychiatric effects following management of chloroquine intoxication with psychotropic polypharmacy. Clinical case reports Maxwell, N. M., Nevin, R. L., Stahl, S., Block, J., Shugarts, S., Wu, A. H., Dominy, S., Solano-Blanco, M. A., Kappelman-Culver, S., Lee-Messer, C., Maldonado, J., Maxwell, A. J. 2015; 3 (6): 379-387


    Susceptibility to quinoline antimalarial intoxication may reflect individual genetic and drug-induced variation in neuropharmacokinetics. In this report, we describe a case of chloroquine intoxication that appeared to be prolonged by subsequent use of multiple psychotropic medications. This case highlights important new considerations for the management of quinoline antimalarial intoxication.

    View details for DOI 10.1002/ccr3.238

    View details for PubMedID 26185633

  • MicroRNA-mediated conversion of human fibroblasts to neurons NATURE Yoo, A. S., Sun, A. X., Li, L., Shcheglovitov, A., Portmann, T., Li, Y., Lee-Messer, C., Dolmetsch, R. E., Tsien, R. W., Crabtree, G. R. 2011; 476 (7359): 228-U123


    Neurogenic transcription factors and evolutionarily conserved signalling pathways have been found to be instrumental in the formation of neurons. However, the instructive role of microRNAs (miRNAs) in neurogenesis remains unexplored. We recently discovered that miR-9* and miR-124 instruct compositional changes of SWI/SNF-like BAF chromatin-remodelling complexes, a process important for neuronal differentiation and function. Nearing mitotic exit of neural progenitors, miR-9* and miR-124 repress the BAF53a subunit of the neural-progenitor (np)BAF chromatin-remodelling complex. After mitotic exit, BAF53a is replaced by BAF53b, and BAF45a by BAF45b and BAF45c, which are then incorporated into neuron-specific (n)BAF complexes essential for post-mitotic functions. Because miR-9/9* and miR-124 also control multiple genes regulating neuronal differentiation and function, we proposed that these miRNAs might contribute to neuronal fates. Here we show that expression of miR-9/9* and miR-124 (miR-9/9*-124) in human fibroblasts induces their conversion into neurons, a process facilitated by NEUROD2. Further addition of neurogenic transcription factors ASCL1 and MYT1L enhances the rate of conversion and the maturation of the converted neurons, whereas expression of these transcription factors alone without miR-9/9*-124 was ineffective. These studies indicate that the genetic circuitry involving miR-9/9*-124 can have an instructive role in neural fate determination.

    View details for DOI 10.1038/nature10323

    View details for Web of Science ID 000293731900041

    View details for PubMedID 21753754

    View details for PubMedCentralID PMC3348862

  • Clinical and Molecular Heterogeneity in Patients with the CblD Inborn Error of Cobalamin Metabolism JOURNAL OF PEDIATRICS Miousse, I. R., Watkins, D., Coelho, D., Rupar, T., Crombez, E. A., Vilain, E., Bernstein, J. A., Cowan, T., Lee-Messer, C., Enns, G. M., Fowler, B., Rosenblatt, D. S. 2009; 154 (4): 551-556


    To describe 3 patients with the cblD disorder, a rare inborn error of cobalamin metabolism caused by mutations in the MMADHC gene that can result in isolated homocystinuria, isolated methylmalonic aciduria, or combined homocystinuria and methylmalonic aciduria.Patient clinical records were reviewed. Biochemical and somatic cell genetic studies were performed on cultured fibroblasts. Sequence analysis of the MMADHC gene was performed on patient DNA.Patient 1 presented with isolated methylmalonic aciduria, patient 3 with isolated homocystinuria, and patient 2 with combined methylmalonic aciduria and homocystinuria. Studies of cultured fibroblasts confirmed decreased synthesis of adenosylcobalamin in patient 1, decreased synthesis of methylcobalamin in patient 3, and decreased synthesis of both cobalamin derivatives in patient 2. The diagnosis of cblD was established in each patient by complementation analysis. Mutations in the MMADHC gene were identified in all patients.The results emphasize the heterogeneous clinical, cellular and molecular phenotype of the cblD disorder. The results of molecular analysis of the MMADHC gene are consistent with the hypothesis that mutations affecting the N terminus of the MMADHC protein are associated with methylmalonic aciduria, and mutations affecting the C terminus are associated with homocystinuria.

    View details for DOI 10.1016/j.jpeds.2008.10.043

    View details for Web of Science ID 000264808000020

    View details for PubMedID 19058814

  • Segregation of ON and OFF retinogeniculate connectivity directed by patterned spontaneous activity JOURNAL OF NEUROPHYSIOLOGY Lee, C. W., Eglen, S. J., Wong, R. O. 2002; 88 (5): 2311-2321


    In many parts of the developing nervous system, the early patterns of connectivity are refined by processes that require neuronal activity. These processes are thought to involve Hebbian mechanisms that lead to strengthening and maintenance of inputs that display correlated pre- and postsynaptic activity and elimination of inputs that fire asynchronously. Here we investigated the role of patterned spontaneous retinal activity and Hebbian synaptic mechanisms on segregation of ON and OFF retinal afferents in the dorsal lateral geniculate nucleus (dLGN) of the developing ferret visual system. We recorded extracellularly the spontaneous spike activity of neighboring pairs of ganglion cells and found that OFF cells have significantly higher mean firing rates than ON cells. Spiking is best correlated between cells of the same sign (ON, ON; OFF, OFF) compared with cells of opposite sign (ON, OFF). We then constructed a simple Hebbian model of retinogeniculate synaptic development based on a correlational framework. Using our recorded activity patterns, together with previous calcium-imaging data, we show that endogenous retinal activity, coupled with Hebbian mechanisms of synaptic development, can drive the segregation of ON and OFF retinal inputs to the dLGN. Segregation occurs robustly when heterosynaptic competition is present within time windows of 50-500 ms. In addition, our results suggest that the initial patterns of connectivity (biases in convergence of inputs) and the strength of inhibition in the network each play a crucial role in determining whether ON or OFF inputs dominate at maturity.

    View details for DOI 10.1152/jn.00372.2002

    View details for Web of Science ID 000179080900015

    View details for PubMedID 12424272

  • A nonlinear Hebbian network that learns to detect disparity in random-dot stereograms NEURAL COMPUTATION Lee, C. W., Olshausen, B. A. 1996; 8 (3): 545-566


    An intrinsic limitation of linear, Hebbian networks is that they are capable of learning only from the linear pairwise correlations within an input stream. To explore what higher forms of structure could be learned with a nonlinear Hebbian network, we constructed a model network containing a simple form of nonlinearity and we applied it to the problem of learning to detect the disparities present in random-dot stereograms. The network consists of three layers, with nonlinear sigmoidal activation functions in the second-layer units. The nonlinearities allow the second layer to transform the pixel-based representation in the input layer into a new representation based on coupled pairs of left-right inputs. The third layer of the network then clusters patterns occurring on the second-layer outputs according to their disparity via a standard competitive learning rule. Analysis of the network dynamics shows that the second-layer units' nonlinearities interact with the Hebbian learning rule to expand the region over which pairs of left-right inputs are stable. The learning rule is neurobiologically inspired and plausible, and the model may shed light on how the nervous system learns to use coincidence detection in general.

    View details for Web of Science ID A1996UA30300006

    View details for PubMedID 8868567

  • Computerized mappings of the cerebral cortex: A multiresolution flattening method and a surface-based coordinate system JOURNAL OF COGNITIVE NEUROSCIENCE Drury, H. A., VANESSEN, D. C., Anderson, C. H., Lee, C. W., COOGAN, T. A., Lewis, J. W. 1996; 8 (1): 1-28


    We present a new method for generating two-dimensional maps of the cerebral cortex. Our computerized, two-stage flattening method takes as its input any well-defined representation of a surface within the three-dimensional cortex. The first stage rapidly converts this surface to a topologically correct two-dimensional map, without regard for the amount of distortion introduced. The second stage reduces distortions using a multiresolution strategy that makes gross shape changes on a coarsely sampled map and further shape refinements on progressively finer resolution maps. We demonstrate the utility of this approach by creating flat maps of the entire cerebral cortex in the macaque monkey and by displaying various types of experimental data on such maps. We also introduce a surface-based coordinate system that has advantages over conventional stereotaxic coordinates and is relevant to studies of cortical organization in humans as well as non-human primates. Together, these methods provide an improved basis for quantitative studies of individual variability in cortical organization.

    View details for Web of Science ID A1996UB45900002

    View details for PubMedID 11539144