Education & Certifications
Bachelor of Science, Brown University, Cognitive Neuroscience (2013)
Deep-Learning for Automated Classification of Inferior Vena Cava Filter Types on Radiographs.
Journal of vascular and interventional radiology : JVIR
PURPOSE: To demonstrate the feasibility and evaluate the performance of a deep-learning convolutional neural network (CNN) classification model for automated identification of different types of inferior vena cava (IVC) filters on radiographs.MATERIALS AND METHODS: In total, 1,375 cropped radiographic images of 14 types of IVC filters were collected from patients enrolled in a single-center IVC filter registry, with 139 images withheld as a test set and the remainder used to train and validate the classification model. Image brightness, contrast, intensity, and rotation were varied to augment the training set. A 50-layer ResNet architecture with fixed pre-trained weights was trained using a soft margin loss over 50 epochs. The final model was evaluated on the test set.RESULTS: The CNN classification model achieved a F1 score of 0.97 (0.92-0.99) for the test set overall and of 1.00 for 10 of 14 individual filter types. Of the 139 test set images, 4 (2.9%) were misidentified, all mistaken for other filter types that appear highly similar. Heat maps elucidated salient features for each filter type that the model used for class prediction.CONCLUSIONS: A CNN classification model was successfully developed to identify 14 types of IVC filters on radiographs and demonstrated high performance. Further refinement and testing of the model is necessary before potential real-world application.
View details for DOI 10.1016/j.jvir.2019.05.026
View details for PubMedID 31542278
Posterior fossa syndrome and increased mean diffusivity in the olivary bodies.
Journal of neurosurgery. Pediatrics
OBJECTIVE: Posterior fossa syndrome (PFS) is a common postoperative complication following resection of posterior fossa tumors in children. It typically presents 1 to 2 days after surgery with mutism, ataxia, emotional lability, and other behavioral symptoms. Recent structural MRI studies have found an association between PFS and hypertrophic olivary degeneration, which is detectable as T2 hyperintensity in the inferior olivary nuclei (IONs) months after surgery. In this study, the authors investigated whether immediate postoperative diffusion tensor imaging (DTI) of the ION can serve as an early imaging marker of PFS.METHODS: The authors retrospectively reviewed pediatric brain tumor patients treated at their institution, Lucile Packard Children's Hospital at Stanford, from 2004 to 2016. They compared the immediate postoperative DTI studies obtained in 6 medulloblastoma patients who developed PFS to those of 6 age-matched controls.RESULTS: Patients with PFS had statistically significant increased mean diffusivity (MD) in the left ION (1085.17 ± 215.51 vs 860.17 ± 102.64, p = 0.044) and variably increased MD in the right ION (923.17 ± 119.2 vs 873.67 ± 60.16, p = 0.385) compared with age-matched controls. Patients with PFS had downward trending fractional anisotropy (FA) in both the left (0.28 ± 0.06 vs 0.23 ± 0.03, p = 0.085) and right (0.29 ± 0.06 vs 0.25 ± 0.02, p = 0.164) IONs compared with age-matched controls, although neither of these values reached statistical significance.CONCLUSIONS: Increased MD in the ION is associated with development of PFS. ION MD changes may represent an early imaging marker of PFS.
View details for DOI 10.3171/2019.5.PEDS1964
View details for PubMedID 31349230
Speech processing and plasticity in the right hemisphere predict variation in adult foreign language learning
2019; 192: 76–87
Foreign language learning in adulthood often takes place in classrooms where learning outcomes vary widely among students, for both initial learning and long-term retention. Despite the fundamental role of speech perception in first language acquisition, its role in foreign language learning outcomes remains unknown. Using a speech discrimination functional magnetic resonance imaging (fMRI) task and resting-state fMRI before and after an intensive, classroom-based, Mandarin Chinese course, we examined how variations in pre-training organization and pre-to-post reorganization of brain functions predicted successful language learning in male and female native English-speakers. Greater pre-training activation in right inferior frontal gyrus (IFG) to Mandarin speech was associated with better Mandarin attainment at the end of the course. After four weeks of class, learners showed overall increased activation in left IFG and left superior parietal lobule (SPL) to Mandarin speech, but in neither region was variation related to learning outcomes. Immediate attainment was associated with greater pre-to-post reduction of right IFG activation to Mandarin speech but also greater enhancement of resting-state connectivity between this region and both left IFG and left SPL. Long-term retention of Mandarin skills measured three months later was more accurately predicted by models using features of neural preparedness (pre-training activation) and neural plasticity (pre-to-post activation change) than models using behavior preparedness and plasticity features (pre-training speech discrimination accuracy and Mandarin attainment, respectively). These findings suggest that successful holistic foreign language acquisition in human adulthood requires right IFG engagement during initial learning but right IFG disengagement for long-term retention of language skills.
View details for DOI 10.1016/j.neuroimage.2019.03.008
View details for Web of Science ID 000463514600007
View details for PubMedID 30853566
One Hundred Years of Innovation: Automatic Detection of Brain Ventricular Volume using Deep Learning in a Large-Scale Multi-Institutional Study
LIPPINCOTT WILLIAMS & WILKINS. 2019
View details for Web of Science ID 000475965905285
Altered cerebral perfusion in children with Langerhans cell histiocytosis after chemotherapy.
Pediatric blood & cancer
Children with Langerhans cell histiocytosis (LCH) may develop a wide array of neurological symptoms, but associated cerebral physiologic changes are poorly understood. We examined cerebral hemodynamic properties of pediatric LCH using arterial spin-labeling (ASL) perfusion magnetic resonance imaging (MRI).A retrospective study was performed in 23 children with biopsy-proven LCH. Analysis was performed on routine brain MRI obtained before or after therapy. Region of interest (ROI) methodology was used to determine ASL cerebral blood flow (CBF) (mL/100 g/min) in the following bilateral regions: angular gyrus, anterior prefrontal cortex, orbitofrontal cortex, dorsal anterior cingulate cortex, and hippocampus. Quantile (median) regression was performed for each ROI location. CBF patterns were compared between pre- and posttreatment LCH patients as well as with age-matched healthy controls.Significantly reduced CBF was seen in posttreatment children with LCH compared to age-matched controls in angular gyrus (P = .046), anterior prefrontal cortex (P = .039), and dorsal anterior cingulate cortex (P = .023). Further analysis revealed dominant perfusion abnormalities in the right hemisphere. No significant perfusion differences were observed in the hippocampus or orbitofrontal cortex.Perfusion in specific cerebral regions may be consistently reduced in children with LCH, and may represent effects of underlying disease physiology and/or sequelae of chemotherapy. Studies that combine a formal cognitive assessment and hemodynamic data may further provide insight into perfusion deficits associated with the disease and the potential neurotoxic effects in children treated by chemotherapy.
View details for DOI 10.1002/pbc.28104
View details for PubMedID 31802628
- Late diagnosis of a rare urea cycle disorder mimicking Kleine-Levin syndrome. Neurology. Clinical practice 2018; 8 (6): e43–e45
Shared neuroanatomical substrates of impaired phonological working memory across reading disability and autism.
Biological psychiatry. Cognitive neuroscience and neuroimaging
2016; 1 (2): 169-177
Individuals with reading disability or individuals with autism spectrum disorder (ASD) are characterized, respectively, by their difficulties in reading or social communication, but both groups often have impaired phonological working memory (PWM). It is not known whether the impaired PWM reflects distinct or shared neuroanatomical abnormalities in these two diagnostic groups.White-matter structural connectivity via diffusion weighted imaging was examined in sixty-four children, ages 5-17 years, with reading disability, ASD, or typical development (TD), who were matched in age, gender, intelligence, and diffusion data quality.Children with reading disability and children with ASD exhibited reduced PWM compared to children with TD. The two diagnostic groups showed altered white-matter microstructure in the temporo-parietal portion of the left arcuate fasciculus (AF) and in the temporo-occipital portion of the right inferior longitudinal fasciculus (ILF), as indexed by reduced fractional anisotropy and increased radial diffusivity. Moreover, the structural integrity of the right ILF was positively correlated with PWM ability in the two diagnostic groups, but not in the TD group.These findings suggest that impaired PWM is transdiagnostically associated with shared neuroanatomical abnormalities in ASD and reading disability. Microstructural characteristics in left AF and right ILF may play important roles in the development of PWM. The right ILF may support a compensatory mechanism for children with impaired PWM.
View details for PubMedID 26949750
View details for PubMedCentralID PMC4776338
Impaired Frontal-Limbic White Matter Maturation in Children at Risk for Major Depression.
Cerebral cortex (New York, N.Y. : 1991)
Depression is among the most common neuropsychiatric disorders. It remains unclear whether brain abnormalities associated with depression reflect the pathological state of the disease or neurobiological traits predisposing individuals to depression. Parental history of depression is a risk factor that more than triples the risk of depression. We compared white matter (WM) microstructure cross-sectionally in 40 children ages 8-14 with versus without parental history of depression (At-Risk vs. Control). There were significant differences in age-related changes of fractional anisotropy (FA) between the groups, localized in the anterior fronto-limbic WM pathways, including the anterior cingulum and the genu of the corpus callosum. Control children exhibited typical increasing FA with age, whereas At-Risk children exhibited atypical decreasing FA with age in these fronto-limbic regions. Furthermore, dorsal cingulate FA significantly correlated with depressive symptoms for At-Risk children. The results suggest maturational WM microstructure differences in mood-regulatory neurocircuitry that may contribute to neurodevelopmental risk for depression. The study provides new insights into neurodevelopmental susceptibility to depression and related disabilities that may promote early preventive intervention approaches.
View details for DOI 10.1093/cercor/bhw250
View details for PubMedID 27578495
- White-matter structure in the right hemisphere predicts Mandarin Chinese learning success JOURNAL OF NEUROLINGUISTICS 2015; 33: 14-28
Modeling healthy male white matter and myelin development: 3 through 60 months of age
2014; 84: 742-752
An emerging hypothesis in developmental and behavioral disorders is that they arise from disorganized brain messaging or reduced connectivity. Given the importance of myelin to efficient brain communication, characterization of myelin development in infancy and childhood may provide salient information related to early connectivity deficits. In this work, we investigate regional and whole brain growth trajectories of the myelin water fraction, a quantitative magnetic resonance imaging measure sensitive and specific to myelin content, in data acquired from 122 healthy male children from 3 to 60months of age. We examine common growth functions to find the most representative model of myelin maturation and subsequently use the best of these models to develop a continuous population-averaged, four-dimensional model of normative myelination. Through comparisons with an independent sample of 63 male children across the same age span, we show that the developed model is representative of this population. This work contributes to understanding the trajectory of myelination in healthy infants and toddlers, furthering our knowledge of early brain development, and provides a model that may be useful for identifying developmental abnormalities.
View details for DOI 10.1016/j.neuroimage.2013.09.058
View details for Web of Science ID 000328868600068
View details for PubMedID 24095814
View details for PubMedCentralID PMC3895775
Pediatric neuroimaging using magnetic resonance imaging during non-sedated sleep
2014; 44 (1): 64-72
Etiological studies of many neurological and psychiatric disorders are increasingly turning toward longitudinal investigations of infant brain development in order to discern predisposing structural and/or functional differences prior to the onset of overt clinical symptoms. While MRI provides a noninvasive window into the developing brain, MRI of infants and toddlers is challenging due to the modality's extreme motion sensitivity and children's difficulty in remaining still during image acquisition.Here, we outline a broad research protocol for successful MRI of children under 4 years of age during natural, non-sedated sleep.All children were imaged during natural, non-sedated sleep. Active and passive measures to reduce acoustic noise were implemented to reduce the likelihood of the children waking up during acquisition. Foam cushions and vacuum immobilizers were used to limit intra-scan motion artifacts.More than 380 MRI datasets have been successfully acquired from 220 children younger than 4 years of age within the past 39 months. Implemented measures permitted children to remain asleep for the duration of the scan and allowed the data to be acquired with an overall 97% success rate.The proposed method greatly advances current pediatric imaging techniques and may be readily implemented in other research and clinical settings to facilitate and further improve pediatric neuroimaging.
View details for DOI 10.1007/s00247-013-2752-8
View details for Web of Science ID 000329357300010
View details for PubMedID 23917588
View details for PubMedCentralID PMC3889986
Breastfeeding and early white matter development: A cross-sectional study
2013; 82: 77-86
Does breastfeeding alter early brain development? The prevailing consensus from large epidemiological studies posits that early exclusive breastfeeding is associated with improved measures of IQ and cognitive functioning in later childhood and adolescence. Prior morphometric brain imaging studies support these findings, revealing increased white matter and sub-cortical gray matter volume, and parietal lobe cortical thickness, associated with IQ, in adolescents who were breastfed as infants compared to those who were exclusively formula-fed. Yet it remains unknown when these structural differences first manifest and when developmental differences that predict later performance improvements can be detected. In this study, we used quiet magnetic resonance imaging (MRI) scans to compare measures of white matter microstructure (mcDESPOT measures of myelin water fraction) in 133 healthy children from 10 months through 4 years of age, who were either exclusively breastfed a minimum of 3 months; exclusively formula-fed; or received a mixture of breast milk and formula. We also examined the relationship between breastfeeding duration and white matter microstructure. Breastfed children exhibited increased white matter development in later maturing frontal and association brain regions. Positive relationships between white matter microstructure and breastfeeding duration are also exhibited in several brain regions, that are anatomically consistent with observed improvements in cognitive and behavioral performance measures. While the mechanisms underlying these structural differences remains unclear, our findings provide new insight into the earliest developmental advantages associated with breastfeeding, and support the hypothesis that breast milk constituents promote healthy neural growth and white matter development.
View details for DOI 10.1016/j.neuroimage.2013.05.090
View details for Web of Science ID 000324568400009
View details for PubMedID 23721722
View details for PubMedCentralID PMC3777218