Stewart Mostofsky, M.D. - US grants

This K08 Award proposal is a three-years plan to develop the candidate into an independent investigator in the area developmental behavioral neurology and includes a study of the cortical and subcortical brain mechanisms that may underlie behaviors observed in children with Attention Deficit Hyperactivity Disorder (ADHD). This proposal provides for the candidate's development through formal course work, extensive mentorship in a collaborative research environment, and implementation of a study that is the first step towards a larger body of research aimed at understanding the neurobiological basis of ADHD, a complex, heterogeneous disorder that has been reported to have multiple etiologies and is in need of specification at the neurobiological level. Primary mentorship will be provided by Martha B. Denckla, M.D., Director of the candidate's department. Dr Denckla is a leading researcher in the field of developmental behavioral neurology who has extensive experience training young investigators. In addition, a panel of seven consultants, most of whom work in a collaborative environment with the candidate and mentor, will each provide training on a specialized area of expertise critical to the candidates development. Immediate career goals include acquiring knowledge and experience in the use of neurobehavioral paradigms and relatively new techniques (volumetric and functional Magnetic Resonance Imaging (MRI) as theses apply to the study of frontal-subcortical systems involved in the pathogenesis of ADHD. Long-term career goals are directed towards using these paradigms and techniques to investigate the role of aberrant brain mechanisms in the pathogenesis of developmental disorders such as ADHD, autism, Tourette's syndrome, and learning disabilities. The study will examine 40 children with ADHD ages 8 through 12 years, and 40 controls. Experimental neurobehavioral paradigms, morphometric MRI and functional MRI (fMRI) will be used to investigate the hypothesis that the syndrome of ADHD is related to dysfunction within an anterior frontal network, comprised of the frontal lobes and interconnected subcortical structures including the basal ganglia and cerebellum.

DESCRIPTION (provided by applicant): The overall goal of this project is to evaluate a new model for understanding the neurologic basis of Attention Deficit Hyperactivity Disorder (ADHD). Theories about neural mechanisms have recently focused on deficient response inhibition as one of the fundamental features driving the pathophysiology of ADHD; however, the nature of these deficits and their neurologic underpinnings remain unclear. Previous investigators have attempted to view response inhibition as a unitary phenomenon, theorizing that its neurologic basis could be localized to a specific frontal region (i.e., a specific frontal sub-cortical circuit), which, once recognized, could then be regarded as the brain basis for ADHD. Recent data from electrophysiology and imaging studies, however, support a multiple-domain model of response inhibition according to which the specific frontal-sub-cortical circuit crucial for response inhibition depends on the nature of the task skeletomotor, oculomotor, cognitive, socioemotional) being performed. In ADHD, there is evidence for impairment on tasks reflecting all four of these domains of response inhibition; furthermore, recent imaging findings suggest that frontal abnormalities in ADHD involve regions within both premotor and prefrontal cortices. Accordingly, we propose to use neurobehavioral testing, anatomic MRI, and functional MRI to investigate the hypothesis that deficient response inhibition associated with ADHD is not a unitary phenomenon localized to one specific frontal region. Rather, impairments across skeletomotor, oculomotor, cognitive, and socioemotional domains of response inhibition are each linked to abnormality within a specific frontal circuit: skeletomotor, oculomotor, dorsolateral prefrontal and orbitoficontal, respectively. The findings will not only contribute to increased understanding of the pathophysiology of ADHD, but could also lay the foundation for neurobiological sub-typing of ADHD and in doing so improve diagnosis and decisions regarding targeted interventions. The conduct of this research project will also provide opportunities to extend, in both scope and in level of depth, investigative work begun during the candidate's training grant and to collaborate with scientists in related fields of discipline, helping to establish the candidate as an independent investigator.

[unreadable] DESCRIPTION (provided by applicant): The overall goal of this project is to use state-of-the-art anatomic and functional imaging methods to evaluate evidence directed towards understanding the pathophysiology of Attention Deficit Hyperactivity Disorder (ADHD), a complex, heterogeneous disorder that is in need of specification at the neurologic level. Hypotheses about neural mechanisms have recently focused on deficient response inhibition as one of the features fundamental to the pathophysiology of ADHD; however, the nature of these deficits and their neurologic underpinnings remain unclear. Previous investigators have theorized that the neural mechanisms underlying response inhibition could be localized to a specific frontal/subcortical region, which, once recognized, could then be regarded as the brain basis for ADHD. Recent data from electrophysiology and imaging studies, however, support a multiple-domain model of response inhibition according to which the specific frontal-subcortical circuit crucial for response inhibition depends on the nature of the task being performed: skeletomotor, oculomotor, cognitive, socioemotional. In ADHD, there is evidence for impairment on tasks reflecting all four of these domains of response inhibition; furthermore, recent imaging findings suggest that frontal abnormalities in ADHD involve regions within both premotor and prefrontal cortices. Accordingly, we propose to investigate the hypothesis that deficient response inhibition associated with ADHD is not attributable to abnormality within one specific frontal/subcortical region. Rather, impairments referable to skeletomotor, oculomotor, cognitive, and socioemotional domains of response inhibition are each linked to abnormality within the corresponding frontal-subcortical circuit - skeletomotor, oculomotor, dorsolateral prefrontal and orbitofrontal, respectively. Anatomic MRI and functional MRI will be used to determine which frontal-subcortical circuits are affected in ADHD and at what level; furthermore, which anatomic component abnormalities are associated with which behavioral/cognitive deficits that define ADHD. The findings will advance our understanding of the neuropathophysiology of ADHD, and in doing so provide a foundation for diagnostic models based on neuroanatomic subtypes that go beyond that of symptomatic presentation; this can lead to an improved framework for understanding the genetic and neurochemical basis of ADHD and for examining therapies targeted at specific fundamental deficits associated with ADHD. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent disorder that can cause substantial impairment in family/social relationships and the ability to succeed in school and occupation;yet, fundamental aspects of the neurobiology of ADHD remain poorly understood. Anomalous motor development is a consistent, but infrequently studied characteristic observed with ADHD that can provide insight into the neurologic basis of the disorder. Children with ADHD fail to meet age-norms on timed repetitive and sequential movements and manifest a greater amount of motor overflow than age-matched controls. These findings, which can accurately distinguish ADHD children from normal controls and children with other neuropsychiatric disorders, suggest that ADHD is associated with abnormalities of motor cortex inhibitory systems. Functional magnetic resonance imaging (fMRI) and transcranial magnetic stimulation (TMS) can localize and quantify inhibition within the motor cortex and thus are particularly suited to study motor inhibition in ADHD. We hypothesize that the neuromotor anomalies present in children with ADHD are associated with deficient intracortical and interhemispheric inhibition of the motor cortex. This hypothesis is consistent with studies suggesting that the central deficit of ADHD is a failure to inhibit or delay a behavioral response. Our preliminary studies using both fMRI and TMS suggest that abnormalities of motor cortex inhibition are present in children with ADHD. The overall goal of this project is to investigate the neurologic basis of motor anomalies associated with ADHD. We propose three specific aims. In the first aim we will measure overflow movements using electromyography (EMG) and accelerometry, which will allow us to detect meaningful correlations between overflow measurements and neurophysiologic and imaging data. The second aim will focus on examining motor cortex inhibition using TMS. The third aim will assess patterns of fMRI activation during simple finger movements, including activation patterns associated with overflow movements. We will then examine the relationship between abnormalities of cortical inhibition and measures of hyperactivity and inattention. The data from these studies will yield critical insights into the neurobiological basis of ADHD and will also set the stage for future development of more effective therapies for this population.

Brain region; Comprehension; Development; Diffusion MRI; Diffusion Magnetic Resonance Imaging; Diffusion Weighted MRI; Functional Magnetic Resonance Imaging; Goals; Grant; Image; Learning; MRI, Functional; Magnetic Resonance Imaging, Functional; Methods and Techniques; Methods, Other; Nervous; Play; Reading; Research; Role; Semantic; Semantics; Techniques; diffusion tensor imaging; fMRI; imaging; neural; neuroimaging; relating to nervous system; social role

DESCRIPTION (provided by applicant): Attention-Deficit/Hyperactivity Disorder (ADHD) is a highly prevalent disorder in which impaired control of unwanted behavior affects family/social relationships and school performance. Motor response control, mediated by the basal ganglia and, at the cortical level, the supplementary motor area (SMA), is crucial for selecting to withhold unwanted responses and thereby necessary for effective inhibition of impulsive, hyperactive and off-task behavior that characterizes ADHD. Findings from the initial grant period established that ADHD is associated with abnormalities in motor response control, including impaired response inhibition and increased intrasubject response variability (ISV);additionally, the degree of impairment on these measures was found to be similar across tasks with differing cognitive demands, suggesting that impaired motor response control is a primary deficit in ADHD. Consistent with these behavioral observations, functional and anatomic MRI studies reveal children with ADHD show abnormalities in the SMA and interconnected regions of the basal ganglia. Functional MRI examination of brain-behavior correlations provided further insight, revealing that for children with ADHD, but not typically developing (TD) children, recruitment of the prefrontal cortex was associated with lower ISV, reflecting improved response control. The combined behavior and imaging findings provide the basis for a shift towards a more developmental model of ADHD in which dysfunction in neural systems critical to motor response control, established early in development, contributes to the pathophysiology of the disorder, while function of later-developing prefrontal systems facilitates compensation. The goals of the current proposal are to: use novel imaging methods to investigate the hypothesis that abnormalities in a neural circuit involving the SMA and basal ganglia contribute to ADHD-associated impairments in response control (Aim 1), investigate the hypothesis that successful motor response control in some children with ADHD depends on recruitment of prefrontal (rather than typical premotor) circuits (Aims 2 and 3), and investigate whether a presentation format that increases vigilance can also result in improved response control (Aim 4). The proposed studies will not only help to identify relevant intermediate endophenotypes of ADHD but also might point to potentially effective treatment strategies. PUBLIC HEALTH RELEVANCE: Attention Deficit Hyperactivity Disorder (ADHD) is a common disorder in childhood in which problems with response control affect social relationships and school performance. The purpose of this research is to identify differences in brain structure contributing to ADHD-associated impairments in response control and to identify mechanisms that help children with ADHD improve task performance. This work may help to improve diagnosis and treatment of ADHD.

DESCRIPTION (provided by applicant): Attention Deficit Hyperactivity Disorder (ADHD) is the most common childhood behavioral diagnosis. In addition, its symptoms of inattention and impulsivity occur pervasively in many genetic and acquired neurological and psychiatric diseases. Despite the short-term efficacy of psychostimulants to treat core ADHD symptoms in childhood, adult outcomes include high rates of academic underachievement, mental illness, substance abuse, and criminal activity. A critical obstacle to improving long term ADHD treatment outcomes is the lack of quantitative markers which correlate with symptoms and reveal neurobiological mechanisms in ways that could point toward more accurate prognosis and more effective future treatments. In research funded during the initial grant period we addressed this barrier by taking advantage of the relationship (in developmental timing and anatomic proximity) between motor control and both cognitive and emotional control to pursue the physiology of inhibitory mechanisms in ADHD. We developed, refined, and compared techniques to easily and precisely evaluate developing motor function and physiology in 8-12 year old children with ADHD. Using Transcranial Magnetic Stimulation (TMS) in motor cortex, we found that Short Interval Cortical Inhibition (SICI), which is mediated by GABAergic interneurons and modulated by dopaminergic/ reward input, is reduced in children with ADHD. Importantly, this SICI reduction correlates with ADHD behavioral symptom severity as well as measures of motor impairment. We also generated novel preliminary findings linking motor cortex GABA, measured with magnetic resonance spectroscopy (MRS), to ADHD and SICI. The broad aim of this application is to 1) develop this ADHD SICI biomarker from resting M1 by extending from baseline (resting) cortical function (rSICI) to informative behavioral (response inhibition) and motivational (reward delay aversion) domains using innovative f(functional)SICI paradigms, 2) clarify the DAergic and GABAergic basis for SICI using pharmacologic challenge and magnetic resonance spectroscopy (MRS) techniques. AIM 1 To quantify fSICI during response inhibition as a biomarker of ADHD. AIM 2 To quantify fSICI during immediate and delayed reward presentation as a biomarker of ADHD. AIM 3 To quantify effects of DA on rSICI and fSICI. AIM 4 To determine whether motor cortex GABA levels 1) differ in ADHD vs. TD and 2) correlate with rSICI and fSICI in Aims 1 and 2. Achieving these aims will lay groundwork for future use of SICI as a pragmatic and biologically meaningful quantitative measure that can be applied to investigations of ADHD treatment, genetics, and risk factors for serious long term outcomes.

DESCRIPTION (provided by applicant): Attention Deficit Hyperactivity Disorder (ADHD) is the most common behavioral diagnosis in childhood. It incurs high medical costs, and can contribute to poor academic achievement, adult mental illness, substance abuse, and criminal behavior. Standard treatments, such as stimulant medications, primarily target symptoms, and long-term follow-up studies of children treated for ADHD reveal that their outcomes remain significantly worse as compared to typically developing (TD) peers. Thus, there are tremendous potential public health benefits for behavioral training programs that could remediate core features of ADHD. To this end, it is worth considering the long-standing observation that children with ADHD often demonstrate difficulties with motor control, including motor impersistence and failure to inhibit motor overflow, that parallel (and correlate with) difficultie with measures of higher-order behavioral control. Further, recent functional imaging and transcranial magnetic stimulation (TMS) findings reveal children with ADHD show abnormal recruitment of cortical inhibitory mechanisms, and that these physiologic measures of motor disinhibition robustly correlate with parent ratings of children's ADHD symptoms. Given these findings, it follows that movement-based interventions that aim to achieve improved behavioral control through engagement of the motor system offer a promising approach for targeting specific biological substrates of ADHD. Gains in cognitive and behavioral control have been observed using mindful movement training, including the focus of this application: Tai Chi. Despite the markedly greater brain plasticity observed in children as compared to older adults, pediatric applications of mindful movement training has been under utilized and under investigated. Addressing this gap, we propose to examine the efficacy of a movement-based Tai Chi training for children with ADHD. Given the recognized need for physiologic biomarkers in mindfulness and movement training studies, we will track changes in specific motor behavioral and physiologic (from TMS) markers, in addition to assessment of core ADHD symptoms. Our proposed design employs a Fast Fail Model to establish feasibility for progression to a clinical trial of efficacy for Tai Chi, and potentially other forms of mindful movement training, for childrn with ADHD. In the R21 phase, children with ADHD will engage in an 8-week Tai Chi training, with pre- and post-training measurement of motor system function via TMS and behavioral testing. Qualitative measures of student and parent experience will also be collected to complement data on core motor system targets. Upon completion of project milestones, the R33 phase will incorporate measures of ADHD symptom severity, as well as Research Domain Criteria (RDoC) measures of cognitive control. Further, the R33 phase will include a mid-training (in addition to pre- and post-training) assessment and we will investigate the hypothesis that changes in motor system measures will precede changes in ADHD symptom severity. The proposed study offers immense potential for the development of novel therapeutic approaches for ADHD with little risk of adverse reaction.

? DESCRIPTION (provided by applicant): Children with ADHD are at risk for a host of deleterious outcomes including impaired social relations, academic difficulties, criminality, and comorbid psychopathology (substance use, depression, anxiety). Many of these difficulties emerge and exacerbate during adolescence; therefore, it is crucial to understand the developmental trajectory of ADHD-associated changes in brain and behavior during this sensitive period. Further, there is increasing recognition that sex may be an important moderator of the clinical manifestations of ADHD, with adolescent boys showing more impulsive risk-taking while girls show more emotional dysregulation. To date, few studies of ADHD have taken a longitudinal approach to examining sex-based differences in behavioral changes and brain development into adolescence. For our currently funded grant, Neural Basis of Response Control in ADHD, we have thus far assessed a large cohort of school-age, pre-pubescent children with ADHD, in which we oversampled for girls. Our findings have revealed sexually dimorphic patterns of ADHD-associated impairments in response control and related structural brain abnormalities. Specifically, boys, but not girls, with ADHD show impaired basic motor response control and abnormalities in premotor structure; in contrast, girls with ADHD show impaired cognitive response control (e.g., when working memory is necessary to guide response selection/inhibition) and a predominance of abnormalities in prefrontal structure. Furthermore, we find that girls, but not boys, with ADHD show abnormalities in limbic structure and fail to show improvements in response control with reward. Given the known sexual dimorphisms in brain development, our findings in pre- pubescent children with ADHD raise important questions: Will these sex differences in abnormal patterns of brain structure and response control persist into adolescence and are they predictive of functional outcomes (e.g., academic, executive function, affective)? Therefore, building upon our results and successful recruitment, the goal of this renewal, is to examine Adolescent Changes in Brain and Behavior in Boys and Girls with ADHD. Using a longitudinal mixed model design, we will delineate and contrast developmental changes in response control and brain structure in ADHD girls and boys with that of typically-developing children, and examine the impact on adolescent mental health and behavioral outcomes. This approach is highly consistent with the strategic research objectives of the National Institute of Mental Health (NIMH) including Defining the developmental trajectories of mental disorders and Developing new ways of classifying mental disorders based on dimensions of observable behavior and neurobiological measures (Strategic Aims 2.1 and 1.4). Furthermore, in line with the Research Domain Criteria (RDoC) approach we are proposing to examine the dimensional construct of response control in children with ADHD from multiple levels of analysis including neural structure and connectivity, behavioral expression, and relation to functional outcomes, with the ultimate goal of identifying bio-behavioral markers of impairment and adjustment in children with ADHD.

? DESCRIPTION (provided by applicant): Tourette Syndrome (TS) is a common, childhood-onset neurological disorder characterized by involuntary and compulsively-performed patterned movements called tics, often preceded by sensory premonitory urges. Severe cases suffer substantial morbidity, despite the use of antipsychotics and even deep brain stimulation. The neurobiology of TS likely involves both atypical motor and sensory development; however, the majority of investigations to date have focused on the pathophysiology and treatment implications of dysfunction in motor system circuits only. Innovative methods are now available to probe the relationship between sensory function and motor control in children, investigations critical for developing better treatments for severe TS. In prior NIH funded research, Drs. Mostofsky and Gilbert, have employed careful clinical phenotyping, transcranial magnetic stimulation (TMS), and magnetic resonance spectroscopy (MRS) in evaluating pathophysiology of neurobehavioral disorders in children. This study builds on prior work as well as new data quantifying sensorimotor adaptation in TS. The over-arching hypothesis of this proposal is that GABAergic dysfunction in sensorimotor cortex results in a combination of abnormal sensory adaptation and impaired inhibitory motor function that underlie the emergence of symptoms of premonitory urges and tics in TS. The primary aim of this study is to quantify the relationship between sensory adaptation, inhibitory motor physiology, and urge and tic severity in children with TS and to compare these metrics with GABA levels in sensorimotor cortex and supplementary motor area. The investigators in this multi-PI study bring a unique combination of complementary scientific and clinical expertise, established scientific collaborations using innovative methods, and solid preliminary data to support this novel study of the pathophysiology of TS. To this end we propose the following aims: Aim 1: To evaluate and compare tactile sensitivity and adaptation in children with Tourette Syndrome (TS) and typically developing children (TDC) and determine whether impaired tactile adaptation is associated with premonitory urge severity and tic severity. Aim 2: To measure and compare inhibition in bilateral motor cortex (M1) using TMS in children with TS and TDC and determine whether reduced M1 inhibition is associated with tic severity. Aim 3: To measure and compare GABA levels in bilateral sensorimotor cortex (S1/M1) and supplementary motor area (SMA) in children with TS and TDC and determine whether S1/M1 GABA correlates with tactile adaptation, M1 inhibition, and premonitory urge and tic severity in TS. This innovative investigation of tactile, motor, and GABAergic dysfunction during emergence of tics in childhood will provide information vital for identifying treatments that reduce symptoms and prevent adverse outcomes in TS.

PROJECT SUMMARY/ABSTRACT ? RESEARCH COMPONENT The goal of this study entitled, ?Sleep and Circadian Dysfunction, Brain and Neurobehavioral Development in Autism? is to examine how disturbed sleep and altered circadian rest/activity rhythms (RARs) affect brain development, cognitive and adaptive function, and symptom severity in children with autism. Disturbed sleep is highly prevalent among children with autism spectrum disorder (ASD); over 60% are estimated to exhibit frequent sleep disturbances, including delayed sleep onset, fragmented nighttime sleep, and early-morning waking. While typically viewed as a consequence of ASD, disturbed sleep may not only have acute effects on cognition, adaptive functioning, and behavioral disturbances in this population. Notably, while many studies have evaluated the relationship between disordered sleep and neuroimaging, and neuroimaging and ASD, no work, to our knowledge, has integrated these topics. Furthermore, rigorous comparison of sleep and RAR metrics via parent-report and actigraphy have not been conducted in ASD. For this project, we propose to address these gaps through refining how sleep/wake problems are assessed in children with ASD by applying novel statistical modeling to both objective (using actigraphy) and parent-report (using CSHQ) measures and examining how disturbed sleep and altered RARs might affect brain structure and function in children with ASD. We embed this project in the Kennedy Krieger Institute-Johns Hopkins University Intellectual and Developmental Disabilities Research Center (IDDRC), leveraging crucial resources in phenotypic assessment, neuroimaging, biostatistics, and behavioral preparation for procedures that are available through the Center. At the same time, this investigative team brings new statistical and epidemiologic expertise to Center Cores, to conduct this study in 200 children with and without ASD. Via this project, Center resources will be used to further integrate this population with 493 children with extant data (but not wrist actigraphy), establishing a large-well phenotyped population for Center research. This work will help to refine and inform clinical and prevention practices among children with disordered sleep and ASD, providing methodological advances in sleep characterization and etiology.

PROJECT SUMMARY Attention Deficit Hyperactivity Disorder (ADHD), the most common childhood behavioral diagnosis, is a heterogeneous disorder linked to poor adult outcomes including high rates of academic underachievement, mental illness, substance abuse, and criminal activity. A critical obstacle to improving long-term ADHD outcomes is the current lack of quantitative neurophysiologic markers that could be used to improve precision treatment (behavioral and pharmacologic). In research funded during two grant periods, we addressed this barrier: Leveraging observed associations of ADHD-associated impairments in motor control with core impairments in cognitive and emotional control to pursue motor physiologic biomarkers linked to ADHD diagnosis and symptom severity. Engaging expertise of two research teams in different cities, we developed, refined, and effectively implemented innovative methods for evaluating motor physiology in school-age children with ADHD. During the first grant period we initiated Transcranial Magnetic Stimulation (TMS) investigations of motor cortex, discovering that Short Interval Cortical Inhibition (SICI) is reduced in children with ADHD and that this reduced SICI correlates with parent-ratings of ADHD symptom severity, with consistent findings across the two sites. During the second grant period, we developed (TMS-compatible) child-friendly games as paradigms for evaluation of motor cortex physiology under conditions requiring cognitive and emotional control. We found that SICI is reduced in children with ADHD during action selection and inhibition (as well as rest). Further, we discovered that TMS-evoked increases in motor evoked potential (MEP) amplitude from rest to task engagement (which we have denoted as ?Task Related Up Modulation? - TRUM) are significantly diminished in ADHD. Interestingly, while both TRUM and SICI robustly correlate with ADHD severity, they show limited correlation with each other. These published findings, in conjunction promising preliminary analyses of TMS- behavioral associations, suggest a model in which these two promising, distinct physiologic biomarkers reflect differential domains of dysfunction in ADHD, with SICI reflecting mechanisms linked to maintaining attention and filtering distractions, while TRUM reflects the capacity to modulate response control particularly under cognitively and emotionally demanding conditions. To pursue validation of SICI and TRUM as physiologic biomarkers, we now propose: AIM 1 To establish test-retest reliability of SICI and TRUM in children with ADHD and typically developing (TD) controls. AIM 2 To quantify stimulant treatment-induced changes in SICI and TRUM in children with ADHD as a validation of utility of these biomarkers for treatment focused studies. AIM 3 To explore the differential association of SICI and TRUM with ADHD-relevant RDoC constructs of Cognitive Control and Emotional Positive/Negative Valence. Achieving these aims will lay groundwork for future use of SICI and TRUM as a pragmatic and biologically meaningful quantitative measure that can be applied to investigations of precision ADHD treatment, genetics, and risk factors for serious long-term outcomes.

Approximately 1 in 54 children in the US is diagnosed with autism spectrum disorder (ASD). Given its high prevalence, there is a need for an automatic and scalable method to inform diagnosis and behavioral therapies. While prior work on finding early-emerging and reliable quantitative biomarkers of ASD has focused on non-motor features, abundant research evidence has revealed patterns of impaired motor imitation in a wide range of children with ASD, making motor imitation deficits a promising avenue to find a phenotypic biomarker. However, traditional imitation assessment methods often rely on expert-based observation, which is costly, time-consuming and error-prone, and lacks objectivity and scalability. Recent advances in computer vision and machine learning make artificial intelligence a promising technology to design an objective, reproducible and highly-scalable multimodal system functioning not only in well-equipped clinical setups but also at home for assessing imitation performance in children with ASD. However, critical challenges such as the design of specific imitation tasks for ASD assessment, the collection and labeling of multimodal data for training machine learning algorithms, and the development of novel fine-grained representations human movements and metrics for comparing such movements need to be addressed to test the validity, scalability and reproducibility of automatic motor imitation assessment algorithms to inform ASD diagnosis.

The overall goal of this project is to design, develop and test an objective, reproducible and highly-scalable multimodal system to observe children performing a brief videogame-like motor imitation task, quantitatively assess their motor imitation performance, and investigate its validity as a phenotypic biomarker for autism. Accomplishing this goal will require an interdisciplinary approach which combines expertise in autism, child development, computer vision and machine learning. Specifically, this project will: (1) design motor imitation tasks that are relevant for ASD assessment, (2) design, test and validate a scalable and flexible system to collect and label multimodal data of children imitating a sequence of movements; (3) design a novel fine-grained representation of human movements that can be learned efficiently and is suitable for comparing the children's movements to the movements they need to imitate; (4) develop novel computer vision and metric learning algorithms for learning and comparing multimodal representations of human movements, and (5) use such metrics to generate candidate imitations scores that can be used as potential quantitative biomarkers for ASD. The motor imitation assessment methods to be developed in this project could be used in a wide variety of applications beyond assessing children with ASD, such as providing imitation performance scores for video-based rehabilitation therapy, surgical skill assessment, athletic activities and other movement-based instructional activities.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.