A5029182367- Autism Spectrum Disorder Research
- Child Development and Digital Technology
- Virology and Viral Diseases
- Mobile Health and mHealth Applications
- Infant Health and Development
- Prostate Cancer Diagnosis and Treatment
- Emotion and Mood Recognition
- Mobile Crowdsensing and Crowdsourcing
- Spinal Dysraphism and Malformations
- Cancer survivorship and care
- Face recognition and analysis
- Speech and Audio Processing
- Neuroendocrine regulation and behavior
- COVID-19 diagnosis using AI
- Family and Disability Support Research
- Team Dynamics and Performance
- Global Cancer Incidence and Screening
- Cerebrospinal fluid and hydrocephalus
- Human-Automation Interaction and Safety
- Neuroscience of respiration and sleep
- Behavioral and Psychological Studies
- Genetics and Neurodevelopmental Disorders
- Digital Mental Health Interventions
- Image and Video Quality Assessment
- Bladder and Urothelial Cancer Treatments
Stanford University
2018-2023
Palo Alto University
2019-2020
University of North Carolina at Chapel Hill
2018
Duke Medical Center
2013-2015
Duke University Hospital
2013-2015
Duke University
2013-2014
Center for Human Genetics
2013
Autism spectrum disorder (ASD) is currently diagnosed using qualitative methods that measure between 20-100 behaviors, can span multiple appointments with trained clinicians, and take several hours to complete. In our previous work, we demonstrated the efficacy of machine learning classifiers accelerate process by collecting home videos US-based children, identifying a reduced subset behavioral features are scored untrained raters classifier determine children's "risk scores" for autism. We...
Autism spectrum disorder (ASD) is a neurodevelopmental that results in altered behavior, social development, and communication patterns. In recent years, autism prevalence has tripled, with 1 44 children now affected. Given traditional diagnosis lengthy, labor-intensive process requires the work of trained physicians, significant attention been given to developing systems automatically detect autism. We toward this goal by analyzing audio data, as prosody abnormalities are signal autism,...
Autism affects 1 in every 59 children the United States, according to estimates from Centers for Disease Control and Prevention's Developmental Disabilities Monitoring Network 2018. Although similar rates of autism are reported rural urban areas, families report greater difficulty accessing resources. An overwhelming number experience long waitlists diagnostic therapeutic services.
Autism spectrum disorder (ASD) is a developmental characterized by deficits in social communication and interaction, restricted repetitive behaviors interests. The incidence of ASD has increased recent years; it now estimated that approximately 1 40 children the United States are affected. Due part to increasing prevalence, access treatment become constrained. Hope lies mobile solutions provide therapy through artificial intelligence (AI) approaches, including facial emotion detection AI...
Mobilized telemedicine is becoming a key, and even necessary, facet of both precision health medicine. In this study, we evaluate the capability potential crowd virtual workers—defined as vetted members popular crowdsourcing platforms—to aid in task diagnosing autism. We workers when providing categorical ordinal behavioral ratings to unstructured public YouTube videos children with autism neurotypical controls. To emerging patterns that are consistent across independent crowds, target from...
Many children with autism cannot receive timely in-person diagnosis and therapy, especially in situations where access is limited by geography, socioeconomics, or global health concerns such as the current COVD-19 pandemic. Mobile solutions that work outside of traditional clinical environments can safeguard against gaps to quality care.The aim study examine engagement level therapeutic feasibility a mobile game platform for autism.We designed application, GuessWhat, which, its form,...
Automated emotion classification could aid those who struggle to recognize emotions, including children with developmental behavioral conditions such as autism. However, most computer vision recognition models are trained on adult and therefore underperform when applied child faces.
There has been significant progress in identifying genes that confer risk for autism spectrum disorders (ASDs). However, the heterogeneity of symptom presentation ASDs impedes detection ASD genes. One approach to understanding genetic influences on expression is evaluate relations between variants candidate and neural endophenotypes unaffected samples. Allelic variations oxytocin receptor (OXTR) gene small but which underlying mechanisms may involve associations variability signaling...
Obtaining a diagnosis of neuropsychiatric disorders such as autism requires long waiting times that can exceed year and be prohibitively expensive. Crowdsourcing approaches may provide scalable alternative accelerate general access to care permit underserved populations obtain an accurate diagnosis.
Abstract Autism Spectrum Disorder is a neuropsychiatric condition affecting 53 million children worldwide and for which early diagnosis critical to the outcome of behavior therapies. Machine learning applied features manually extracted from readily accessible videos (e.g., smartphones) has potential scale this diagnostic process. However, nearly unavoidable variability in video quality can lead missing that degrade algorithm performance. To manage uncertainty, we evaluated impact values...
Standard medical diagnosis of mental health conditions requires licensed experts who are increasingly outnumbered by those at risk, limiting reach. We test the hypothesis that a trustworthy crowd non-experts can efficiently annotate behavioral features needed for accurate machine learning detection common childhood developmental disorder Autism Spectrum Disorder (ASD) children under 8 years old. implement novel process identifying and certifying distributed workforce video feature...
Chiari Type I Malformation (CMI) is characterized by displacement of the cerebellar tonsils below base skull, resulting in significant neurologic morbidity. Although multiple lines evidence support a genetic contribution to disease, no genes have been identified. We therefore conducted largest whole genome linkage screen date using 367 individuals from 66 families with at least two presenting nonsyndromic CMI or without syringomyelia. Initial findings across all showed minimal for due...
Chiari Type I Malformation (CMI) is characterized by herniation of the cerebellar tonsils through base skull. Although tonsillar (CTH) hypothesized to result from an underdeveloped posterior cranial fossa (PF), patients are frequently diagnosed extent CTH without morphometric assessment. We recently completed largest CMI whole genome qualitative linkage screen date. Despite initial lack statistical evidence, stratified analyses using clinical criteria reduce heterogeneity resulted in a...
Digitally-delivered healthcare is well suited to address current inequities in the delivery of care due barriers access facilities. As COVID-19 pandemic phases out, we have a unique opportunity capitalize on familiarity with telemedicine approaches and continue advocate for mainstream adoption remote delivery. In this paper, specifically focus ability GuessWhat? smartphone-based charades-style gamified therapeutic intervention autism spectrum disorder (ASD) generate signal that distinguishes...
Previous research has found accumulating evidence for atypical reward processing in autism spectrum disorders (ASD), particularly the context of social rewards. Yet, this line focused largely on positive reinforcement, while little is known about negative reinforcement individuals with ASD.The present study examined neural responses to (a face displaying affect) and non-social (monetary loss) children ASD relative typically developing children, using functional magnetic resonance imaging...
Artificial Intelligence (A.I.) solutions are increasingly considered for telemedicine. For these methods to serve children and their families in home settings, it is crucial ensure the privacy of child parent or caregiver. To address this challenge, we explore potential global image transformations provide while preserving quality behavioral annotations. Crowd workers have previously been shown reliably annotate features unstructured videos, allowing machine learning classifiers detect...
Facial expression recognition (FER) is a critical computer vision task for variety of applications. Despite the widespread use FER, there dearth racially diverse facial emotion datasets which are enriched children, teens, and adults. To bridge this gap, we have built database using publicly available videos from TikTok, video-focused social networking service. We describe construction TikTok database. The dataset extracted 6428 scraped TikTok. consist 9392 distinct individuals labels 15...
Background: Automated emotion classification could aid those who struggle to recognize emotions, including children with developmental behavioral conditions such as autism. However, most computer vision recognition models are trained on adult and therefore underperform when applied child faces. Objective: We designed a strategy gamify the collection labeling of emotion-enriched images boost performance automatic level closer what will be needed for digital health care approaches. Methods:...
ABSTRACT Standard medical diagnosis of mental health conditions often requires licensed experts who are increasingly outnumbered by those at risk, limiting reach. We test the hypothesis that a trustworthy crowd non-experts can efficiently label features needed for accurate machine learning detection common childhood developmental disorder autism. implement novel process creating distributed workforce video feature extraction, selecting 102 workers from pool 1,107. Two previously validated...
Large, whole-genome sequencing (WGS) data sets containing families provide an important opportunity to identify crossovers and shared genetic material in siblings. However, the high variant calling error rates of WGS some areas genome can result spurious crossover calls, special inheritance status X Chromosome presents challenges. We have developed a hidden Markov model that addresses these issues by modeling variants presence error-prone regions inherited deletions. call our method...
Background: This project explores a new model of care that enhances survivorship planning and promotes health for men with localized prostate cancer transitioning to posttreatment self-management. Survivorship is important patients because its high incidence rate in the United States, frequent occurrence treatment-related side effects, reduced quality life (QOL) both their partners. A key component comprehensive plans (SCPs), documents summarize diagnosis, treatment, follow-up care. However,...