Autism Spectrum Disorder (ASD) is a condition that carries high costs for families and the healthcare system, requiring extensive management both in terms of diagnosis and treatment. The implementation of AI-based systems in clinical practice represents a possible supportive solution that can help clinicians by providing more systematic meth- ods to monitor child behavior. The main advantage over more traditional observational approaches is to offer quantitative and refined analysis solutions that can be ecological at the same time. The relevance of AI in clinical applications can have a role both in the challenge of early detection and in designing intervention programs better tai- lored to the specific functioning of children with ASD. The research project presented in this dissertation focused on developing AI-based systems for fine-grained analysis of autism-related social behaviors and their validation in concrete clinical environments. Specifically, in Chapter 2, our first study is presented, which targets on implementing a computational phenotyping system to address the need for new early markers of the condition. Through fine-grained analytics of facial dynamics in videos, we identified a set of features that distinguished young (6-12 months) infants with ASD (18 ASD, 15 non-ASD) during unconstrained at-home interactions. In Chapters 3 and 4, we introduce EYE-C, a Behavior Imaging model for robust analysis of eye contact episodes in eco- logical therapist-child interactions. The system was validated in the clinical setting for personalized early intervention. First, we investigated the influence of extracted features in categorizing spectrum heterogeneity across a sample of 62 preschool (<6 years) chil- dren with ASD. Further, we tested our metrics as predictors of early intensive treatment outcomes in a sub-sample of 18 subjects with ASD. The project aims to demonstrate the feasibility of effective computational systems that are robust to the high variability of unstructured interactions, with emphasis on the applicative value in real-world scenar- ios. Even though based on limited sample sizes, the work presented may offer interesting insights into the perspective of integrating AI into clinical practice. The research project was funded by an FBK scholarship and developed in a double in- ternship at ODFLab (University of Trento) and the FBK Data Science for Health (DSH) research unit.

No description available.

No description available.

No description available.

No description available.

<p><strong>Background:</strong> Developmental neuroscience research suggests that relative differences in emerging social skills between very young children with ASD may influence the trajectories of multiple important developmental domains, such as language ability. Such “cascade” associations between developmental trajectories may contribute in important ways to the substantial heterogeneity in symptoms and functioning seen in children with ASD. However, longitudinal research has yet to test such “cascade” models of change in children diagnosed with ASD.</p> <p><strong>Objectives:</strong> In this dissertation I aimed to model cascade patterns of association between social competence and language ability pathways in the year after diagnosis of ASD in preschool-aged children. Data for 365 participants aged 2-4 years old who had been recently diagnosed with ASD and followed prospectively were obtained from the “Pathways To Better Outcomes in ASD Study”.</p> <p><strong>Methods:</strong> Study 1 aimed to determine whether social competence and structural language ability could be measured as distinct constructs that were invariant over time and between clinically relevant groups of young children with ASD. Study 2 modeled longitudinal reciprocal associations between these domains. Study 3 addressed the issue of variability and change within this sample, and tested whether baseline levels and rate of change in one domain were associated with subsequent growth in the other.</p> <p><strong>Results:</strong> Social competence and language ability constructs were measurably distinct and invariant in a young sample comprising verbal and non-verbal children with ASD. Only small reciprocal cascade effects were evident between social and language pathways. Levels of social competence at time of diagnosis were significantly associated with subsequent language growth.</p> <p><strong>Conclusions:</strong> In preschoolers with ASD, advantages in social competence as measured at time of diagnosis appear to “spill over” in a feed-forward cascade model to influence progress in language ability. Social competence and language ability then appear to develop along more specialized, less interrelated pathways.</p> / Doctor of Philosophy (PhD)

Background. Mothers of children with Autism Spectrum Disorder (ASD) report high levels of stress, depression, and poor well-being, with little sense of social support due to the high demands of their child’s care. The Incredible Years® for Autism and Language Delays (IY-ASLD) program offers a unique model in that there is a dual focus on both child and parent outcomes. However, there is a dearth of literature on whether IY-ASLD can improve parental mental health outcomes. While there is some evidence to suggest that in-vivo therapist-to-parent supportive statements reduce parental resistance to treatment, there remains a gap in the literature regarding whether these behaviors directly impact parental mental health outcomes. Therefore, the present study investigated the degree to which exposure to the IY-ASLD telehealth curriculum (total minutes in attendance across sessions) along with in-vivo therapist-parent group verbal support was related to parental mental health outcomes and parental verbal approval statements of the child at post-treatment after controlling for baseline mental health and verbal approval. Methods. Participants were seventeen mother-child dyads, with children ages 2-years-8-months to 5-years-old recruited from a preschool utilizing an Applied Behavior Analysis (ABA) approach to schooling. All or virtually all children referred to this school by their school district are at-risk for or have been diagnosed with ASD. All of the children had an IEP or IFSP with mandated speech services, 71% of the children met criteria for ASD (N = 12) on the Childhood Autism Rating Scale- 2-SF, and 2/3 of the sample had adaptive behavior composites below a standard score of 70 on the Vineland Adaptive Behavior Scales- Third Edition Comprehensive Teacher Form (Sparrow et al., 2016). Participants were assigned to one of four treatment groups based on language level of the child and convenience of the group meeting time. Groups met for 12-weekly IY-ASLD sessions via telehealth. Data were collected at the beginning, mid-point, and end of treatment in the form of a parent questionnaire covering demographics, parental well-being (World Health Organization- Five Well-Being Index; WHO, 1998), depressive symptoms (Patient Health Questionnaire; Kroenke et al., 2001), and parenting stress (Parenting Stress Index-Fourth Edition, Short Form; Abidin, 2012). At post-treatment, parents reported satisfaction with the IY-ASLD-T program (IY-ASLD Parent Program Satisfaction Questionnaire). Data were also collected throughout treatment by recording IY-ASLD parent sessions. From the recorded parent sessions, two primary variables were developed. The first was treatment dosage, calculated as the total minutes each parent attended over the course of the 12-week intervention. The second was exposure to supportive therapist verbalizations during the first 10-minutes of each session, which was coded using the Therapy Process Code (TPC) to measure therapist supportive verbalizations during therapist-parent interaction (Chamberlain et al., 1986). Additionally, data were collected via a virtual parent-child play session (PCI) at the beginning, mid-point, and end of treatment, and later coded using the Verbal Behavior Developmental Language Coding System (VBDT) to measure parental verbal approval to child (Greer & Ross, 2008). Results. Mothers in this study had low levels of depression (PHQ-9; M = 4.59, SD = 4.69), average parenting stress (PSI-4-SF; M = 78.12, SD = 23.78), and low levels of well-being (WHO-5; M = 12.82, SD = 4.47). There were no statistically significant changes in parental depression, parental stress, or observed verbal approval from pre- to post-treatment and the effect sizes were small based on the partial eta squared and Kendall’s W values as per Cohen’s (1988, 1994) guidelines (small = .20-.49, medium = .50-.79, large > .80). However, there was a statistically significant change in caregiver well-being from baseline, mid-treatment, to post-treatment and a small effect size based on Kendall’s W (W = .31). A Friedman test indicated that caregivers demonstrated significant improvements in well-being from baseline to mid-treatment, and the improvement remained at post-treatment, χ2 (2) = 10.04, p < .01. There was no significant relationship between treatment dosage or supportive verbalizations and change in caregiver well-being. There was a significant negative correlation between treatment dosage and well-being at mid-treatment (r = -.493, p = .044), trending in the same direction at post-treatment (r = -.369, p = .146), possibly suggesting that parents with lower well-being, and therefore in greater need of support, attended IY-ASLD more frequently. An additional unexpected finding indicated that parents who were exposed to more supportive verbalizations had statistically significantly lower well-being at post-treatment. One explanation proposed by the authors is that parents who had lower well-being, and attended IY-ASLD sessions more frequently, received more supportive therapist verbalizations, possibly because they pulled for supportive verbalizations from the therapist. Results from the Incredible Years satisfaction questionnaire indicated there was high satisfaction with the program and the group leaders. High ratings of satisfaction with the group leaders and the program overall indicated that group leaders in the program may have successfully accomplished providing a “good enough” sense of support and reinforcement to parents over the course of the intervention. Therefore, it’s possible that specific therapist-to-parent supportive verbalizations were less related to improvements in parental well-being, and rather, the overall sense of support from the therapist to the parents in the group, along with support between the parents in the group, were sufficient to result in improvements in maternal well-being over the course of the intervention. Conclusions. Over the course of the IY-ASLD-T intervention, mothers demonstrated significant improvements in their overall well-being. While exposure to treatment and to supportive verbalizations from therapist to parents were not related to residualized change in well-being, these seem to be important factors in understanding the needs of parents with young children with ASD (e.g., parents with low well-being had higher attendance and received more supportive verbalizations). Taken together, the results suggest maternal well-being is an important and promising target for evidence-based behavioral parenting interventions such as IY-ASLD.

Human genetics is one of the strongest risk factors for complex diseases. Understandingthe effects of genetic variations not only serves as a fundamental approach to studying disease mechanisms but also offers unprecedented opportunities for improved clinical screening, disease diagnosis and therapeutic discoveries. Despite decades of extensive DNA sequencing and genetic research involving large cohorts, two major challenges remain. First, the majority of disease risk genes remain unidentified due to limited statistical power. Second, the functional effects of rare variants, especially missense variants, in disease risk genes are understudied. In this thesis, I describe new computational approaches to address those challenges using statistical genetics and machine learning methods implementing intuition of biological mechanisms. First, I worked on a statistical framework that can identify disease related pathways from de novo coding variants data. I applied this framework to study the genetics of esophageal atresia / tracheoesophageal fistula (EA/TEF) and identified several potential disease causal pathways that involved in endosome trafficking. Next, I developed a new method to identifying disease risk genes by integrating genetic (rare de novo variants) and functional genomics data. Identifying risk genes using rare variants typically has low statistical power due to the rarity of genotype data. Using functional genomics data has the potential to address this challenge as it serves as informative priors of disease risk. Therefore, I developed a statistical method called VBASS. VBASS is a semi-supervised algorithm that uses a neural network to encode biological priors, such as cell type-specific expression values, into a rigorous Bayesian statistical model to increase statistical power. On simulated data, VBASS demonstrated proper error rate control and better power than current state-of-the-art methods. We applied VBASS to congenital heart disease (CHD) and autism spectrum disorder (ASD), identifying several novel disease risk genes along with their associated cell types. Finally, I focused on predicting the functional mechanisms of missense variants that cause diseases. Pathogenic missense variants may act through different modes of action (e.g., gain-of-function or loss-of-function) by affecting various aspects of protein function. These variants may result in distinct clinical conditions requiring different treatments, yet current computational tools cannot distinguish between them because their predictions heavily relied on evolutional conservation data. The recent breakthrough of AI-powered protein structure prediction tools provides an opportunity to address this challenge because the functional mechanisms of variants is intrinsically embedded in its structural properties. Therefore, I developed a deep learning method called PreMode. PreMode is a pretrained SE(3)-equivariant graph neural network model designed to capture the effects of missense variants from their structural contexts and evolutionary information. I pretrained PreMode using labeled pathogenicity data to enable the model to learn a general representation of variant effects, followed by protein-specific transfer learning to predict mode-of-action effects. I applied PreMode to the mode-of-action predictions of 17 genes and demonstrated that PreMode achieved state-of-the-art performance compared to existing models. PreMode has various applications, including identifying novel gain/loss-of-function variants, improving the study design of deep mutational scans and optimization in protein engineering.

Thesis (MEdPsych (Educational Psychology))--University of Stellenbosch, 2009. / This thesis focused on three adults with Autistic Spectrum Disorder (ASD) who are living in a group home in Cape Town. The purpose of this exploratory study was to explore and describe the three occupants' perceived quality of life and their experiences of living in the group home. The group home was established in 2005 as a pioneering project by Autism Western Cape, a regional non-profit organisation (NGO). It is designed for adults with high-functioning ASD who require low care. In this interpretative study, a multiple case study method of enquiry was used. The Personal Wellbeing Index – Intellectual Disability (PWI-ID), as developed by Cummins and Lau (2005), was used as the primary subjective outcome measure of quality of life. In addition, the study made use of participant observations undertaken during two semi-structured interviews with each participant and two focus group interviews. All three participants felt that they had enjoyed a higher level of quality of life, especially in the domain of 'personal relationships', since their arrival at the group home. However, they expressed a desire for more independence; a higher income; female companionship and better security at the home. These suggestions have been made available to the relevant sponsors of the housing project.

Thesis (MEdPsych)--Stellenbosch University, 2012. / Includes bibiliography / ENGLISH ABSTRACT: At birth every human being is at the starting point of many different journeys: journeys of discovery and change, and journeys of mental and physical growth. Most children follow a similar path of physical and mental growth to adulthood, achieving predetermined milestones at approximately the same age. But what happens to a child who cannot follow this path, a child born without a map? How do the diagnosis and subsequent educational journey of the child affect the parents of that child, parents who find themselves disabled by their experiences of parenthood and life? This study is an autoethnography. It was undertaken to reflect on the physical and emotional journey two parents experienced in finding an educational niche for their son who was diagnosed with Pervasive Developmental Delay – Not Otherwise Specified (PDD-NOS), an Autism Spectrum Disorder. In an attempt to engage and involve the non-academic audience, while at the same time addressing the analytical needs of the researcher audience, evocative autoethnographic co-constructed narratives were combined with analytic autoethnography. Not only did I aim to fill in some of the gaps in researcher knowledge about South African parents’ experiences in finding educational support for their children with pervasive developmental delays, but I also wanted to provide knowledge, hope and encouragement to other parents, especially those parents who are at the start of a journey leading to a brighter future for their child with special needs. / AFRIKAANSE OPSOMMING: By geboorte bevind elke mens hom by die beginpunt van verskeie reise: reise van ontdekking en verandering, en reise van geestelike en fisieke groei. Die meeste kinders volg ‘n gelyksoortige roete van geestelike en fisieke groei na volwassenheid, deur voorafbepaalde doelwitte op naastenby ooreenstemmende ouderdomme te bereik. Maar wat gebeur met ‘n kind wat nie hierdie pad kan volg nie, ‘n kind wat sonder ‘n roetekaart gebore word? Hoe beïnvloed die diagnose en gevolglike opvoedkundige reis van daardie kind sy of haar ouers, ouers wat hulself gestremd bevind in hul ervaring van ouerskap en die lewe? Hierdie studie is ‘n outo-etnografie. Dit reflekteer op die fisieke en emosionele reis deur twee ouers onderneem, in hul soeke na ‘n geskikte onderwysnis vir hul seun wat met PDD-NOS1, ‘n Outisme Spektrumversteuring, gediagnoseer is. In ‘n poging om die nie-akademiese gehoor te betrek, maar terselfdertyd die analitiese behoeftes van die navorsergehoor aan te spreek, is die tegnieke van stemmingsvolle outo-etnografiese mede-saamgestelde narratiewe en analitiese outo-etnografie gekombineer. Ek het nie slegs ten doel gehad om sommige gapings in navorsing rondom die ervarings van Suid-Afrikaanse ouers van kinders met Outisme Spektrumversteurings te vul nie, maar ook om kennis, hoop en aanmoediging te gee aan ander ouers, veral daardie ouers wat aan die begin staan van ‘n reis na ‘n beter toekoms vir hul kind met spesiale behoeftes.