Browsing by Author "Kana, Rajesh"

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    Accuracy of Machine Learning Algorithms for the Diagnosis of Autism Spectrum Disorder: Systematic Review and Meta-Analysis of Brain Magnetic Resonance Imaging Studies
    (JMIR, 2019) Moon, Sun Jae; Hwang, Jinseub; Kana, Rajesh; Torous, John; Kim, Jung Won; Ewha Womans University; Daegu University; University of Alabama Tuscaloosa; Harvard University; Beth Israel Deaconess Medical Center; Harvard Medical School; University of Alabama Birmingham
    Background: In the recent years, machine learning algorithms have been more widely and increasingly applied in biomedical fields. In particular, their application has been drawing more attention in the field of psychiatry, for instance, as diagnostic tests/tools for autism spectrum disorder (ASD). However, given their complexity and potential clinical implications, there is an ongoing need for further research on their accuracy. Objective: This study aimed to perform a systematic review and meta-analysis to summarize the available evidence for the accuracy of machine learning algorithms in diagnosing ASD. Methods: The following databases were searched on November 28, 2018: MEDLINE, EMBASE, CINAHL Complete (with Open Dissertations), PsycINFO, and Institute of Electrical and Electronics Engineers Xplore Digital Library. Studies that used a machine learning algorithm partially or fully for distinguishing individuals with ASD from control subjects and provided accuracy measures were included in our analysis. The bivariate random effects model was applied to the pooled data in a meta-analysis. A subgroup analysis was used to investigate and resolve the source of heterogeneity between studies. True-positive, false-positive, false-negative, and true-negative values from individual studies were used to calculate the pooled sensitivity and specificity values, draw Summary Receiver Operating Characteristics curves, and obtain the area under the curve (AUC) and partial AUC (pAUC). Results: A total of 43 studies were included for the final analysis, of which a meta-analysis was performed on 40 studies (53 samples with 12,128 participants). A structural magnetic resonance imaging (sMRI) subgroup meta-analysis (12 samples with 1776 participants) showed a sensitivity of 0.83 (95% CI 0.76-0.89), a specificity of 0.84 (95% CI 0.74-0.91), and AUC/pAUC of 0.90/0.83. A functional magnetic resonance imaging/deep neural network subgroup meta-analysis (5 samples with 1345 participants) showed a sensitivity of 0.69 (95% CI 0.62-0.75), specificity of 0.66 (95% CI 0.61-0.70), and AUC/pAUC of 0.71/0.67. Conclusions: The accuracy of machine learning algorithms for diagnosis of ASD was considered acceptable by few accuracy measures only in cases of sMRI use; however, given the many limitations indicated in our study, further well-designed studies are warranted to extend the potential use of machine learning algorithms to clinical settings.
  • Thumbnail Image
    Item
    Community-Based Transition Support Programming for Autistic Emerging Adults
    (University of Alabama Libraries, 2024) Brewe, Alexis; White, Susan W.
    Becoming an independent adult is a critical life transition, characterized by new roles and responsibilities in employment, relationships, and education. For adolescents with Autism Spectrum Disorder (ASD), this transition is marked by unique challenges including unemployment/underemployment, social isolation, and difficulties living independently. Despite growing evidence of the utility of programs that prepare autistic individuals for adulthood, these programs are rarely adopted into routine clinical practice. The current study uses an implementation science approach to refine and test an existing, evidence-based transition support program, the Stepped Transition to Employment and Postsecondary Success program (STEPS) for community implementation. In Phase 1, qualitative feedback was sought from stakeholders (total n = 45; i.e., autistic adults, caregivers, and professionals involved in the adult transition process) on several barriers and facilitators to implementation of STEPS in the community. In Phase 2, STEPS was piloted with 12 autistic individuals aged 16-35 in a local community agency to examine program feasibility, acceptability, and initial clinical impact. Results supported feasibility of STEPS implementation, evidenced by 93% of treatment objectives being delivered as intended by the STEPS therapist and moderately strong therapeutic alliance (average rating = 3.29, possible range = 0-5) established with clients. Results also supported acceptability, evidenced by low attrition (91.67% retention rate), high session attendance (96.27% sessions attended as scheduled), high homework completion (84.10% homework completion), and high participant- and caregiver-reported program satisfaction (average ratings of >4 across all items, possible range 1-5). Results also partially support the clinical impact of STEPS, evidenced by clinically significant change and reliable improvement in participants' transition readiness, as well as secondary measures of adult functioning, self-knowledge, self-determination, and self-regulation. In Phase 3, participants from Phase 2 completed exit interviews to provide final input on STEPS content, which was used to prepare STEPS for community implementation. Findings informed future community implementation of STEPS, and produced a fully scalable, stakeholder-informed program that was developed to address implementation challenges and ready for community deployment. Future research could utilize innovative implementation approaches (i.e., hybrid effectiveness-implementation trials) to test strategies that might promote adoption and long-term sustainability of STEPS in communities.
  • Thumbnail Image
    Item
    Dignity During a Pandemic: Dignity Therapy Delivered Through Telehealth is not Feasible in the Deep South
    (University of Alabama Libraries, 2024) Reel, Candice Danelle; Allen, Rebecca S.
    As time is limited, creation of a legacy document, particularly when aided by a care partner, is an effective method of facilitating a sense of dignity. However, access to care has been a problem for many individuals enrolled in community dwelling hospice care. Providing Dignity Therapy, (DT) a short-term individualized psychotherapy intervention for those at end of life, via telehealth could be one possible response to address this lack of access. Enrollment in hospice is often late in the disease process, indicating a need for short term interventions and a consideration of hospice participant attrition rates. The current study examined feasibility and efficacy of a telehealth delivery of the DT protocol to community dwelling hospice patients and their care partners and investigated challenges associated with hospice research recruitment through semi-structured interviews with hospice staff. Results of feasibility showed three potential participants were recruited but none consented to participate. The results from the feasibility study precluded our ability to assess efficacy as planned. Seven members of the hospice staff completed qualitative interviews designed to understand the lack of feasibility of this study. Results identified four main themes that point to the value of the DT intervention, an overwhelming disapproval of telehealth delivery of interventions, a close consideration of research methods, and the need for future research to further the advancement and clinical use of this effective intervention, particularly in rural and underserved areas.
  • Thumbnail Image
    Item
    Parental Burnout in Mother and Fathers of Children with and without Autism Spectrum Disorder During the COVID-19 Pandemic
    (University of Alabama Libraries, 2022) Paisley, Courtney Ann; Tomeny, Theodore S.; University of Alabama Tuscaloosa
    Objective: This project aimed to identify if and how experiences and functioning differ for mothers and fathers of typically developing (TD) children and mothers and fathers of children with autism spectrum disorder (ASD) in the midst of the COVID-19 pandemic. The primary focus is on parental burnout and associated mental health problems, parenting behaviors, and child behavior problems. An exploratory aim examined the differences in parental resilience. Method: The sample was comprised of 185 parents of children with and without ASD ages 4 and 16 years. Parents self-reported on measures of psychological functioning, parental burnout, behaviors, and resilience, and child behaviors. Results: The ASD group was found to have higher levels of depression, anxiety, and all types of parental burnout. Fathers in the ASD group reported higher levels of anxiety, depression, and burnout than mothers. No differences were found between mothers and fathers or between groups in level of acceptance, but group and gender differences were found in use of psychological and firm control. Fathers in both groups reported lower levels of resilience related to knowledge of their child's characteristics relative to mothers. Fathers in the ASD group also reported lower levels of social support than mothers in the ASD group and fathers in the TD group. However, no differences were found between groups or between mothers and fathers in positive perception of parenting. Conclusions: This study sheds light on how parents' experiences of children with and without ASD differed during the COVID-19 pandemic. Given the high percentage of parents of children with ASD who reported parental burnout, it is essential for clinicians to assess parents' level of functioning and feelings related to their parenting role. This study also suggest that fathers are struggling more psychologically and are more severely burned out than mothers, which highlights the importance of the inclusion of fathers in both research and clinical services.
  • Thumbnail Image
    Item
    Psychometric Properties of the Youth Psychopathic Traits Inventory – Short Version (YPI-S): an Investigation of Factor Structure, Item Function, and Convergent and Divergent Validity
    (University of Alabama Libraries, 2022) Bontemps, Andrew Price; Salekin, Randall T.; University of Alabama Tuscaloosa
    Few studies have investigated the factor structure or item functioning of the Youth Psychopathic Traits Inventory (YPI) family of measures and fewer still have investigated the short form of the YPI (Andershed, et al., 2002), the YPI-S (vanBaardewijk, et al. 2010). The current study attempts to fill a gap in the literature by investigating the factor structure and factor invariance of the YPI-S and its subscales as well as using an Item Response Theory (IRT) approach to investigate item functioning overall and differential item functioning (DIF) based on race and gender. A diverse group of high school students (N = 288, Mage = 15.53, 44.8% male, 60% White) was recruited from a mid-sized Southeastern city in the United States. Confirmatory factor analyses were conducted to determine the best-fitting factor structure which was then subjected to invariance testing across race and gender. Analyses showed good support for measure invariance across race and gender using a hierarchical factor structure. IRT analyses revealed overall strong functioning of most items. DIF analyses were conducted at the whole-measure and subscale level. At the whole-measure level no DIF was found based on race, but six items were identified as displaying DIF across gender showing greater discrimination for boys on two items (from the GM subscale) and greater discrimination for girls on four items (from the CU and II subscales). The DIF analysis suggests that there may be a difference in item functioning across gender, especially within the Grandiose-Manipulative and Callous-Unemotional subscales of the YPI-S.
  • Thumbnail Image
    Item
    The spike-and-slab elastic net as a classification tool in Alzheimer's disease
    (PLOS, 2022) Alzheimer's Dis Neuroimaging Intia; Leach, Justin M.; Edwards, Lloyd J.; Kana, Rajesh; Visscher, Kristina; Yi, Nengjun; Aban, Inmaculada; University of Alabama Birmingham; University of Alabama Tuscaloosa
    Alzheimer's disease (AD) is the leading cause of dementia and has received considerable research attention, including using neuroimaging biomarkers to classify patients and/or predict disease progression. Generalized linear models, e.g., logistic regression, can be used as classifiers, but since the spatial measurements are correlated and often outnumber subjects, penalized and/or Bayesian models will be identifiable, while classical models often will not. Many useful models, e.g., the elastic net and spike-and-slab lasso, perform automatic variable selection, which removes extraneous predictors and reduces model variance, but neither model exploits spatial information in selecting variables. Spatial information can be incorporated into variable selection by placing intrinsic autoregressive priors on the logit probabilities of inclusion within a spike-and-slab elastic net framework. We demonstrate the ability of this framework to improve classification performance by using cortical thickness and tau-PET images from the Alzheimer's Disease Neuroimaging Initiative (ADNI) to classify subjects as cognitively normal or having dementia, and by using a simulation study to examine model performance using finer resolution images.