← Back to Research Papers

Machine learning methods evaluation for identification of cognitive phenotypes in multiple sclerosis and their MRI correlates.

Authors: Romaniszyn-Kania P, Galus W, Wyszomirska J, Zawiślak-Fornagiel K, Bożek O, Ledwoń D, Kania D, Tuszy A, Siuda J, Mitas AW
Journal: Frontiers in neuroscience
mental health psychology open access

Abstract

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition defined by persistent deficits in social communication and interaction, alongside restricted, repetitive patterns of behavior, interests, or activities (). Under the current DSM-5-TR framework, ASD is conceptualized as a single spectrum diagnosis differentiated by severity levels rather than discrete subtypes, reflecting the wide heterogeneity in clinical presentation across individuals (). Global reviews now estimate that approximately 1 in 127 individuals are autistic, with substantial variation across studies, regions, and diagnostic practices (). In high-income countries, prevalence estimates from national surveillance data have reached 1 in 36 children, reflecting improved detection and broadened diagnostic criteria (). In the Arab region, including Palestine, systematic epidemiological data remain limited, though growing service utilization at autism care centers across the West Bank suggests a substantial and insufficiently documented burden of ASD in this population. The neurobiology of ASD involves complex genetic, epigenetic, and environmental interactions that affect early brain development. Neuroimaging and genetic studies have consistently implicated disruptions in synaptic connectivity, cortical organization, and the development of social brain networks (). These neurobiological disruptions have downstream effects on sensory processing, behavioral regulation, and homeostatic functions including appetite, gastrointestinal motility, and sleep architecture domains that are directly relevant to this study (, ). Feeding difficulties are among the most frequently reported comorbidities in ASD, affecting an estimated 46%–89% of affected children across studies (). These difficulties arise from multiple converging mechanisms. First, sensory processing differences are pervasive in ASD: children may exhibit hypersensitivity or hyposensitivity to gustatory, olfactory, tactile, and proprioceptive stimuli, resulting in strong food preferences or refusals, insistence on specific food textures, and an inability to tolerate novel foods (). Second, the rigid and repetitive behavioral patterns characteristic of ASD often manifest as strong adherence to limited food repertoires and heightened distress when food routines are disrupted (). Third, gastrointestinal (GI) dysfunction is common in ASD, estimated at 23%–70% across studies and contributes to discomfort-driven food avoidance and oral aversion (). Together, these factors place children with ASD at significant risk for nutritional deficiencies, growth faltering, and atypical oral behaviors, including the ingestion of non-food substances.