BRAIN. Broad Research in Artificial Intelligence and Neuroscience

e-ISSN: 2067-3957

Schizophrenia and Oral Health: A Comorbidity Model Integrating Oxidative Stress, Nutrition and Perspectives for Predictive Artificial Intelligence

Authors:

Video Presentation Landscape

Abstract

Schizophrenia is a severe psychiatric disorder characterized by profound cognitive, emotional and behavioral impairments, frequently accompanied by somatic comorbidities that remain underrecognized. Among these, oral health deterioration emerges as a significant concern. This pilot study compares male patients with schizophrenia (n = 20) and a matched control group without psychiatric or systemic disorders, examining the interplay between salivary oxidative stress biomarkers, dietary habits and oral health indicators, with the aim of constructing a schizophrenia-centred comorbidity model. Oral health was assessed using the DMFT index (Decayed, Missing and Filled Teeth), a standard measure reflecting cumulative dental caries experience. Oxidative stress was evaluated through malondialdehyde (MDA) levels, a marker of lipid peroxidation and superoxide dismutase (SOD) activity, an antioxidant enzyme reflecting the body’s defence against free radicals. Nutritional habits were recorded to explore lifestyle contributions to oral pathology. Moderate correlations between oxidative stress biomarkers and oral health parameters suggest that schizophrenia-related lifestyle and metabolic factors contribute to dental deterioration. No artificial intelligence (AI) methods were implemented in this study; however, the findings highlight the conceptual potential for future AI-based predictive models that integrate biochemical, behavioural and clinical data to identify oral health risks specifically in populations with schizophrenia. This pilot study highlights the interplay between oxidative stress, nutrition and oral health in schizophrenia. Overall, this pilot work provides a preliminary dataset and conceptual framework that may inform future computational and AI-driven predictive studies in larger, gender-balanced and more diverse cohorts.

Links

Full article
Video file (MP4)