👤 Maksymilian A Brzezicki

Diagnosis of affective disorders among adolescent population links with the high risk of suicide attempt. The use of clinical psychological scales and biological markers may help to understand the background of suicidal process. Here we present the exploratory data study on retrospective suicide attempt risk factors and classification model of diagnosis conversion from major depressive disorder to bipolar disorder among adolescent population. This retrospective classification study was conducted on 45 adolescent/early-adulthood patients with the diagnosis of major depressive disorders. The psychological profile of patients was assessed with the use of standard clinical scales, like: Defence Style Questionnaire, Barrat Impulsiveness Scale, Beck Depression Inventory, Family APGAR, Emotional Intelligence Questionnaire and Temperament and Character Inventory. We assessed also the baseline concentration of blood-serum proteins: brain-derived neurotrophic factor, proBDNF, epidermal growth factor, macrophage migration inhibitory protein, and Stem Cell Factor. Suicide attempt history was determined at baseline (lifetime occurrence). The machine learning were used to assess the classification of the risk of suicidal attempt as well as diagnosis conversion from major depression to bipolar disorder. The winning models of machine learning were logistic regression and random forest. Regarding the suicidal attempt risk classification, significant coefficient were found mainly in Hamilton Depression Rating Scale (both factor and item assessment) and Temperament and Character Inventory (AUC = 0.74 (95% CI: 0.53-0.91), permutation p = 0.003). Serum biomarkers showed no discriminative ability (AUC = 0.35-0.40, p > 0.5) for suicide attempts in the past. We found not reliable clinical and biological data on the diagnosis conversion prediction. Clinical psychological scales, not peripheral biomarkers, distinguished suicide attempters in this exploratory analysis. Show less

Early-onset psychosis presents diagnostic challenges due to overlapping clinical presentations and complex comorbidities, typically requiring specialized tertiary care with extensive neuroimaging, neuropsychometric testing, and multidisciplinary evaluation. This case-control study investigated whether machine learning could integrate multiple diagnostic modalities to create an objective diagnostic framework for early-onset psychosis. We recruited 45 patients with early-onset psychosis and 34 healthy controls from a tertiary referral centre. Participants underwent comprehensive assessment including serum protein biomarker analysis (brain-derived neurotrophic factor, proBDNF, p75 neurotrophin receptor, S100B), neuropsychometric testing (Iowa Gambling Task, Simple Response Time, Zabor Verbal Task), and demographic evaluation. Four machine learning algorithms (logistic regression, support vector machine, random forest, XGBoost) were trained on five feature combinations using nested cross-validation with hyperparameter optimization. XGBoost demonstrated superior performance, achieving optimal classification with the complete multimodal dataset (accuracy: 0.91 ± 0.08, precision: 0.92 ± 0.08, area under curve: 0.97 ± 0.04). Feature importance analysis revealed cognitive measures, particularly Zabor Verbal Task errors and response time parameters, as most discriminative, with brain-derived neurotrophic factor pathway components showing highest biomarker importance. Machine learning effectively integrated neuropsychometric and protein biomarker data for high-accuracy early-onset psychosis classification, with multimodal approaches outperforming single-domain assessments. Show less