Machine learning for oral frailty factors in hospitalized schizophrenia patients: two-stage feature selection and SHAP analysis - Scorecard - MDSpire
Advertisement
Utilizing Machine Learning to Identify Oral Frailty Determinants in Hospitalized Patients with Schizophrenia: A Two-Stage Feature Selection and SHAP Analysis Approach
Clinical Scorecard: Utilizing Machine Learning to Identify Oral Frailty Determinants in Hospitalized Patients with Schizophrenia: A Two-Stage Feature Selection and SHAP Analysis Approach
At a Glance
Category
Detail
Condition
Oral Frailty in Schizophrenia
Key Mechanisms
Machine Learning for risk factor identification and predictive modeling
Target Population
Long-term hospitalized patients with schizophrenia
Care Setting
Psychiatric hospitals
Key Highlights
Prevalence of Oral Frailty (OF) in this population is 69.3%
Random forest model achieved an AUC of 0.779 after optimization
Core risk factors identified include Number of Teeth, Psychiatric Hospitalizations, Self-discontinuation of Medication, Marital Status, and Age
Traditional regression models are limited in high-dimensional data processing
SHAP analysis enhances interpretability of machine learning models
Guideline-Based Recommendations
Diagnosis
Utilize the Oral Frailty Index-8 for assessment of OF
Management
Develop individualized oral intervention programs based on identified risk factors
Monitoring & Follow-up
Regularly assess oral health and frailty status in hospitalized SZ patients
Risks
Monitor for complications such as dysphagia and increased risk of aspiration pneumonia
Patient & Prescribing Data
Long-term hospitalized patients with schizophrenia
Address oral health issues through targeted interventions based on identified determinants
Clinical Best Practices
Implement machine learning approaches for risk assessment in oral health
Incorporate SHAP for model interpretability in clinical settings
Focus on comprehensive evaluations of oral and systemic health
