这篇论文发了个叫PsyBridge的框架，能同时评估抑郁、焦虑、认知和人格，准确率0.84比单用量表高。搞心理健康AI的可以看看。
PsyBridge提出了一个混合智能框架，整合临床验证的PHQ-9和GAD-7量表、认知评估和人格特征，通过加权聚合生成可解释的心理健康风险分类。基于500个患者画像的半合成数据集，整体准确率达0.84，优于单独使用PHQ-9和GAD-7。敏感性和消融实验表明，整合认知和人格组件在中等风险预测中减少不一致性，提升分类稳定性。该框架为数字医疗和远程医疗环境提供可扩展、可解释的AI辅助决策支持。
PsyBridge: A Hybrid Intelligent Framework for Multi-Dimensional Mental Health Assessment and Decision Support
Mental health assessment commonly relies on isolated screening instruments or data-driven models that often lack interpretability and multi-dimensional integration. Existing approaches frequently focus on individual indicators such as depression or anxiety while providing limited support for comprehensive and explainable decision-making. To address this limitation, this study proposes PsyBridge, a hybrid intelligent decision-support framework designed for multi-dimensional mental health assessment through the integration of clinically validated screening tools, cognitive evaluation, and personality profiling within a unified architecture. The proposed framework incorporates PHQ-9 and GAD-7 assessments alongside cognitive and behavioural indicators using a modular design and a weighted aggregation mechanism to generate interpretable mental health risk classifications and recommendations. To evaluate the framework, a semi-synthetic dataset consisting of 500 patient profiles representing varying severity levels was constructed based on clinically grounded score distributions. Experimental results demonstrate that PsyBridge achieves an overall accuracy of 0.84, outperforming standalone PHQ-9 and GAD-7 assessments while improving precision, recall, and F1-score. Sensitivity analysis and ablation studies further indicate that integrating cognitive and personality components contributes to more stable classification performance and reduces inconsistencies in moderate-risk prediction. The findings suggest that PsyBridge provides a scalable and interpretable approach for AI-assisted mental health decision support, particularly within digital healthcare and telehealth environments.