一个可解释的预测模型,用于焦虑症状风险在中国老年人腹部肥胖使用机器学习和夏普利添加物扩展方法
Tengfei Niu1, Shiwei Cao2, Jingyu Cheng3
1Department of Basic Courses, Chongqing Medical and Pharmaceutical College, Chongqing, China.
Frontiers in psychiatry
|December 25, 2024
概括
在患有腹部肥胖的老年人中,早期发现焦虑是健康衰老的关键. 一个机器学习模型准确地预测了焦虑风险,确定了"看好前景"作为一个关键的保护因素.
科学领域:
- 老年学是一门学科.
- 公共卫生 公共卫生
- 计算医学是一种计算医学.
背景情况:
- 老年人的腹部肥胖与焦虑症状负担增加有关.
- 在这个人口群体中早期发现焦虑症可以改善健康的衰老结果.
- 针对性干预对于管理肥胖的老年人焦虑至关重要.
研究的目的:
- 开发和验证一个预测模型的焦虑症状在中国老年人与腹部肥胖.
- 确定这一群体中焦虑的关键风险和保护因素.
- 利用机器学习进行精确的焦虑风险预测.
主要方法:
- 利用了2017-2018年中国长度健康长寿调查 (CLHLS) 的2,427名参与者的数据.
- 采用LASSO回归用于变量选择和XGBoost (极端梯度提升) 用于预测建模.
- 应用夏普利添加式解释 (SHAP) 用于模型解释和特征重要性分析.
主要成果:
- 确定了9个关键变量,包括"看好前景",自我报告的经济状况和生活质量.
- XGBoost 模型对焦虑症状有很好的预测效果.
- "看看光明的一面"是最重要的保护因素,而"自我报告的生活质量"是最不重要的.
结论:
- 机器学习,特别是XGBoost和SHAP,为评估腹部肥胖的老年人焦虑风险提供了准确和可解释的方法.
- 该模型促进了及时,有针对性的干预措施,以减轻焦虑和促进健康的衰老.
- 识别乐观等关键因素可以为个性化预防策略提供信息.
相关概念视频
Generalized Anxiety Disorder
29
Generalized Anxiety Disorder (GAD) is a chronic condition characterized by excessive and uncontrollable worry that persists for at least six months, significantly interfering with daily functioning. Unlike situational anxiety, which arises in response to specific stressors, GAD often occurs without a clear cause. Individuals may experience disproportionate worry about work, health, or relationships. For instance, a person might continuously fear poor health despite normal medical evaluations or...
29
Theory of Romantic Attachment in Adulthood
42.8K
Attachment is a long-standing connection or bond with others. While Attachment Theory was conceived in developmental psychology to describe infant-caregiver bonding, it's been extended into adulthood to include romantic relationships.
42.8K


