基于XGBoost-SHAP的阿尔茨海默病的可解释诊断框架
Fuliang Yi1, Hui Yang1, Durong Chen1
1Department of Health Statistics, School of Public Health, Shanxi Medical University, 56 South XinJian Road, Taiyuan, 030001, P.R. China.
BMC medical informatics and decision making
|July 25, 2023
概括
这项研究引入了一个可解释的机器学习框架,XGBoost-SHAP,通过解决阶级不平衡来改善阿尔茨海默病 (AD) 诊断. 该框架提高了分类性能,并确定了临床决策的关键预测因素.
科学领域:
- 机器学习在神经退行性疾病研究中的应用.
- 为医疗保健开发可解释的人工智能模型.
背景情况:
- 阿尔茨海默病 (AD) 进展中的阶级不平衡对基于机器学习 (ML) 的辅助诊断提出了挑战.
- 由于数据不平衡问题,现有的ML模型具有较低的诊断性能.
研究的目的:
- 开发一个可解释的框架,XGBoost-SHAP,以解决AD进展中的阶级不平衡.
- 为了实现正常认知 (NC),轻度认知障碍 (MCI) 和AD的准确多重分类.
- 确定AD诊断中临床决策的关键预测特征.
主要方法:
- 利用来自阿尔茨海默病神经成像计划 (ADNI) 和国家阿尔茨海默病协调中心 (NACC) 数据库的患者数据.
- 采用极端梯度提升 (XGBoost),对不平衡数据进行样本重量调整.
- 综合的沙普利添加式解释 (SHAP) 用于模型解释性和特征分析.
主要成果:
- 与其他ML模型相比,XGBoost-SHAP框架展示了优越的分类性能.
- 获得了高精度 (87.57%在ADNI上,80.52%在NACC上) 和AUC (0.91在ADNI上,0.88在NACC上).
- 确定了与AD风险有不同的关联的顶部特征 (例如CDRSB,ADAS13,心室体积).
结论:
- 可解释的XGBoost-SHAP框架有效地处理AD多分类的不平衡数据.
- 通过最佳特征子集提供有价值的临床决策指导.
- 为AD预防和治疗策略提供了新的研究方向.
更多相关视频
04:22Author Spotlight: Exploring Sex-Specific Glial Signatures and Therapeutic Leads for Alzheimer's Disease
Published on: May 20, 2024
897
09:38Generalized Psychophysiological Interaction PPI Analysis of Memory Related Connectivity in Individuals at Genetic Risk for Alzheimer's Disease
Published on: November 14, 2017
15.0K
相关概念视频
01:26Alzheimer's Disease: Overview
524
Alzheimer's Disease (AD) is a continually advancing neurodegenerative disorder, distinguished by escalating memory loss, cognitive dysfunction, and dementia. The disease unfolds in three stages: preclinical, mild cognitive impairment (MCI), and dementia. Its onset is insidious, and the progression gradual, with the cause not well explained by other disorders.
The clinical diagnosis of AD hinges on the presence of memory and other cognitive impairments. Biomarkers, such as changes in Aβ...
The clinical diagnosis of AD hinges on the presence of memory and other cognitive impairments. Biomarkers, such as changes in Aβ...
524
01:22Alzheimer's Disease: Treatment
219
Alzheimer's Disease (AD), a neurodegenerative disorder, is pathologically identified by amyloid plaques and neurofibrillary tangles composed of tau protein. AD pharmacotherapy aims to manage cognitive symptoms, delay disease progression, and treat behavioral symptoms. The treatment is primarily symptomatic and palliative, with no definitive disease-modifying therapy available. Cholinesterase inhibitors, including donepezil (Aricept), rivastigmine (Exelon), and galantamine (Razadyne), are...
219