相关实验视频
Updated: Jan 15, 2026

A Murine Model of Hemodialysis Access-Related Hand Dysfunction
Published on: May 31, 2022
可解释的机器学习模型用于预测维护血液透析患者的蛋白质能量浪费
Genlian Cai1,2, Yujiao Zhang1,2, Mengyan Pan1,2
1Department of Nursing, The First Affiliated Hospital, Zhejiang University School of Medicine, No. 79 Qingchun Road, Shangcheng District, Hangzhou, Zhejiang, 310003, China.
一个可解释的机器学习模型可以识别维护血液透析 (MHD) 患者的蛋白质能量浪费 (PEW) 风险. 这种工具有助于早期营养干预,以获得更好的临床结果.
科学领域:
- 腎臟病學 (nephrology) 是一種醫學專業.
- 人工智能的人工智能
- 营养科学 营养科学
背景情况:
- 蛋白质能量浪费 (PEW) 是维持血液透析 (MHD) 患者的重大并发症,与不良临床结果有关.
- 早期识别和营养干预对于在这个人群中管理PEW至关重要.
研究的目的:
- 开发和验证一种可解释的机器学习 (ML) 模型,用于识别MHD患者的PEW风险.
- 为及时和个性化的营养管理提供决策支持工具.
主要方法:
- 分析了908名MHD患者的队列,数据分为培训 (635) 和测试 (273) 套.
- 最小绝对收缩和选择运算符 (LASSO) 回归被用于特征选择.
- 训练和验证了七个ML算法;使用Shapley添加式解释 (SHAPs) 解释了最好的模型,并将其部署为web应用程序.
主要成果:
- XGBoost模型显示出最高的预测性能,AUC为0.827.
- PEW风险的关键预测因素包括透析前肌,手握强度,非HDL-C,KT/V和hs-CRP.
- 该模型实现了良好的灵敏度 (0.727),特异性 (0.762) 和精度 (0.755).
结论:
- 一个可解释的ML模型为医疗保健专业人员提供了一种实际的方法,以检测MHD患者的PEW风险.
- 该模型支持早期,个性化的营养策略,以改善患者的治疗结果.
相关概念视频
Drug Dosing in Renal Diseases: Estimation of Glomerular Filtration Rate Based on Serum Creatinine Concentration
Chronic Kidney Disease III: Interprofessional Care
Overview of Protein Metabolism
Amino acids play various roles in the body once they are absorbed into cells. They are restructured...
Physiological Pharmacokinetic Models: Assumption with Protein Binding
Dialysis
Acute kidney injury develops suddenly and can be caused by pre-renal causes (e.g., hypovolemia, shock), intrinsic renal causes (e.g., acute tubular necrosis), or post-renal causes (e.g., urinary obstruction). In contrast, chronic renal failure progresses gradually over time and is often...
Nephrotic Syndrome II : Assessment and Medical Management
