基于MIMIC-IV的特征选择和糖尿病患者的酸性糖尿病风险预测
Yang Liu1,2, Wei Mo1,2, He Wang1,2
1Endocrinology, The Fifth Clinical College of Guangzhou University of Chinese Medicine, Guangzhou, China.
Frontiers in endocrinology
|April 11, 2024
概括
糖尿病酸性症 (DKA) 的风险与血红素,血红蛋白,离子间隙,年龄和并发症指数有关. 使用这些因素的早期检测可以帮助预防糖尿病患者的DKA.
科学领域:
- 生物医学信息学 生物医学信息学
- 临床医学 临床医学
- 数据科学数据科学数据科学
背景情况:
- 糖尿病酸性酸症 (DKA) 是糖尿病 (DM) 的严重并发症.
- DKA显著影响患者的健康,需要及时管理.
- 识别DKA的早期预测因素对于及时干预至关重要.
研究的目的:
- 通过机器学习识别可预测DKA开发的关键基线特征.
- 为了实现针对性和早期预防DKA的策略.
主要方法:
- 使用MIMIC-IV数据集对2382名糖尿病患者 (1193名患有DKA,1189名没有).
- 采用皮尔森相关性和随机森林从42个基线特征中进行特征选择.
- 应用后勤回归,XGBoost,决策树,随机森林,SVM和k-NN用于DKA预测.
主要成果:
- 确定了DKA的顶级预测因素:平均血红素,平均血红蛋白,平均离子间隙,年龄和查尔森并发症指数.
- 平均血位 (F1=1.000) 和平均血红蛋白 (F1=0.978) 个别显示出高预测值.
- 机器学习模型 (逻辑回归,XGBoost,决策树,随机森林) 获得了F1的1.000分.
结论:
- 血红素,血红蛋白,离子间隙,年龄和查尔森并发症指数与DKA密切相关.
- 这五个因素对于在临床实践中早期发现和治疗DKA至关重要.
- 专注于这些特征可以减少DKA发生率.
更多相关视频
相关概念视频
01:15Receiver Operating Characteristic Plot
150
A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...
150
01:15Diabetes: Symptoms, Diagnosis, and Complications
528
For most patients, experiencing several weeks of polyuria, polydipsia, fatigue, and significant weight loss may indicate the presence of diabetes. Furthermore, adults displaying the phenotypic appearance of type 2 diabetes (particularly those who are obese and not initially insulin-requiring), may have islet cell autoantibodies, suggesting autoimmune-mediated β cell destruction and a diagnosis of latent autoimmune diabetes of adults (LADA). The categorization of glucose homeostasis is...
528