使用电子健康记录进行深度度度度学习的2型糖尿病机会性查
Qixuan Jin1, Haoran Zhang2, Lukasz Szczerbinski3,4,5,6,7
1Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA. qixuanj@mit.edu.
Scientific reports
|November 25, 2025
概括
本研究引入了一种深度度度度学习模型,用于统一的2型糖尿病 (T2D) 发病预测和使用电子健康记录进行亚型化. 该模型准确地预测T2D发病,并识别出不同的患者亚型以进行个性化干预.
科学领域:
- 人工智能的人工智能
- 医疗信息学 医疗信息学
- 内分泌学 在内分泌学.
背景情况:
- 电子健康记录 (EHR) 为2型糖尿病 (T2D) 的机会性查提供了潜力.
- 目前用于T2D发病预测和亚型的方法通常是分开的,这限制了它们的临床应用.
- 准确的预测和分类对于开发有针对性的干预措施至关重要.
研究的目的:
- 引入一种新的深度度度度学习 (DML) 模型,统一T2D发病预测和亚型化.
- 评估DML模型的性能与T2D发病预测的现有方法相比.
- 识别不同的T2D亚型并分析它们的临床特征和治疗反应.
主要方法:
- 开发了一个深度度度度学习 (DML) 模型,以基于样本相似性来学习潜在空间,统一发病预测和子类型.
- 利用来自我们所有人计划 (n=7567) 和马萨诸塞州总医院布里格姆生物库 (n=3298) 的EHR数据进行模型培训和验证.
- 与DML模型的发病预测性能 (AUC=0.754) 与后勤回归,临床风险因素和血糖指标进行比较.
主要成果:
- 该DML模型实现了0.754的AUC预测T2D发病前7年,超过了基线方法.
- 确定了三种不同的T2D亚型,其特点是不同的并发症 (肥胖,心血管,心理健康).
- 具有较少并发症的亚型表现出较早的甲福林启动和更大的HbA1c降低,表明不同的治疗反应.
结论:
- 统一的DML模型在T2D发病预测方面表现出卓越的性能,并使强大的亚型化成为可能.
- 已识别的T2D亚型具有不同的临床特征和对治疗的不同反应,这表明个性化医学的潜力.
- 该模型在多样化的队列中的可转移性突显了其在T2D管理中广泛临床实施的潜力.
相关概念视频
01:22Diabetes Mellitus: Type 2 and Gestational
4.3K
Type 2 diabetes, characterized by insulin resistance, arises when the insulin receptors on cells lose responsiveness to insulin, diminishing the cell's capacity to take up glucose, resulting in elevated blood glucose levels. To receive a diagnosis of Type 2 diabetes, a series of blood glucose tests are necessary to assess whether the blood glucose falls within normal parameters. If the result is out of the normal range, a patient may be diagnosed as prediabetic or diabetic, depending on the...
4.3K
01:15Diabetes: Symptoms, Diagnosis, and Complications
2.0K
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...
2.0K
01:36Carbohydrate Metabolism
13.7K
Carbohydrates are polymers composed of molecules containing atoms of carbon, hydrogen and oxygen. One gram of carbohydrate can provide four kilo-calories of energy, which makes it the most efficient instant energy source.
Starch accounts for approximately 60% of the carbohydrates consumed by humans. Since amylase enzymes cannot function in the stomach's acidic environment, starch can only be digested in the mouth and small intestine. Simple sugars are found naturally in milk and fruits in...
Starch accounts for approximately 60% of the carbohydrates consumed by humans. Since amylase enzymes cannot function in the stomach's acidic environment, starch can only be digested in the mouth and small intestine. Simple sugars are found naturally in milk and fruits in...
13.7K