相关实验视频
Updated: Jul 26, 2025

03:25
Simple Continuous Glucose Monitoring in Freely Moving Mice
Published on: February 24, 2023
5.6K
通过数据增强模拟现实的连续葡萄糖监测时间序列
Louis A Gomez1, Adedolapo Aishat Toye1, R Stanley Hum2
1Stevens Institute of Technology, Hoboken, NJ, USA.
Journal of diabetes science and technology
|June 23, 2023
概括
研究人员开发了一种新方法来改进模拟血糖 (BG) 数据,用于1型糖尿病研究. 这种方法通过结合现实的数据缺失和错误来增强BG预测算法测试,弥合模拟和现实世界性能之间的差距.
科学领域:
- 生物医学工程 生物医学工程
- 数据科学数据科学数据科学
- 糖尿病 技术 技术
背景情况:
- 模拟数据对于对比血糖 (BG) 预测和控制算法至关重要.
- 专家创建的模型和黑子方法,如GANs提供有限的现实主义和诊断能力,用于现实世界的性能.
- 现有的模拟方法缺乏真实连续血糖监测 (CGM) 数据的复杂特征,阻碍了准确的算法评估.
研究的目的:
- 开发一种新的方法来增强模拟的BG数据,并从真实的CGM数据中获得现实的缺失和错误属性.
- 提高模拟CGM数据的准确性,以便对BG预测算法进行更严格的测试和基准测试.
- 为了减少在模拟数据和真实CGM数据上测试的算法之间的性能差距.
主要方法:
- 从真实CGM数据集中学习缺失和错误特征 (OpenAPS,OhioT1DM,RCT,种族差异).
- 增强模拟BG数据与这些学习的属性来模仿现实世界的数据挑战.
- 使用增强模拟数据对比标准模拟实践 (随机丢失,高斯噪声,CGM错误模型) 评估了BG预测性能.
主要成果:
- 拟议的方法证明了与真实数据相比的最小性能差异,与随机丢失和高斯噪声相比,对于缺失的数据和单独的错误效应.
- 结合方法在大多数数据集中显著优于高斯噪声和随机丢失,但俄俄T1DM除外.
- 开发的错误模型显著提高了各种数据集的结果,表明模拟现实性得到了改进.
结论:
- 在基于模拟数据和真实数据的BG预测之间存在显著的绩效差距.
- 拟议的方法有效地弥补了这一差距,使得性能估计更为现实.
- 研究人员现在可以严格测试算法,并获得可靠的现实世界的性能见解,而无需过度装配或广泛的数据收集.
相关概念视频
Glucose Homeostasis: Regulation of Blood Glucose
1.8K
Carbohydrates consumed through foods are converted into glucose, a crucial energy source for the body. In the prandial state, high blood glucose levels stimulate the secretion of insulin from the pancreas. Insulin inhibits hepatic glucose production and stimulates glucose uptake and metabolism by muscle and adipose tissue. The excess glucose is converted into glycogen and stored in the liver and muscles.
During fasting, when blood glucose levels are low, the pancreas secretes glucagon. it...
During fasting, when blood glucose levels are low, the pancreas secretes glucagon. it...
1.8K
Sampling Continuous Time Signal
282
In signal processing, a continuous-time signal can be sampled using an impulse-train sampling technique, followed by the zero-order hold method. Impulse-train sampling involves the use of a periodic impulse train, which consists of a series of delta functions spaced at regular intervals determined by the sampling period. When a continuous-time signal is multiplied by this impulse train, it generates impulses with amplitudes corresponding to the signal's values at the sampling points.
In the...
In the...
282

