Department of Information Engineering, University of Padua, Padua, Italy.
Harvard John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA.
J Diabetes Sci Technol. 2022 May;16(3):641-648. doi: 10.1177/1932296821997854. Epub 2021 Mar 9.
Personal insulin pumps have shown to be effective in improving the quality of therapy for people with type 1 diabetes (T1D). However, the safety of this technology is limited by the possible infusion site failures, which are linked with hyperglycemia and ketoacidosis. Thanks to the large availability of collected data provided by modern therapeutic technologies, machine learning algorithms have the potential to provide new way to identify failures early and avert adverse events.
A clinical dataset ( = 20) is used to evaluate a novel method for detecting real-time infusion site failures using unsupervised anomaly detection algorithms, previously proposed and developed on in-silico data. An adapted feature engineering procedure is introduced to make the method able to operate in the absence of a closed-loop (CL) system and meal announcements.
In the optimal configuration, we obtained a performance of 0.75 Sensitivity (15 out of 20 total failures detected) and 0.08 FP/day, outperforming previously proposed literature algorithms. The algorithm was able to anticipate the replacement of the malfunctioning infusion sets by ~2 h on average.
On the considered dataset, the proposed algorithm showed the potential to improve the safety of patients treated with sensor-augmented pump systems.
个人胰岛素泵已被证明能有效提高 1 型糖尿病 (T1D) 患者的治疗质量。然而，这项技术的安全性受到输注部位故障的限制，这些故障与高血糖和酮症酸中毒有关。由于现代治疗技术提供了大量可收集的数据，机器学习算法有可能提供新的方法来尽早发现故障并避免不良事件。
使用临床数据集（= 20）评估一种使用无监督异常检测算法实时检测输注部位故障的新方法，该方法先前是基于模拟数据提出和开发的。引入了一种经过改进的特征工程过程，使该方法能够在没有闭环 (CL) 系统和膳食通知的情况下运行。
在最佳配置下，我们获得了 0.75 的灵敏度（总共检测到 15 次故障中的 15 次）和 0.08 的 FP/天，优于先前提出的文献算法。该算法平均能够提前约 2 小时预测到出现故障的输注套件的更换。
在考虑的数据集上，所提出的算法显示出有潜力提高接受传感器增强型泵系统治疗的患者的安全性。
