Department of Applied Mathematics and Computer Science, DTU Compute, Kgs. Lyngby, Denmark.
Novo Nordisk A/S, Device R&D, Hillerød, Denmark.
J Diabetes Sci Technol. 2021 Jan;15(1):98-108. doi: 10.1177/1932296820912411. Epub 2020 Apr 16.
Lack of treatment adherence can lead to life-threatening health complications for people with type 2 diabetes (T2D). Recent improvements and availability in continuous glucose monitoring (CGM) technology have enabled various possibilities to monitor diabetes treatment. Detection of missed once-daily basal insulin injections can be used to provide feedback to patients, thus improving their diabetes management. In this study, we explore how (ML) based on CGM data can be used for detecting to once-daily basal insulin injections.
In-silico CGM data were generated to simulate a cohort of T2D patients on once-daily insulin injection (Tresiba®). methods within ML based on automatic feature extraction including convolutional neural networks were explored and compared with simple feature-engineered ML classification models for adherence detection. It was further investigated whether fused expert-dependent and automatically learned features could improve performance, resulting in a comparison of six different detection models. Adherence was detected throughout each day with an increasing amount of CGM data available.
The adherence detection accuracy improved as more CGM data became available on the day of classification. The three classification models based on expert-engineered features obtained mean accuracies of 78.6%, 78.2%, and 78.3%. The classification model based purely on learned features obtained a mean accuracy of 79.7%. The two classification models fusing expert-engineered and learned features obtained mean accuracies of 79.7% and 79.8%. All the mentioned results were obtained 16 hours after time of injection.
The results suggest that adherence detection based on CGM data is feasible. Even though our study based on in-silico data indicates only slightly improved performance of more complex models, the question remains whether advanced models would outperform the simple in a real-world setting. Thus, future studies on adherence monitoring using real CGM data are relevant.
治疗依从性差可能导致 2 型糖尿病（T2D）患者出现危及生命的健康并发症。连续血糖监测（CGM）技术的最新改进和可用性使各种监测糖尿病治疗的可能性成为可能。检测漏用一次每日基础胰岛素注射可以为患者提供反馈，从而改善他们的糖尿病管理。在这项研究中，我们探索了如何使用基于 CGM 数据的机器学习（ML）来检测一次每日基础胰岛素注射的漏用情况。
为了模拟一次每日胰岛素注射（Tresiba®）的 T2D 患者队列，生成了基于计算机的 CGM 数据。探索了基于自动特征提取的 ML 中的分类方法，包括卷积神经网络，并将其与基于简单特征工程的 ML 分类模型进行了比较，以进行依从性检测。进一步研究了融合专家依赖和自动学习的特征是否可以提高性能，从而比较了六个不同的检测模型。在每天的每个时间点，随着 CGM 数据的增加，检测依从性。
随着分类日当天 CGM 数据的可用性增加，依从性检测的准确性也随之提高。基于专家设计特征的三个分类模型的平均准确率分别为 78.6%、78.2%和 78.3%。基于纯粹学习特征的分类模型的平均准确率为 79.7%。融合专家设计特征和学习特征的两个分类模型的平均准确率分别为 79.7%和 79.8%。所有提到的结果都是在注射后 16 小时获得的。
结果表明，基于 CGM 数据的依从性检测是可行的。尽管我们的研究基于计算机数据仅表明更复杂的模型的性能略有提高，但仍存在一个问题，即高级模型在实际环境中是否会优于简单模型。因此，使用真实 CGM 数据进行依从性监测的未来研究是相关的。
