Modelling endogenous insulin concentration in type 2 diabetes during closed-loop insulin delivery

A Glucose-Only Model to Extract Physiological Information from Postprandial Glucose Profiles in Subjects with Normal Glucose Tolerance

BACKGROUND

Current mathematical models of postprandial glucose metabolism in people with normal and impaired glucose tolerance rely on insulin measurements and are therefore not applicable in clinical practice. This research aims to develop a model that only requires glucose data for parameter estimation while also providing useful information on insulin sensitivity, insulin dynamics and the meal-related glucose appearance (GA).

METHODS

The proposed glucose-only model (GOM) is based on the oral minimal model (OMM) of glucose dynamics and substitutes the insulin dynamics with a novel function dependant on glucose levels and GA. A Bayesian method and glucose data from 22 subjects with normal glucose tolerance are utilised for parameter estimation. To validate the results of the GOM, a comparison to the results of the OMM, obtained by using glucose and insulin data from the same subjects is carried out.

RESULTS

The proposed GOM describes the glucose dynamics with comparable precision to the OMM with an RMSE of 5.1 ± 2.3 mg/dL and 5.3 ± 2.4 mg/dL, respectively and contains a parameter that is significantly correlated to the insulin sensitivity estimated by the OMM (r = 0.7) Furthermore, the dynamic properties of the time profiles of GA and insulin dynamics inferred by the GOM show high similarity to the corresponding results of the OMM.

CONCLUSIONS

The proposed GOM can be used to extract useful physiological information on glucose metabolism in subjects with normal glucose tolerance. The model can be further developed for clinical applications to patients with impaired glucose tolerance under the use of continuous glucose monitoring data.

Machine Learning-Based Adherence Detection of Type 2 Diabetes Patients on Once-Daily Basal Insulin Injections

BACKGROUND

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 machine learning (ML) based on CGM data can be used for detecting adherence to once-daily basal insulin injections.

METHODS

In-silico CGM data were generated to simulate a cohort of T2D patients on once-daily insulin injection (Tresiba®). Deep learning 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.

RESULTS

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.

CONCLUSION

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.

Feature extraction using traditional image processing and convolutional neural network methods to classify white blood cells: a study

White blood cells play a vital role in monitoring health condition of a person. Change in count and/or appearance of these cells indicate hematological disorders. Manual microscopic evaluation of white blood cells is the gold standard method, but the result depends on skill and experience of the hematologist. In this paper we present a comparative study of feature extraction using two approaches for classification of white blood cells. In the first approach, features were extracted using traditional image processing method and in the second approach we employed AlexNet which is a pre-trained convolutional neural network as feature generator. We used neural network for classification of WBCs. The results demonstrate that, classification result is slightly better for the features extracted using the convolutional neural network approach compared to traditional image processing approach. The average accuracy and sensitivity of 99% was obtained for classification of white blood cells. Hence, any one of these methods can be used for classification of WBCs depending availability of data and required resources.

A deep learning approach to adherence detection for type 2 diabetics

Diabetes has become one of the biggest health problems in the world. In this context, adherence to insulin treatment is essential in order to avoid life-threatening complications. In this pilot study, a novel adherence detection algorithm using Deep Learning (DL) approaches was developed for type 2 diabetes (T2D) patients, based on simulated Continuous Glucose Monitoring (CGM) signals. A large and diverse amount of CGM signals were simulated for T2D patients using a T2D adapted version of the Medtronic Virtual Patient (MVP) model for T1D. By using these signals, different classification algorithms were compared using a comprehensive grid search. We contrast a standard logistic regression baseline to Multi- Layer Perceptrons (MLPs) and Convolutional Neural Networks (CNNs). The best classification performance with an average accuracy of 77:5% was achieved with CNN. Hence, this indicates the potential of DL, when considering adherence detection systems for T2D patients.

A critical review and analysis of ethical issues associated with the artificial pancreas

The artificial pancreas combines a hormone infusion pump with a continuous glucose monitoring device, supported by a dosing algorithm currently installed on the pump. It allows for dynamic infusions of insulin (and possibly other hormones such as glucagon) tailored to patient needs. For patients with type 1 diabetes the artificial pancreas has been shown to prevent more effectively hypoglycaemic events and hyperglycaemia than insulin pump therapy and has the potential to simplify care. However, the potential ethical issues associated with the upcoming integration of the artificial pancreas into clinical practice have not yet been discussed. Our objective was to identify and articulate ethical issues associated with artificial pancreas use for patients, healthcare professionals, industry and policymakers. We performed a literature review to identify clinical, psychosocial and technical issues raised by the artificial pancreas and subsequently analysed them through a common bioethics framework. We identified five sensitive domains of ethical issues. Patient confidentiality and safety can be jeopardized by the artificial pancreas' vulnerability to security breaches or unauthorized data sharing. Public and private coverage of the artificial pancreas could be cost-effective and warranted. Patient selection criteria need to ensure equitable access and sensitivity to patient-reported outcomes. Patient coaching and support by healthcare professionals or industry representatives could help foster realistic expectations in patients. Finally, the artificial pancreas increases the visibility of diabetes and could generate issues related to personal identity and patient agency. The timely consideration of these issues will optimize the technological development and clinical uptake of the artificial pancreas.

A data driven nonlinear stochastic model for blood glucose dynamics

The development of adequate mathematical models for blood glucose dynamics may improve early diagnosis and control of diabetes mellitus (DM). We have developed a stochastic nonlinear second order differential equation to describe the response of blood glucose concentration to food intake using continuous glucose monitoring (CGM) data. A variational Bayesian learning scheme was applied to define the number and values of the system's parameters by iterative optimisation of free energy. The model has the minimal order and number of parameters to successfully describe blood glucose dynamics in people with and without DM. The model accounts for the nonlinearity and stochasticity of the underlying glucose-insulin dynamic process. Being data-driven, it takes full advantage of available CGM data and, at the same time, reflects the intrinsic characteristics of the glucose-insulin system without detailed knowledge of the physiological mechanisms. We have shown that the dynamics of some postprandial blood glucose excursions can be described by a reduced (linear) model, previously seen in the literature. A comprehensive analysis demonstrates that deterministic system parameters belong to different ranges for diabetes and controls. Implications for clinical practice are discussed. This is the first study introducing a continuous data-driven nonlinear stochastic model capable of describing both DM and non-DM profiles.