Using Iterative Learning for Insulin Dosage Optimization in Multiple-Daily-Injections Therapy for People With Type 1 Diabetes

Cescon, Marzia; Deshpande, Sunil; Nimri, Revital; Doyle, Francis J.; Dassau, Eyal

doi:10.1109/tbme.2020.3005622

Cited by 9 publications

(13 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This baseline algorithm is taken from recent work from Cescon et al. where an iterative learning control (ILC) is proposed to optimize the basal insulin dose ( 16 ). This ILC algorithm finds the new optimal dose

to be administered such that the SMBG values are driven as close as possible to the desired reference trajectory G target :…”

Section: Methodsmentioning

confidence: 99%

“…where g is a gain set to 1, G target is set to 110 mg/dL, and F(q) is a discrete filter described in (16). Similarly to (16), the DC gain of the filter F(q) is individualized to each subject as 3 BW .…”

Section: Control To Referencementioning

confidence: 99%

“…Because of changes in people's lifestyle and metabolism, this dose is periodically titrated to achieve the desired glycemic goals (1). The standard-of-care clinical approach to basal insulin titration involves adjusting the insulin dosage by few units based on self-monitored blood glucose (SMBG) data measured daily in fasting conditions (i.e., pre-breakfast) (15)(16)(17)(18). Continuous glucose monitoring (CGM) devices offer the possibility to record glucose levels almost continuously (e.g., 5-minute sampling time), for the whole day.…”

Section: Introductionmentioning

confidence: 99%

“…Not requiring carbohydrate records renders the method suitable to serve people using MDI therapy, since most do not perform precise carbohydrate counting (24). The proposed method is shown to outperform traditional SMBGbased approaches for the titration of basal insulin (16), in two insilico experiments performed within the University of Virginia/ Padova T1D Simulator (25)(26)(27). Further, the robustness of the approach to missing carbohydrate records is assessed by comparing the changes to insulin dose suggested by the algorithm with and without carbohydrate information.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Titration of Long-Acting Insulin Using Continuous Glucose Monitoring and Smart Insulin Pens in Type 1 Diabetes: A Model-Based Carbohydrate-Free Approach

2022

View full text Add to dashboard Cite

ObjectiveMultiple daily injections (MDI) therapy is the most common treatment for type 1 diabetes (T1D), consisting of long-acting insulin to cover fasting conditions and rapid-acting insulin to cover meals. Titration of long-acting insulin is needed to achieve satisfactory glycemia but is challenging due to inter-and intra-individual metabolic variability. In this work, a novel titration algorithm for long-acting insulin leveraging continuous glucose monitoring (CGM) and smart insulin pens (SIP) data is proposed.MethodsThe algorithm is based on a glucoregulatory model that describes insulin and meal effects on blood glucose fluctuations. The model is individualized on patient’s data and used to extract the theoretical glucose curve in fasting conditions; the individualization step does not require any carbohydrate records. A cost function is employed to search for the optimal long-acting insulin dose to achieve the desired glycemic target in the fasting state. The algorithm was tested in two virtual studies performed within a validated T1D simulation platform, deploying different levels of metabolic variability (nominal and variance). The performance of the method was compared to that achieved with two published titration algorithms based on self-measured blood glucose (SMBG) records. The sensitivity of the algorithm to carbohydrate records was also analyzed.ResultsThe proposed method outperformed SMBG-based methods in terms of reduction of exposure to hypoglycemia, especially during the night period (0 am–6 am). In the variance scenario, during the night, an improvement in the time in the target glycemic range (70–180 mg/dL) from 69.0% to 86.4% and a decrease in the time in hypoglycemia (<70 mg/dL) from 10.7% to 2.6% was observed. Robustness analysis showed that the method performance is non-sensitive to carbohydrate records.ConclusionThe use of CGM and SIP in people with T1D using MDI therapy has the potential to inform smart insulin titration algorithms that improve glycemic control. Clinical studies in real-world settings are warranted to further test the proposed titration algorithm.SignificanceThis algorithm is a step towards a decision support system that improves glycemic control and potentially the quality of life, in a population of individuals with T1D who cannot benefit from the artificial pancreas system.

show abstract

to be administered such that the SMBG values are driven as close as possible to the desired reference trajectory G target :…”

Section: Methodsmentioning

confidence: 99%

Section: Control To Referencementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Titration of Long-Acting Insulin Using Continuous Glucose Monitoring and Smart Insulin Pens in Type 1 Diabetes: A Model-Based Carbohydrate-Free Approach

2022

View full text Add to dashboard Cite

show abstract

“…Iterative learning control (ILC) is an effective control methodology that can enhance the tracking accuracy of dynamical systems by learning from previous iterations data [1,2], and has been applied to many applications like biomedical engineering [3,4], batch processes [5][6][7], robotic systems [8,9], urban traffic control [10,11], and so forth [12][13][14][15][16]. A more detailed discussion about various ILC designs in the continuous-or discrete-time domain can be found in [1,17].…”

Section: Introductionmentioning

confidence: 99%

Robust convergence conditions of iterative learning control for time‐delay systems under random non‐repetitive uncertainties

Shokri-Ghaleh

Ganjefar

Shahri

2022

IET Control Theory & Appl

View full text Add to dashboard Cite

Some of the most fundamental assumptions in the design of iterative learning control (ILC) for uncertain systems are the strict repetitiveness (i.e. iteration-independence) of uncertainties in all factors of plant dynamic, external disturbances, initial conditions, and reference trajectory, which may not always hold in practical applications. The simultaneous relaxation of all these assumptions is a challenging problem that has been addressed only for delay-free systems. This problem is still open for time-delay systems, especially when the time-delay factor also has non-repetitive (i.e. iteration-dependent/varying) uncertainty. Hence, this study extends the problem of robust ILC for a class of time-delay systems under nonrepetitive uncertainties in not only plant dynamic, external disturbances, initial conditions, and reference trajectory but also time-delay. By using the ILC scheme introduced in this work, and based on the frequency domain analysis, it is shown that both monotonic convergence and boundedness of the expected tracking error can be achieved (in the sense of L 2 -norm) when the non-repetitiveness (i.e. iteration-dependence) of all existing uncertainties from a random viewpoint are taken into account. The effectiveness of the proposed strategy is verified by two simulation examples.

show abstract

Data-enabled learning and control algorithms for intelligent glucose management: The state of the art

Cai

Cescon

et al. 2023

Annual Reviews in Control

View full text Add to dashboard Cite

Using Iterative Learning for Insulin Dosage Optimization in Multiple-Daily-Injections Therapy for People With Type 1 Diabetes

Cited by 9 publications

References 28 publications

Titration of Long-Acting Insulin Using Continuous Glucose Monitoring and Smart Insulin Pens in Type 1 Diabetes: A Model-Based Carbohydrate-Free Approach

Titration of Long-Acting Insulin Using Continuous Glucose Monitoring and Smart Insulin Pens in Type 1 Diabetes: A Model-Based Carbohydrate-Free Approach

Robust convergence conditions of iterative learning control for time‐delay systems under random non‐repetitive uncertainties

Data-enabled learning and control algorithms for intelligent glucose management: The state of the art

Contact Info

Product

Resources

About