A Deep Learning Framework for Automatic Meal Detection and Estimation in Artificial Pancreas Systems

Daniels, John; Herrero, Pau; Georgiou, Pantelis

doi:10.3390/s22020466

Cited by 21 publications

(12 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several approaches have been published for meal detection 8,9,[17][18][19] using data collected in clinical studies. The amount of missing information and outliers are lower in this case.…”

Section: Discussion Of Resultsmentioning

confidence: 99%

Detection of Meals and Physical Activity Events From Free-Living Data of People With Diabetes

Rashid

Sun

Sevil

et al. 2022

J Diabetes Sci Technol

View full text Add to dashboard Cite

Background: Predicting carbohydrate intake and physical activity in people with diabetes is crucial for improving blood glucose concentration regulation. Patterns of individual behavior can be detected from historical free-living data to predict meal and exercise times. Data collected in free-living may have missing values and forgotten manual entries. While machine learning (ML) can capture meal and exercise times, missing values, noise, and errors in data can reduce the accuracy of ML algorithms. Methods: Two recurrent neural networks (RNNs) are developed with original and imputed data sets to assess detection accuracy of meal and exercise events. Continuous glucose monitoring (CGM) data, insulin infused from pump data, and manual meal and exercise entries from free-living data are used to predict meals, exercise, and their concurrent occurrence. They contain missing values of various lengths in time, noise, and outliers. Results: The accuracy of RNN models range from 89.9% to 95.7% for identifying the state of event (meal, exercise, both, or neither) for various users. “No meal or exercise” state is determined with 94.58% accuracy by using the best RNN (long short-term memory [LSTM] with 1D Convolution). Detection accuracy with this RNN is 98.05% for meals, 93.42% for exercise, and 55.56% for concurrent meal-exercise events. Conclusions: The meal and exercise times detected by the RNN models can be used to warn people for entering meal and exercise information to hybrid closed-loop automated insulin delivery systems. Reliable accuracy for event detection necessitates powerful ML and large data sets. The use of additional sensors and algorithms for detecting these events and their characteristics provides a more accurate alternative.

show abstract

“…Several approaches have been published for meal detection 8,9,[17][18][19] using data collected in clinical studies. The amount of missing information and outliers are lower in this case.…”

Section: Discussion Of Resultsmentioning

confidence: 99%

Detection of Meals and Physical Activity Events From Free-Living Data of People With Diabetes

Rashid

Sun

Sevil

et al. 2022

J Diabetes Sci Technol

View full text Add to dashboard Cite

show abstract

Section: Hypoglycaemia Unawarenessmentioning

confidence: 99%

“…With the help of more recent technologies, algorithms for meal detection or estimation of undisclosed meals based on multitask quantile regression and neural networks will be developed, helping to further lower the danger of hypoglycemia. 55 In the United Kingdom, the National Institute for Health and Care Excellence (NICE) has recommended all patients with T1DM should be provided with CGM technology to assist their care (either real time or intermittent). 56 This includes all adults and children above 4 years old with T1DM.…”

Section: Management Of Hypoglycaemia In Diabetic Patients Preventionmentioning

confidence: 99%

Managing Severe Hypoglycaemia in Patients with Diabetes: Current Challenges and Emerging Therapies

Demirbilek

Vurallı

Haris

et al. 2023

DMSO

View full text Add to dashboard Cite

Hypoglycaemia is common in patients with diabetes mellitus and is a limiting factor for achieving adequate glycaemic control. In the vast majority of cases, hypoglycaemia develops due to the imbalance between food intake and insulin injections. As recurrent hypoglycaemia leads to significant morbidity and mortality, the recognition and immediate treatment of hypoglycaemia in diabetic patients is thus important. In the last 20 years, the introduction of improved insulin analogues, insulin pump therapy, continuous glucose monitoring (CGM), and sensor-augmented pump therapy have all made significant improvements in helping to reduce and prevent hypoglycaemia. In terms of treatment, the American Diabetes Association recommends oral glucose as the first-line treatment option for all conscious patients with hypoglycaemia. The second line of treatment (or first line in unconscious patients) is the use of glucagon. Novel formulations of glucagon include the nasal form, the Gvoke HypoPen which is a ready-to-deliver auto-injector packaged formulation and finally a glucagon analogue, Dasiglucagon. The Dasiglucagon formulation has recently been approved for the treatment of severe hypoglycaemia. It is a ready-to-use, similar to endogenous glucagon and its potency is also the same as native glucagon. It does not require reconstitution before injection and therefore ensures better compliance. Thus, significant improvements including development of newer insulin analogues, insulin pump therapy, continuous glucose monitoring (CGM), sensor-augmented pump therapy and novel formulations of glucagon have all contributed to reducing and preventing hypoglycaemia in diabetic individuals. However, considerable challenges remain as not all patients have access to diabetes technologies and to the newer glucagon formulations to help reduce and prevent hypoglycaemia.

show abstract

“…They are then classified as 2 categories, true or false [ 38 ]. One author [ 49 ] proposes an algorithm based on a deep learning for multitask quantile regression by using a sequence-sequence (seq2seq) encoder- decoder LSTM for predicting the last 20 minutes of glucose using historical CGM measures, meals, and insulin [ Figure 4 ].…”

Section: Meal Detection and Estimation Techniquesmentioning

confidence: 99%

Artificial Pancreas: A Review of Meal Detection and Carbohydrates Counting Techniques

Rodriguez

Villamizar

2022

Review of Diabetic Studies

View full text Add to dashboard Cite

OBJECTIVE: The development of an artificial pancreas is an open research problem that faces the challenge of creating a control algorithm capable of dosing insulin automatically and driving blood glucose to healthy levels. Many of these approaches, including artificial intelligence, are based on techniques that could result in and undesirable outcome because most of them include neither detect meal intake or meal size information. To overcome that issue, some meal count-detection algorithms reported in scientific publications have shown not only a good performance on blood glucose regulation but fewer hypoglicemia and hyperglycemia events too. METHODS: We reviewed the most relevant authors and publications and main databases (particularly SCOPUS and Google Scholar), focusing on algorithms of detection and estimation of meal intake from multiple approaches. RESULTS: A wide range of approaches and proposals have been found. The majority of them include trials on in silico patients rather than in vivo ones. Most of procedures require as inputs glucose samples from continuous glucose monitoring devices as basal insulin and bolus as well. Most of approaches could be grouped by 2 categories: mathematical model based and not model based. CONCLUSION: A combination of methods seems to reach better results.

show abstract

A Deep Learning Framework for Automatic Meal Detection and Estimation in Artificial Pancreas Systems

Cited by 21 publications

References 33 publications

Detection of Meals and Physical Activity Events From Free-Living Data of People With Diabetes

Detection of Meals and Physical Activity Events From Free-Living Data of People With Diabetes

Managing Severe Hypoglycaemia in Patients with Diabetes: Current Challenges and Emerging Therapies

Artificial Pancreas: A Review of Meal Detection and Carbohydrates Counting Techniques

Contact Info

Product

Resources

About