| 1. |
- Cederblad, Lars, et al.
(författare)
-
Classification of Hypoglycemic Events in Type 1 Diabetes Using Machine Learning Algorithms
- 2023
-
Ingår i: Diabetes Therapy. - : Springer Nature. - 1869-6953 .- 1869-6961. ; 14:6, s. 953-965
-
Tidskriftsartikel (refereegranskat)abstract
- IntroductionTo improve the utilization of continuous- and flash glucose monitoring (CGM/FGM) data we have tested the hypothesis that a machine learning (ML) model can be trained to identify the most likely root causes for hypoglycemic events.MethodsCGM/FGM data were collected from 449 patients with type 1 diabetes. Of the 42,120 identified hypoglycemic events, 5041 were randomly selected for classification by two clinicians. Three causes of hypoglycemia were deemed possible to interpret and later validate by insulin and carbohydrate recordings: (1) overestimated bolus (27%), (2) overcorrection of hyperglycemia (29%) and (3) excessive basal insulin presure (44%). The dataset was split into a training (n = 4026 events, 304 patients) and an internal validation dataset (n = 1015 events, 145 patients). A number of ML model architectures were applied and evaluated. A separate dataset was generated from 22 patients (13 ‘known’ and 9 ‘unknown’) with insulin and carbohydrate recordings. Hypoglycemic events from this dataset were also interpreted by five clinicians independently.ResultsOf the evaluated ML models, a purpose-built convolutional neural network (HypoCNN) performed best. Masking the time series, adding time features and using class weights improved the performance of this model, resulting in an average area under the curve (AUC) of 0.921 in the original train/test split. In the dataset validated by insulin and carbohydrate recordings (n = 435 events), i.e. ‘ground truth,’ our HypoCNN model achieved an AUC of 0.917.ConclusionsThe findings support the notion that ML models can be trained to interpret CGM/FGM data. Our HypoCNN model provides a robust and accurate method to identify root causes of hypoglycemic events.
|
|
| 2. |
- Wahlström Johnsson, Inger, 1973-, et al.
(författare)
-
High birth weight was not associated with altered body composition or impaired glucose tolerance in adulthood
- 2019
-
Ingår i: Acta Paediatrica. - : Wiley. - 0803-5253 .- 1651-2227. ; 108:12, s. 2208-2213
-
Tidskriftsartikel (refereegranskat)abstract
- Aim To investigate whether a high birth weight was associated with an increased proportion of body fat or with impaired glucose tolerance in adulthood.Methods Our cohort comprised 27 subjects with birth weights of 4,500 g or more, and 27 controls with birth weights within ±1 SDS, born at Uppsala University Hospital 1975-1979. The subjects were 34-40 years old at the time of study.Anthropometric data was collected, and data on body composition was obtained by air plethysmography and bioimpedance and was estimated with a three compartment model. Indirect calorimetry, blood sampling for fasting insulin and glucose as well as a 75 g oral glucose tolerance test were also performed. Insulin sensitivity was assessed using homeostasis model assessment 2 (HOMA2) and Matsuda index. Areas under the curves were calculated for insulin and glucose.Results There were no differences in body mass index, body composition or insulin sensitivity between subjects with a high birth weight and controls.Conclusion Adult subjects, born with a moderately high birth weight, did not differ from those with birth weights within ±1 SDS regarding body composition or glucose tolerance
|
|
