Continuous monitoring of blood glucose (BG) levels is a key aspect of diabetes management. Patients with Type-1 diabetes (T1D) require an effective tool to monitor these levels in order to make appropriate decisions regarding insulin administration and food intake to keep BG levels in target range. Effectively and accurately predicting future BG levels at multi-time steps ahead benefits a patient with diabetes by helping them decrease the risks of extremes in BG including hypo- and hyperglycemia. In this study, we present a novel multi-component deep learning model BG-Predict that predicts the BG levels in a multi-step look ahead fashion. The model is evaluated both quantitatively and qualitatively on actual blood glucose data for 97 patients. For the prediction horizon (PH) of 30 mins, the average values for root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and normalized mean squared error (NRMSE) are [Formula: see text] mg/dL, 16.77 ± 4.87 mg/dL, [Formula: see text] and [Formula: see text] respectively. When Clarke and Parkes error grid analyses were performed comparing predicted BG with actual BG, the results showed average percentage of points in Zone A of [Formula: see text] and [Formula: see text] respectively. We offer this tool as a mechanism to enhance the predictive capabilities of algorithms for patients with T1D.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8692478 | PMC |
| http://dx.doi.org/10.1038/s41598-021-03341-5 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
On hyper-parameter selection for guaranteed convergence of RMSProp.
Cogn Neurodyn
December 2024
Department of Electrical and Electronic Engineering, Imperial College London, London, SW7 2AZ UK.
RMSProp is one of the most popular stochastic optimization algorithms in deep learning applications. However, recent work has pointed out that this method may not converge to the optimal solution even in simple convex settings. To this end, we propose a time-varying version of RMSProp to fix the non-convergence issues.
View Article and Find Full Text PDFBenchmarking a dual-scale hybrid simulation framework for small globular proteins combining the CHARMM36 and Martini2 models.
J Mol Graph Model
December 2024
CSIR-National Chemical Laboratory, Dr. Homi Bhabha Road, Pune 411008, India; Academy of Scientific and Innovative Research (AcSIR), Ghaziabad, 201 002, India. Electronic address:
Multi-scale models in which varying resolutions are considered in a single molecular dynamics simulation setup are gaining importance in integrative modeling. However, combining atomistic and coarse-grain resolutions, especially for coarse-grain force fields derived from top-down approaches, have not been well explored. In this study, we have implemented and tested a dual-resolution simulation approach to model globular proteins in atomistic detail (represented by the CHARMM36 model) with the surrounding solvent in Martini2 coarse-grain detail.
View Article and Find Full Text PDFA proposal to simplify the definition of pediatric hypertension: bridging the gap between perception and action.
BMC Med
December 2024
Department of Epidemiology/Department of Maternal, Child and Adolescent Care, School of Public Health, Qilu Hospital, Cheeloo College of Medicine, Shandong University, Jinan, 250012, China.
Background: The importance of routine hypertension screening in children and adolescents is now well recognized. However, it is often undiagnosed in clinical practice, partly due to the reliance on a complex blood pressure (BP) percentile-based table with hundreds of cutoffs by age, sex, and height.
Main Text: Many studies have explored simplified tools for screening hypertension in children and adolescents, such as simplified formulas, simplified BP tables by age and sex group, by age group, or by height group, and the BP to height ratio.
Effects of timely case conferencing between general practitioners and specialist palliative care services on symptom burden in patients with advanced chronic disease: results of the cluster-randomised controlled KOPAL trial.
BMC Palliat Care
December 2024
Department of General Practice and Primary Care, University Medical Center Hamburg-Eppendorf, Martinistr. 52, 20246, Hamburg, Germany.
Background: Patients with advanced chronic non-malignant conditions often experience significant symptom burden. Therefore, overcoming barriers to interprofessional collaboration between general practitioners (GPs) and specialist palliative home care (SPHC) teams is essential to facilitate the timely integration of palliative care elements. The KOPAL trial aimed to examine the impact of case conferences between GPs and SPHC teams on symptom burden and pain in patients with advanced chronic heart failure, chronic obstructive pulmonary disease, and dementia.
View Article and Find Full Text PDFCapillary-driven self-assembly of soft ellipsoidal microgels at the air-water interface.
Proc Natl Acad Sci U S A
December 2024
Institute of Physical Chemistry, Rheinisch-Westfälische Technische Hochschule Aachen University, Aachen 52074, Germany.
The adsorption of ellipsoidal colloidal particles on liquid interfaces induces interfacial deformation, resulting in anisotropic interface-mediated interactions and the formation of superstructures. Soft prolate-shaped microgels at the air-water interface offer an ideal model for studying spontaneous capillary-driven self-assembly due to their tunable aspect ratio, controlled functionality, and softness. These microgels consist of a polystyrene core surrounded by a cross-linked, fluorescently labeled poly([Formula: see text]-isopropylmethylacrylamide) shell.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!