Portal de Murcia

www.portaldemurcia.com

• Home
• Business
• Hotels
• Restaurants
• News
• Advertisement

The researcher from the University of Murcia (UMU) Ignacio Rodríguez Rodríguez proposes a system that enables greater precision when predicting blood glucose and thereby deciding on insulin doses for diabetic patients, resulting in the improvement of their quality of life .

This study, published in prestigious magazines such as Sensors or Journal of Diabetes Research, shows a comprehensive management system for type 1 diabetes, when the pancreas no longer produces insulin, with biosensors that allow estimating the evolution of glucose in a diabetic patient automatically.

The glucose prediction is obtained from the complete monitoring of biomedical variables, with continuous blood glucose meters, smart wristbands and the use of algorithms.

Smart algorithms and variable monitoring

Seven variables have been monitored: insulin dose, amount of carbohydrates ingested, sugar values ​​of the person during the previous hours, physical exercise, heart rate, hours of sleep and schedule of activities.

All of them have been measured in real situations, during fourteen days, to 25 volunteers with type 1 diabetes.

With the information collected, using intelligent algorithms, a glycemic prediction has been reached in the following 45 minutes with an error margin of 18.60 mg / dL.

"Through machine-learning algorithms, it is possible to predict the glucose of a diabetic patient with a very acceptable error," says Rodríguez.

The Internet of Things and health

Through this research, we have tried to exploit the concept of the Internet of Things.

It is a total interconnection with the network and all the potential that this implies in terms of data storage, cloud computing or ubiquity, as well as the extraction of patterns and prediction made possible by intelligent algorithms.

The algorithms chosen to predict glucose have been run on devices with limited resources to check their viability on portable devices, such as mobile phones, that the patient can carry.

An attempt has been made to find out to what extent the control system can operate independently with a mobile phone in the event of a loss of coverage or network failure, achieving "fairly good glucose prediction values ​​at acceptable execution times", according to the investigator.

Likewise, this work has been possible thanks to the popularization of continuous glucose meters, which allow obtaining a glucose value every five minutes.

"Smart bracelets complete the path towards a global monitoring system, because with them we continuously know other circumstances that affect the person with diabetes, such as exercise or hours of sleep," says Rodríguez.

About type 1 diabetes

Diabetes is a disease characterized by difficulties in metabolizing blood glucose.

In the specific case of type 1, said metabolization becomes impossible because the pancreas no longer produces insulin, so the person has to inject it daily.Up to 14% of the Spanish population suffers from diabetes and, of this percentage, between 1% and 5% is type 1.

An investigation with multiple thanks

The articles published on this study are part of the thesis of Ignacio Rodríguez Rodríguez (currently a researcher at the University of Malaga), whose directors are the researcher at the UMU Miguel Ángel Zamora and the researcher at the Polytechnic University of Cartagena (UPCT) José Víctor Rodríguez.

In addition, the SODICAR Association (Society of Diabetics of Cartagena) has facilitated the search for volunteers to study, as well as the Morales Meseguer Hospital in Murcia and its Endocrinology team.

In the international arena, the contributions of Professor Ioannis Chatzigiannakis of the Dpartimento di Ingegneria Informatica, Automática e Gestionale de La SapienzaUniversità di Roma (Italy), co-author of two of the publications, have been of great relevance.

References:

Rodríguez-Rodríguez, I., Zamora-Izquierdo, M. Á., & Rodríguez, JV (2018).

Towards an ICT-based platform for type 1 diabetes mellitus management.Applied Sciences, 8 (4), 511. https://doi.org/10.3390/app8040511

Rodríguez-Rodríguez, I., Rodríguez, JV, & Zamora-Izquierdo, M. Á.

(2018).

Variables to Be Monitored via Biomedical Sensors for Complete Type 1 Diabetes Mellitus Management: An Extension of the "On-Board" Concept. Journal of diabetes research, 2018.

https://doi.org/10.1155/2018/4826984

Rodríguez-Rodríguez, I., Rodríguez, JV, González-Vidal, A., & Zamora, M. Á.

(2019).

Feature Selection for Blood Glucose Level Prediction in Type 1 Diabetes Mellitus by Using the Sequential Input Selection Algorithm (SISAL) .Symmetry, 11 (9), 1164. https://doi.org/10.3390/sym11091164

Rodríguez-Rodríguez, I., Chatzigiannakis, I., Rodríguez, JV, Maranghi, M., Gentili, M., & Zamora-Izquierdo, M. Á.

(2019).

Utility of Big Data in Predicting Short-Term Blood Glucose Levels in Type 1 Diabetes Mellitus Through Machine Learning Techniques. Sensors, 19 (20), 4482.https: //doi.org/10.3390/s19204482

Rodríguez-Rodríguez, I., Rodríguez, JV, Chatzigiannakis, I., & Zamora Izquierdo, M. Á.

(2019).

On the Possibility of Predicting Glycaemia 'On the Fly'with Constrained IoT Devices in Type 1 Diabetes Mellitus Patients.Sensors, 19 (20), 4538. https://doi.org/10.3390/s19204538

Source: Universidad de Murcia