• Je něco špatně v tomto záznamu ?

Ensemble methods in combination with compartment models for blood glucose level prediction in type 1 diabetes mellitus

K. Saiti, M. Macaš, L. Lhotská, K. Štechová, P. Pithová

. 2020 ; 196 (-) : 105628. [pub] 20200627

Jazyk angličtina Země Irsko

Typ dokumentu časopisecké články

Perzistentní odkaz   https://www.medvik.cz/link/bmc21019866

Backgroung: Type 1 diabetes is a disease that adversely affects the daily life of a large percentage of people worldwide. Daily glucose levels regulation and useful advices provided to patients regarding their diet are essential for diabetes treatment. For this reason, the interest of the academic community has focused on developing innovative systems, such as decision support systems, based on glucose prediction algorithms. The present work presents the predictive capabilities of ensemble methods compared to individual algorithms while combining each method with compartment models for fast acting insulin absorption simulation. Methods: An approach of combining widely used glycemia prediction algorithms is proposed and three different ensemble methods (Linear, Bagging and Boosting metaregressor) are applied and evaluated on their ability to provide accurate predictions for 30, 45 and 60 minutes ahead prediction horizon. Moreover, glycemia levels, long and short acting insulin dosages and consumed carbohydrates from six type one people with diabetes are used as input data and the results are evaluated in terms of root-mean square error and Clarke error grid analysis. Results: According to results, ensemble methods can provide more accurate glucose concentration in comparison to individual algorithms. Bagging metaregressor, specifically, performed better than individual algorithms in all prediction horizons for small datasets. Bagging ensemble method improved the percentage in zone A according to Clarkes error grid analysis by 4% and in some cases by 9%. Moreover, compartment models are proved to improve results in combination with any method at any prediction horizon. This strengthen the potential practical usefulness of the ensemble methods and the importance of building accurate compartment models.

Citace poskytuje Crossref.org

000      
00000naa a2200000 a 4500
001      
bmc21019866
003      
CZ-PrNML
005      
20210830101450.0
007      
ta
008      
210728s2020 ie f 000 0|eng||
009      
AR
024    7_
$a 10.1016/j.cmpb.2020.105628 $2 doi
035    __
$a (PubMed)32640369
040    __
$a ABA008 $b cze $d ABA008 $e AACR2
041    0_
$a eng
044    __
$a ie
100    1_
$a Saiti, Kyriaki $u Department of Cybernetics, Czech Technical University in Prague, Czech Republic. Electronic address: saitikyr@fel.cvut.cz
245    10
$a Ensemble methods in combination with compartment models for blood glucose level prediction in type 1 diabetes mellitus / $c K. Saiti, M. Macaš, L. Lhotská, K. Štechová, P. Pithová
520    9_
$a Backgroung: Type 1 diabetes is a disease that adversely affects the daily life of a large percentage of people worldwide. Daily glucose levels regulation and useful advices provided to patients regarding their diet are essential for diabetes treatment. For this reason, the interest of the academic community has focused on developing innovative systems, such as decision support systems, based on glucose prediction algorithms. The present work presents the predictive capabilities of ensemble methods compared to individual algorithms while combining each method with compartment models for fast acting insulin absorption simulation. Methods: An approach of combining widely used glycemia prediction algorithms is proposed and three different ensemble methods (Linear, Bagging and Boosting metaregressor) are applied and evaluated on their ability to provide accurate predictions for 30, 45 and 60 minutes ahead prediction horizon. Moreover, glycemia levels, long and short acting insulin dosages and consumed carbohydrates from six type one people with diabetes are used as input data and the results are evaluated in terms of root-mean square error and Clarke error grid analysis. Results: According to results, ensemble methods can provide more accurate glucose concentration in comparison to individual algorithms. Bagging metaregressor, specifically, performed better than individual algorithms in all prediction horizons for small datasets. Bagging ensemble method improved the percentage in zone A according to Clarkes error grid analysis by 4% and in some cases by 9%. Moreover, compartment models are proved to improve results in combination with any method at any prediction horizon. This strengthen the potential practical usefulness of the ensemble methods and the importance of building accurate compartment models.
650    _2
$a algoritmy $7 D000465
650    _2
$a krevní glukóza $7 D001786
650    _2
$a selfmonitoring glykemie $7 D015190
650    12
$a diabetes mellitus 1. typu $x farmakoterapie $7 D003922
650    _2
$a lidé $7 D006801
650    _2
$a inzulin $7 D007328
655    _2
$a časopisecké články $7 D016428
700    1_
$a Macaš, Martin $u Czech Institute of Informatics, Robotics and Cybernetics, Czech Technical University in Prague, Czech Republic
700    1_
$a Lhotská, Lenka $u Czech Institute of Informatics, Robotics and Cybernetics, Czech Technical University in Prague, Czech Republic
700    1_
$a Štechová, Kateřina $u FN Motol University Hospital, Prague, Czech Republic
700    1_
$a Pithová, Pavlína $u FN Motol University Hospital, Prague, Czech Republic
773    0_
$w MED00001214 $t Computer methods and programs in biomedicine $x 1872-7565 $g Roč. 196, č. - (2020), s. 105628
856    41
$u https://pubmed.ncbi.nlm.nih.gov/32640369 $y Pubmed
910    __
$a ABA008 $b sig $c sign $y p $z 0
990    __
$a 20210728 $b ABA008
991    __
$a 20210830101450 $b ABA008
999    __
$a ok $b bmc $g 1690631 $s 1140312
BAS    __
$a 3
BAS    __
$a PreBMC
BMC    __
$a 2020 $b 196 $c - $d 105628 $e 20200627 $i 1872-7565 $m Computer methods and programs in biomedicine $n Comput Methods Programs Biomed $x MED00001214
LZP    __
$a Pubmed-20210728

Najít záznam

Citační ukazatele

Možnosti archivace