• Je něco špatně v tomto záznamu ?

Predicting glucose level with an adapted branch predictor

T. Koutny, M. Mayo

. 2022 ; 145 (-) : 105388. [pub] 20220319

Jazyk angličtina Země Spojené státy americké

Typ dokumentu časopisecké články, práce podpořená grantem

Perzistentní odkaz   https://www.medvik.cz/link/bmc22018129

BACKGROUND AND OBJECTIVE: Diabetes mellitus manifests as prolonged elevated blood glucose levels resulting from impaired insulin production. Such high glucose levels over a long period of time damage multiple internal organs. To mitigate this condition, researchers and engineers have developed the closed loop artificial pancreas consisting of a continuous glucose monitor and an insulin pump connected via a microcontroller or smartphone. A problem, however, is how to accurately predict short term future glucose levels in order to exert efficient glucose-level control. Much work in the literature focuses on least prediction error as a key metric and therefore pursues complex prediction methods such a deep learning. Such an approach neglects other important and significant design issues such as method complexity (impacting interpretability and safety), hardware requirements for low-power devices such as the insulin pump, the required amount of input data for training (potentially rendering the method infeasible for new patients), and the fact that very small improvements in accuracy may not have significant clinical benefit. METHODS: We propose a novel low-complexity, explainable blood glucose prediction method derived from the Intel P6 branch predictor algorithm. We use Meta-Differential Evolution to determine predictor parameters on training data splits of the benchmark datasets we use. A comparison is made between our new algorithm and a state-of-the-art deep-learning method for blood glucose level prediction. RESULTS: To evaluate the new method, the Blood Glucose Level Prediction Challenge benchmark dataset is utilised. On the official test data split after training, the state-of-the-art deep learning method predicted glucose levels 30 min ahead of current time with 96.3% of predicted glucose levels having relative error less than 30% (which is equivalent to the safe zone of the Surveillance Error Grid). Our simpler, interpretable approach prolonged the prediction horizon by another 5 min with 95.8% of predicted glucose levels of all patients having relative error less than 30%. CONCLUSIONS: When considering predictive performance as assessed using the Blood Glucose Level Prediction Challenge benchmark dataset and Surveillance Error Grid metrics, we found that the new algorithm delivered comparable predictive accuracy performance, while operating only on the glucose-level signal with considerably less computational complexity.

Citace poskytuje Crossref.org

000      
00000naa a2200000 a 4500
001      
bmc22018129
003      
CZ-PrNML
005      
20220804134557.0
007      
ta
008      
220720s2022 xxu f 000 0|eng||
009      
AR
024    7_
$a 10.1016/j.compbiomed.2022.105388 $2 doi
035    __
$a (PubMed)35349798
040    __
$a ABA008 $b cze $d ABA008 $e AACR2
041    0_
$a eng
044    __
$a xxu
100    1_
$a Koutny, Tomas $u Department of Computer Science and Engineering, University of West Bohemia, Univerzitni 8, 301 00, Plzen, Czech Republic. Electronic address: txkoutny@kiv.zcu.cz
245    10
$a Predicting glucose level with an adapted branch predictor / $c T. Koutny, M. Mayo
520    9_
$a BACKGROUND AND OBJECTIVE: Diabetes mellitus manifests as prolonged elevated blood glucose levels resulting from impaired insulin production. Such high glucose levels over a long period of time damage multiple internal organs. To mitigate this condition, researchers and engineers have developed the closed loop artificial pancreas consisting of a continuous glucose monitor and an insulin pump connected via a microcontroller or smartphone. A problem, however, is how to accurately predict short term future glucose levels in order to exert efficient glucose-level control. Much work in the literature focuses on least prediction error as a key metric and therefore pursues complex prediction methods such a deep learning. Such an approach neglects other important and significant design issues such as method complexity (impacting interpretability and safety), hardware requirements for low-power devices such as the insulin pump, the required amount of input data for training (potentially rendering the method infeasible for new patients), and the fact that very small improvements in accuracy may not have significant clinical benefit. METHODS: We propose a novel low-complexity, explainable blood glucose prediction method derived from the Intel P6 branch predictor algorithm. We use Meta-Differential Evolution to determine predictor parameters on training data splits of the benchmark datasets we use. A comparison is made between our new algorithm and a state-of-the-art deep-learning method for blood glucose level prediction. RESULTS: To evaluate the new method, the Blood Glucose Level Prediction Challenge benchmark dataset is utilised. On the official test data split after training, the state-of-the-art deep learning method predicted glucose levels 30 min ahead of current time with 96.3% of predicted glucose levels having relative error less than 30% (which is equivalent to the safe zone of the Surveillance Error Grid). Our simpler, interpretable approach prolonged the prediction horizon by another 5 min with 95.8% of predicted glucose levels of all patients having relative error less than 30%. CONCLUSIONS: When considering predictive performance as assessed using the Blood Glucose Level Prediction Challenge benchmark dataset and Surveillance Error Grid metrics, we found that the new algorithm delivered comparable predictive accuracy performance, while operating only on the glucose-level signal with considerably less computational complexity.
650    _2
$a algoritmy $7 D000465
650    _2
$a krevní glukóza $7 D001786
650    12
$a selfmonitoring glykemie $7 D015190
650    12
$a diabetes mellitus 1. typu $7 D003922
650    _2
$a lidé $7 D006801
650    _2
$a inzulin $7 D007328
655    _2
$a časopisecké články $7 D016428
655    _2
$a práce podpořená grantem $7 D013485
700    1_
$a Mayo, Michael $u Department of Computer Science, University of Waikato, New Zealand. Electronic address: michael.mayo@waikato.ac.nz
773    0_
$w MED00001218 $t Computers in biology and medicine $x 1879-0534 $g Roč. 145, č. - (2022), s. 105388
856    41
$u https://pubmed.ncbi.nlm.nih.gov/35349798 $y Pubmed
910    __
$a ABA008 $b sig $c sign $y p $z 0
990    __
$a 20220720 $b ABA008
991    __
$a 20220804134551 $b ABA008
999    __
$a ok $b bmc $g 1821955 $s 1169372
BAS    __
$a 3
BAS    __
$a PreBMC
BMC    __
$a 2022 $b 145 $c - $d 105388 $e 20220319 $i 1879-0534 $m Computers in biology and medicine $n Comput Biol Med $x MED00001218
LZP    __
$a Pubmed-20220720

Najít záznam

Citační ukazatele

Nahrávání dat ...

Možnosti archivace

Nahrávání dat ...