Table 2.

Summary of reviewed studies addressing detection of adverse glycemic events: prediction horizon (PH) in minutes, objective population criteria, number of participants in the cohort, mean number of monitored days per patient, mean number of monitored hours per day, type of monitoring technology, existence of monitoring during the overnight period (O), and inclusion of exercise or physical activity information (E),

PH (min)	Population	Cohort	Days	Time	Measurements	O	E	Method	Refa	Year
0 min	T1Db	6	1 day	10 h	EEGc	✓		ANNd	[65]	2010
0 min	T1D	30	80-247 days	—	SMBGe		✓	RFf, SVMg	[66]	2012
0 min	T1D	15	1 day	10 h	CGMh	✓		ANN, PSOi	[67]	2012
0 min	T1D	10	30 days	4 h	CGM			SVM	[68]	2013
30, 60 min	T1D	15	12.5 days	24 h	CGM	✓	✓	SVM	[69]	2013
0 min	T1D	10	4.5 days	6 h	CGM			SVM	[70]	2013
30 min	T1D	10	17.3 days	24 h	CGM	✓		DTj	[71]	2013
24 h	T2Dk	163	—l	—l	SMBG			RF	[72]	2015
0 min	T1D	15	1 day	4 h	CGM			ANN	[73]	2014
0 min	T1D	10	—m	—m	SMBG, ECG	✓		ANN	[74]	2014
2, 7, 30, 61-90 days	T1D, T2D	201, 323	—n	—n	SMBG			Pattern recognition	[75]	2014
0 min	T1D	15	1 day	10 h	ECG	✓	✓	ANN	[76]	2016
Past events	T2D	119695	>12 days	—	EHRo			NLPp	[77]	2016
0 min	T1D, T2D	500	1 day	2 h	SMBG			DT, ANN	[78]	2017

^aRef: reference.

^bT1D: type 1 diabetes.

^cEEG: electroencephalogram.

^dANN: artificial neural network.

^eSMBG: self-monitoring blood glucose.

^fRF: random forest.

^gSVM: support vector machine.

^hCGM: continuous glucose monitoring.

ⁱPSO: particle swarm optimization.

^jDT: decision tree.

^kT2D: type 2 diabetes.

^l344 data points.

^m18 data points.

ⁿ787 data points.

^oEHR: electronic health record.

^pNLP: natural language processing.