Table 3.

Summary of studies addressing risk and patient stratification.

Stratification	Challenge	Period	Cohort	Population	Methods	Year	Refa
Complications	Group risks of retinopathy	5 years	55	T1Db	DTc, ANNd	2010	[95]
Disease complexity	Group combinations of comorbid conditions	2 years	15480	Chronic diseases	Hierarchical clustering	2011	[96]
Disease complexity	Group management profiles	3 months	239	T1D	Hierarchical clustering	2011	[97]
Disease complexity	Group management profiles	3.5 months	70	T2De	K-means	2012	[98]
Complications	Group biomechanical foot profiles	6 months	97	T1D, T2D	K-means	2013	[99]
Disease complexity	Group by drug purchases	7 years	953	T2D	Knowledge discovery	2015	[100]
Complications	Group risks of renal disease	3 years	109	T2D	K-means	2015	[101]
Complications	Group risks of complications	10 years	1441	T1D	Learning models	2015	[102]
Complications	Group risks of complications	5 years	84	T1D, T2D	RAf and ANN	2015	[103]
Complications	Group risks of retinopathy	<1 year	345	T2D	SVMg, RFh, DT, NBi	2015	[104]
Disease complexity	Groups of blood glucose profiles	4 months	10	T1D	Hierarchical clustering	2016	[105]
Complications	Group personal networks types	5 years	1862	T2D	RA, K-means;	2016	[106]
Disease progression	Group risks of T2D progression	5 years	24331	T2D	NB	2016	[107]
Complications	Group risks of retinopathy	2 years	323378	T2D	RF	2017	[108]
Disease complexity	Group blood glucose profiles	2 years	27005	T2D	RF	2017	[109]
Disease complexity	Groups of HbA1c profiles	5 years	684	T2D	RAs, KNN	2017	[110]
Weight intervention	Group of BMI profiles	31 years	2540	T2D	RA	2017	[111]
Complications	Group by retinopathy, neuropathy, or nephropathy	3,5,7 years	943	T2D	RF, RA	2018	[112]

^aref: reference.

^bT1D: type 1 diabetes.

^cDT: decision tree.

^dANN: artificial neural network.

^eT2D: type 2 diabetes.

^fRA: regression algorithm.

^gSVM: support vector machine.

^hRF: random forest.

ⁱNB: Naïve Bayes.