Genetic Folding (GF) Algorithm with Minimal Kernel Operators to Predict Stroke Patients

Figures & data

Table 1. List of features used in the experiment.

Features	Description	Mean	SD
Gender (gen)	“Male” (M), “Female” (F)	-	-
Age	age of the patient	43.2	22.6
Hypertension (hyp)	0 if the patient does not have hypertension, 1 if the patient has hypertension	0.097	0.30
Heart disease (hrt)	0 if the patient does not have any heart diseases, 1 if the patient has a heart disease	0.054	0.23
Work type (work)	“children” (Chi), “Govt_jov” (Gov), “Never_worked” (Never), “Private” (P) or “Self-employed” (Self)	-	-
avg_glucose_level (gluco)	average glucose level in blood	106.1	45.2
BMI	body mass index	28.8	7.69
Smoking status (somke)	“formerly smoked” (form), “never smoked” (never), “smokes” (sm) or “Unknown” (Unk)	-	-
Residence_type (resi)	“Rural” or “Urban”	-	-
Stroke (str)	1 if the patient had a stroke or 0 if not	0.048	0.22
ever_married (mar)	“No” or “Yes”	-	-

Table 2. Sample of stroke dataset.

gen	age	hyp	hrt	mar	work	resi	gluco	bmi	smoke	str
M	67.0	0	1	Yes	P	Urban	228.69	36.6	form	1
F	61.0	0	0	Yes	Self	Rural	202.21	NaN	never	1
M	80.0	0	1	Yes	P	Rural	105.92	32.5	never	1
F	49.0	0	0	Yes	P	Urban	171.23	34.4	smok	1

Table 3. Scaled sample of stroke dataset.

gen	age	hyp	hrt	mar	work	resi	gluco	bmi	unk	form	nev	sm	str
0	0.1	0	0	0	1	0	0.2	0.1	1	0	0	0	0
1	0.9	1	0	1	0	0	0.2	0.4	0	0	0	1	0
1	0.4	0	0	0	0	0	0.1	0.3	1	0	0	0	0
1	0.9	0	0	1	0.1	0	0.1	0.2	0	0	1	0	0
1	0.9	0	0	1	0	0	0.2	0.1	0	0	0	1	0
0	0.8	0	0	1	0.1	0	0.2	0.4	1	0	0	0	0
1	0.8	1	0	1	0	1	0.2	0.5	0	0	0	1	0
0	0.1	0	0	0	1	1	0.3	0.1	1	0	0	0	0
1	0.8	0	0	0	0.1	0	0.1	0.5	0	0	0	1	0
1	0.5	0	0	1	0	1	0.1	0.5	0	0	1	0	0

Figure 1. Balanced data.

Figure 2. Unbalanced data.

Figure 3. Proposed methodology.

Figure 4. GF algorithm producing SVM-kernels.

Figure 5. mGF algorithm producing SVM-kernels.

Table 4. mGF parameters.

Parameters	Values
Population	50
Generation	25
GF size	25
Crossover rate	0.5
Mutation rate	0.01

Figure 6. mGF pseudo-code.

Figure 7. mGF results.

Table 5. Accuracy comparisons.

Model	Ref.	Accuracy (%)	K-Fold Mean Accuracy (%)
SVM	(Veena and Ramakrishna 2021)	80.0	-
RF		73.0	-
DT		66.0	-
KNN		80.0	-
NB		82.0	-
mGF	Proposed	80.8	83.2
GF		80.1	81.7
LR		77.20	79.45
BernoulliNB		60.57	72.94
GaussianNB		19.28	57.33

Table 6. Performance comparison results for state-of-the-Art approaches and proposed mGF.

Model Name	(Almadani and Alshammari 2018) after tuning	(Almadani and Alshammari 2018)	(Sung et al. 2015)	(Adam, Yousif, and Bashir 2016)	Proposed mGF
Oversampling technique
XGBoost	0.75	0.52	-	-	-
RF	0.73	0.53	0.58	-	-
LR	-	0.75	0.68	-	-
KNeighbours	-	0.60	0.70	-	-
GussianNB	-	0.56	-	-	-
BernoulliNB	-	0.71	-	-	-
DT	-	0.52	0.57	-	-
Undersampling technique
DT	-	-	0.41	0.66	-
LR	-	-	-	0.78	-
RF	-	-	0.44	0.73	-
NB	-	-	-	0.82	-
KNN	-	-	-	0.80	-
Linear-SVM	-	-	-	-	0.80
Polynomial-SVM	-	-	-	-	0.78
RBF-SVM	-	-	-	-	0.81
mGF	-	-	-	-	0.88

Veena, G., and M. Ramakrishna. 2021. A survey on image encryption using chaos-based techniques. International Journal of Advanced Computer Science and Applications 12 (1). doi:10.14569/IJACSA.2021.0120145.

 Web of Science ®Google Scholar

Almadani, O., and R. Alshammari. 2018. Prediction of stroke using data mining classification techniques. International Journal of Advanced Computer Science and Applications 9 (1). doi:10.14569/IJACSA.2018.090163.

 Web of Science ®Google Scholar

Sung, S.-F., C.-Y. Hsieh, Y.-H.K Yang, H.-J. Lin, C.-H. Chen, Y.-W. Chen, and Y.-H. Hu. 2015. Developing a stroke severity index based on administrative data was feasible using data mining techniques. Journal of Clinical Epidemiology 68 (11):1292–300. doi:10.1016/j.jclinepi.2015.01.009.

 PubMed Web of Science ®Google Scholar

Adam, S. Y., A. Yousif, and M. B. Bashir. 2016. Classification of ischemic stroke using machine learning algorithms. International Journal of Computer Applications 149 (10):26–3852. doi:10.5120/ijca2016911607.

 Google Scholar