J Clin Med Res

Journal of Clinical Medicine Research, ISSN 1918-3003 print, 1918-3011 online, Open Access

Article copyright, the authors; Journal compilation copyright, J Clin Med Res and Elmer Press Inc

Journal website https://jocmr.elmerjournals.com

Original Article

Volume 17, Number 11, November 2025, pages 653-662

Comparison of the SeizCT Primer and Optimized Models for Predicting Positive Computed Tomography Findings in Patients With Non-Traumatic Seizures

↓ Figure 1. Study flow diagram. GCS: Glasgow Coma Scale; AuROC: area under the receiver operating characteristic curve; O:E ratio: observed-to-expected ratio; CITL: calibration-in-the-large.

↓ Figure 2. Receiver operating characteristic (ROC) curves comparing the SeizCT primer and optimized models using equivalence testing. CI: confidence interval.

↓ Figure 3. Calibration plots of the SeizCT primer and optimized models. AUC: area under the curve; O:E ratio: observed-to-expected ratio; CITL: calibration-in-the-large; CI: confidence interval.

↓ **Table 1.** Baseline Characteristics of Patients in the SeizCT Primer, SeizCT Optimized, and External Validation Cohorts
Baseline characteristics	SeizCT primer (n = 362), n (%)	SeizCT optimized (n = 625), n (%)	Validation (n = 312), n (%)	P value
SD: standard deviation; SBP: systolic blood pressure; DBP: diastolic blood pressure; GCS: Glasgow Coma Scale; IQR: interquartile range; GCS change: GCS change from baseline; CT: computed tomography.
Male	267 (73.8)	468 (74.9)	210 (67.3)	0.045
Age (years), mean ± SD	55.2 ± 16.2	54.8 ± 16.8	53.4 ± 17.5	0.347
SBP (mm Hg), mean ± SD	141.2 ± 29.2	140.6 ± 28.2	137.1 ± 29.0	0.126
DBP (mm Hg), mean ± SD	83.9 ± 18.6	84.6 ± 18.4	82.1 ± 19.3	0.152
GCS, median (IQR)	15 (10, 15)	15 (10, 15)	15 (11, 15)	0.076
Eye, median (IQR)	4 (3, 4)	4 (4, 4)	4 (4, 4)	0.414
Verbal, median (IQR)	5 (1, 5)	5 (2, 5)	5 (2, 5)	0.003
Motor, median (IQR)	6 (5, 6)	6 (5, 6)	6 (5, 6)	0.271
GCS change, median (IQR)	0 (0, 5)	0 (0, 4)	0 (0, 2)	0.006
Focal neurological deficit	62 (17.1)	153 (24.5)	92 (29.5)	0.001
Previous CT
No previous CT	172 (47.5)	311 (49.8)	125 (40.1)	0.003
Normal finding	91 (25.1)	110 (17.6)	73 (23.4)
Abnormal finding	99 (27.3)	204 (32.6)	114 (36.5)
Prior stroke more than 3 months	74 (20.4)	111 (17.8)	90 (28.9)	0.001
Current cancer	15 (4.1)	31 (5.0)	21 (6.7)	0.302
Epilepsy	86 (23.8)	151 (24.2)	77 (24.7)	0.960
End-stage renal disease	25 (6.9)	44 (7.0)	15 (4.8)	0.398
Chronic liver disease	5 (1.4)	16 (2.6)	33 (10.6)	< 0.001
Anticoagulant use	11 (3.1)	33 (5.3)	12 (3.9)	0.257
Hypertension	124 (34.3)	207 (33.1)	138 (44.2)	0.003
Diabetes mellitus	65 (18.0)	106 (17.0)	65 (20.8)	0.347
Alcohol consumption	147 (40.7)	237 (38.3)	77 (24.7)	< 0.001
Substance use	6 (1.7)	22 (3.5)	8 (2.6)	0.224
Alcohol withdrawal symptoms	58 (16.0)	94 (15.0)	34 (10.9)	0.001
Positive CT findings	71 (19.6)	118 (18.9)	58 (18.6)	0.940

↓ Table 1. Baseline Characteristics of Patients in the SeizCT Primer, SeizCT Optimized, and External Validation Cohorts

Baseline characteristics

SeizCT primer (n = 362), n (%)

SeizCT optimized (n = 625), n (%)

Validation (n = 312), n (%)

P value

SD: standard deviation; SBP: systolic blood pressure; DBP: diastolic blood pressure; GCS: Glasgow Coma Scale; IQR: interquartile range; GCS change: GCS change from baseline; CT: computed tomography.

Male

267 (73.8)

468 (74.9)

210 (67.3)

0.045

Age (years), mean ± SD

55.2 ± 16.2

54.8 ± 16.8

53.4 ± 17.5

0.347

SBP (mm Hg), mean ± SD

141.2 ± 29.2

140.6 ± 28.2

137.1 ± 29.0

0.126

DBP (mm Hg), mean ± SD

83.9 ± 18.6

84.6 ± 18.4

82.1 ± 19.3

0.152

GCS, median (IQR)

15 (10, 15)

15 (11, 15)

0.076

Eye, median (IQR)

4 (3, 4)

4 (4, 4)

0.414

Verbal, median (IQR)

5 (1, 5)

5 (2, 5)

0.003

Motor, median (IQR)

6 (5, 6)

0.271

GCS change, median (IQR)

0 (0, 5)

0 (0, 4)

0 (0, 2)

0.006

Focal neurological deficit

62 (17.1)

153 (24.5)

92 (29.5)

0.001

Previous CT

No previous CT

172 (47.5)

311 (49.8)

125 (40.1)

0.003

Normal finding

91 (25.1)

110 (17.6)

73 (23.4)

Abnormal finding

99 (27.3)

204 (32.6)

114 (36.5)

Prior stroke more than 3 months

74 (20.4)

111 (17.8)

90 (28.9)

0.001

Current cancer

15 (4.1)

31 (5.0)

21 (6.7)

0.302

Epilepsy

86 (23.8)

151 (24.2)

77 (24.7)

0.960

End-stage renal disease

25 (6.9)

44 (7.0)

15 (4.8)

0.398

Chronic liver disease

5 (1.4)

16 (2.6)

33 (10.6)

< 0.001

Anticoagulant use

11 (3.1)

33 (5.3)

12 (3.9)

0.257

Hypertension

124 (34.3)

207 (33.1)

138 (44.2)

0.003

Diabetes mellitus

65 (18.0)

106 (17.0)

65 (20.8)

0.347

Alcohol consumption

147 (40.7)

237 (38.3)

77 (24.7)

< 0.001

Substance use

6 (1.7)

22 (3.5)

8 (2.6)

0.224

Alcohol withdrawal symptoms

58 (16.0)

94 (15.0)

34 (10.9)

0.001

Positive CT findings

71 (19.6)

118 (18.9)

58 (18.6)

0.940

↓ **Table 2.** Predictive Performance of the SeizCT Primer and SeizCT Optimized Models in the External Validation Cohort
Parameters	SeizCT primer model	SeizCT optimized model	P value
^aP value from a two-sided equivalence test. AuROC: area under the receiver operating characteristic curve; O:E ratio: observed-to-expected ratio; CITL: calibration-in-the-large.
AuROC (original)	0.8156 (0.7586, 0.8727)	0.8221 (0.7813, 0.8629)	0.798
AuROC (validation)	0.7218 (0.6476, 0.7961)	0.7394 (0.6663, 0.8124)	< 0.015^a
Slope	0.582 (0.358, 0.805)	0.701 (0.454, 0.948)	0.002
O:E ratio	0.934	0.971	0.030
CITL	-0.112 (-0.441, 0.216)	-0.048 (-0.371, 0.276)	0.783

↓ Table 2. Predictive Performance of the SeizCT Primer and SeizCT Optimized Models in the External Validation Cohort

Parameters

SeizCT primer model

SeizCT optimized model

P value

^aP value from a two-sided equivalence test. AuROC: area under the receiver operating characteristic curve; O:E ratio: observed-to-expected ratio; CITL: calibration-in-the-large.

AuROC (original)

0.8156 (0.7586, 0.8727)

0.8221 (0.7813, 0.8629)

0.798

AuROC (validation)

0.7218 (0.6476, 0.7961)

0.7394 (0.6663, 0.8124)

< 0.015^a

Slope

0.582 (0.358, 0.805)

0.701 (0.454, 0.948)

0.002

O:E ratio

0.934

0.971

0.030

CITL

-0.112 (-0.441, 0.216)

-0.048 (-0.371, 0.276)

0.783

↓ **Table 3.** Confusion Matrix of the SeizCT Primer and SeizCT Optimized Models in the External Validation Cohort
CT brain	SeizCT primer model	SeizCT optimized model
CT: computed tomography; PPV: positive predictive value; NPV: negative predictive value.
Positive (n = 58)	34 (10.9)	24 (7.7)	33 (10.6)	25 (8.0)
Negative (n = 254)	59 (18.9)	195 (62.5)	53 (17.0)	201 (64.4)
Total	93	219	86	226
Sensitivity	58.6 (44.9, 71.4)	56.9 (43.2, 69.8)
Specificity	76.8 (71.1, 81.8)	79.1 (73.6, 84.0)
PPV	36.6 (26.8, 47.2)	38.4 (28.1, 49.5)
NPV	89.0 (84.1, 92.9)	88.9 (84.1, 92.7)

↓ Table 3. Confusion Matrix of the SeizCT Primer and SeizCT Optimized Models in the External Validation Cohort

CT brain

SeizCT primer model

SeizCT optimized model

Cut-point probabilities

> 20

≤ 20

> 22.57

≤ 22.57

CT: computed tomography; PPV: positive predictive value; NPV: negative predictive value.

Positive (n = 58)

34 (10.9)

24 (7.7)

33 (10.6)

25 (8.0)

Negative (n = 254)

59 (18.9)

195 (62.5)

53 (17.0)

201 (64.4)

Total

219

226

Sensitivity

58.6 (44.9, 71.4)

56.9 (43.2, 69.8)

Specificity

76.8 (71.1, 81.8)

79.1 (73.6, 84.0)

PPV

36.6 (26.8, 47.2)

38.4 (28.1, 49.5)

NPV

89.0 (84.1, 92.9)

88.9 (84.1, 92.7)