Page 114 - 《南京医科大学学报(自然科学版)》2026年第3期
P. 114
第46卷第3期
·422 · 南 京 医 科 大 学 学 报 2026年3月
A 48 46 45 26 1 B 49 49 49 49 47 46 45 44 35 28 18 10 3
10 1.5
5
Coefficients -5 0 Binomial deviance 1.4
-10 1.3
1.2
-15
-10 -8 -6 -4 -2 -10 -8 -6 -4 -2
log(λ) log(λ)
C D
10 LASSO Boruta
Importance 5 G - CHD
Age
Hemoglobin
0
Creatinine
APTT Urea nitrogen Temperature
-5 Serum sodium D⁃dimer
Total bilirubin Procalcitonin
x.cl
x.hr
x.mn
x.vis
x.plt
x.pri
x.ca
x.chd
x.cre
x.age
x.G
x.ne
x.ly
x.alb
x.fib
x.sbp
x.bg
x.mv
ShadowMin x.cancer x.stroke shadowMean x.wbc x.other x.sofa x.aptt
x.alt
A:The coefficient path diagram of LASSO regression. B:The cross⁃validation curve for LASSO regression analysis. C:The box plot resulting from
the Boruta algorithm. D:The Venn diagram used as a filter variable comparison between LASSO regression analysis and the Boruta algorithm.
图2 预测模型筛选变量路径图
Figure 2 Path diagram of filter variables for the prediction model
表3 脓毒症患者发生SAMI的多因素Logistic回归分析
Table 3 Multivariate logistic regression of developing SAMI in sepsis patients
Variate OR 95%CI P
Age(per 1⁃year increase) 1.017 1.000-1.035 0.049
Coronary heart disease(Yes) 3.158 1.464-7.227 0.005
Admission creatinine(per 1 μmol/L increase) 1.003 1.000-1.006 0.089
Admission urea nitrogen(per 1 mmol/L increase) 1.065 1.021-1.114 0.005
D⁃dimer(per 1 mg/L increase) 1.043 1.013-1.078 0.007
Procalcitonin(per 1 ng/mL increase) 1.008 1.000-1.015 0.043
0 10 20 30 40 50 60 70 80 90 100
Points
Age(years)
10 30 50 70 90
Yes
Coronary heart disease
No
Serum creatinine(μmol/L)
0 100 300 500 700 900 1 100 1 300
Serum urea nitrogen(mmol/L)
0 10 20 30 40 50 60 70 80 90
D⁃dimer(mg/L) 0 10 20 30 40
Procalcitonin(mg/L)
0 40 80 120 160 200 240
Total points
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
Risk of SAMI
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0.99
图3 SAMI预测列线图
Figure 3 The establishment of nomogram for predicting SAMI

