Skip to main content

Table 2 Results of predicting the occurrence of HF using three MLR models based on the most discriminative features, extracted for (i) all patients, (ii) MASLD( +) patients, and (iii) MASLD(−) patients of Dataset A

From: Machine learning identification of risk factors for heart failure in patients with diabetes mellitus with metabolic dysfunction associated steatotic liver disease (MASLD): the Silesia Diabetes-Heart Project

Method

Sensitivity

Specificity

CC with event [%]

CC without event [%]

CC All [%]

(i) All (5 parameters)

0.81

0.70

80.76

70.17

72.10

(i) MASLD( +) subgroup (5 parameters)

0.68

0.80

67.53

79.55

77.19

(ii) MASLD( +) subgroup (3 parameters)

0.69

0.75

69.07

74.65

73.56

(i) MASLD(−) subgroup (5 parameters)

0.84

0.74

83.74

74.07

75.67

(iii) MASLD(−) subgroup (2 parameters)

0.84

0.70

83.74

70.37

72.58

  1. The best results in the subgroup analysis are boldfaced
Back to article page

Cardiovascular Diabetology

ISSN: 1475-2840