Table 3 Data characteristics of the internal (SMC) and external (Intermountain) dataset.

From: PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging

 

Overall

Train

Validation

Test

External test

(intermountain)

Number of studies

1797

1461

167

169

200

Median age (IQR)

66.14 (53.24–82.40)

66.13 (53.14–82.95)

64.10 (50.88–78.38)

67.24 (56.62–82.76)

55.3 (42.0–69.5)

Number of patients (Female %)

1773 (57.07%)

1414 (56.64%)

162 (67.36%)

163 (52.08%)

198 (58.5%)

Median number of slices (IQR)

386 (134)

385 (136)

388 (132)

388 (139)

324

Number of positive PE

655

488

82

85

94

Number of negative PE

1142

973

85

84

106

  1. The internal SMC dataset was divided into training, validation and test. The training set was used to optimize model parameters and the validation set was used to select the best model and operating points. The hold-out test set was used to evaluate the model’s performance. The external Intermountain dataset was used solely for evaluation.
Back to article page