A Practical Deep Learning Model in Differentiating Pneumonia-Type Lung Carcinoma from Pneumonia on CT Images: ResNet Added with Attention Mechanism
Table 1
Summary of training and independent testing datasets.
Training set
Independent testing set
Parameter
PTLC
Pneumonia
PTLC
Pneumonia
value
value
No. of patients
88 (43.6)
114 (56.4)
43 (43)
57 (57)
Male patients
40 (19.8)
68 (33.7)
0.550
27 (27.0)
33 (33.0)
0.468
Age (y)
68.23 ± 12.46
66.28 ± 16.16
0.347
69.42 ± 16.54
69.34 ± 15.21
0.424
Pathology examination
Adenocarcinoma, n = 67; squamous cell carcinomas, n = 14; small cell undifferentiated carcinoma, n = 7
Bacterial pneumonia, n = 24; viral pneumonia, n = 11; CAP, n = 79
Adenocarcinoma, n = 33; squamous cell carcinomas, n = 7; small cell undifferentiated carcinoma, n = 3
Bacterial pneumonia, n = 16; viral pneumonia, n = 7; CAP, n = 34
Values in parentheses are percentages. PTLC, pneumonic-type lung carcinoma; CAP, community-acquired pneumonia. Ages are reported as means ± standard deviations.
