From: Convolutional neural network for detecting rib fractures on chest radiographs: a feasibility study
CNN model | Chest radiograph | Total | |
---|---|---|---|
With rib fractures | Without rib fractures | ||
(a) CNN model | |||
Detected fractures | 117 | 13 | 130 |
Undetected fractures | 4 | 67 | 71 |
Total | 121 | 80 | 201 |
Senior radiologist | Chest radiographs | Total | |
---|---|---|---|
With rib fractures | Without rib fractures | ||
(b) Senior radiologist | |||
Detected fractures | 117 | 8 | 125 |
Undetected fractures | 4 | 72 | 76 |
Total | 121 | 80 | 201 |
Junior radiologist | Chest radiographs | Total | |
---|---|---|---|
With rib fractures | Without rib fractures | ||
(c) Junior radiologist | |||
Detected fractures | 94 | 3 | 97 |
Undetected fractures | 27 | 77 | 104 |
Total | 121 | 80 | 201 |
Model | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | Accuracy (%) |
---|---|---|---|---|---|
(d) Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy in the independent testing set, based on the case level | |||||
CNN model | 96.7% | 83.8% | 90.0% | 94.4% | 91.5% |
Senior radiologist | 96.7 | 90.0 | 93.6 | 94.7 | 94.0 |
Junior radiologist | 77.7 | 96.3 | 96.9 | 74.0 | 85.1 |