TY - JOUR
T1 - Development and Validation of a Convolutional Neural Network for Automated Detection of Scaphoid Fractures on Conventional Radiographs
AU - Hendrix, Nils
AU - Scholten, Ernst
AU - Vernhout, Bastiaan
AU - Bruijnen, Stefan
AU - Maresch, Bas
AU - de Jong, Mathijn
AU - Diepstraten, Suzanne
AU - Bollen, Stijn
AU - Schalekamp, Steven
AU - de Rooij, Maarten
AU - Scholtens, Alexander
AU - Hendrix, Ward
AU - Samson, Tijs
AU - Ong, Sharon
AU - Postma, Eric
AU - van Ginneken, Bram
AU - Rutten, Matthieu
N1 - Funding Information:
We would like to acknowledge the resources provided by the hospitals Jeroen Bosch Ziekenhuis and Radboud University Medical Center for conducting this study. We also thank Chris Peters, MSc, PhD, and Peter Pijnenburg for their assistance in data acquisition, and Willem Huijbers for his feedback and guidance when setting up the research project.
Publisher Copyright:
© RSNA, 2021.
PY - 2021/7
Y1 - 2021/7
N2 - Purpose: To compare the performance of a convolutional neural network (CNN) to that of 11 radiologists in detecting scaphoid bone fractures on conventional radiographs of the hand, wrist, and scaphoid. Materials and Methods: At two hospitals (hospitals A and B), three datasets consisting of conventional hand, wrist, and scaphoid radiographs were retrospectively retrieved: a dataset of 1039 radiographs (775 patients [mean age, 48 years ± 23 {standard deviation}; 505 female patients], period: 2017–2019, hospitals A and B) for developing a scaphoid segmentation CNN, a dataset of 3000 radiographs (1846 patients [mean age, 42 years ± 22; 937 female patients], period: 2003–2019, hospital B) for developing a scaphoid fracture detection CNN, and a dataset of 190 radiographs (190 patients [mean age, 43 years ± 20; 77 female patients], period: 2011–2020, hospital A) for testing the complete fracture detection system. Both CNNs were applied consecutively: The segmentation CNN localized the scaphoid and then passed the relevant region to the detection CNN for fracture detection. In an observer study, the performance of the system was compared with that of 11 radiologists. Evaluation metrics included the Dice similarity coefficient (DSC), Hausdorff distance (HD), sensitivity, specificity, positive predictive value (PPV), and area under the receiver operating characteristic curve (AUC). Results: The segmentation CNN achieved a DSC of 97.4% ± 1.4 with an HD of 1.31 mm ± 1.03. The detection CNN had sensitivity of 78% (95% CI: 70, 86), specificity of 84% (95% CI: 77, 92), PPV of 83% (95% CI: 77, 90), and AUC of 0.87 (95% CI: 0.81, 0.91). There was no difference between the AUC of the CNN and that of the radiologists (0.87 [95% CI: 0.81, 0.91] vs 0.83 [radiologist range: 0.79–0.85]; P = .09). Conclusion: The developed CNN achieved radiologist-level performance in detecting scaphoid bone fractures on conventional radiographs of the hand, wrist, and scaphoid.
AB - Purpose: To compare the performance of a convolutional neural network (CNN) to that of 11 radiologists in detecting scaphoid bone fractures on conventional radiographs of the hand, wrist, and scaphoid. Materials and Methods: At two hospitals (hospitals A and B), three datasets consisting of conventional hand, wrist, and scaphoid radiographs were retrospectively retrieved: a dataset of 1039 radiographs (775 patients [mean age, 48 years ± 23 {standard deviation}; 505 female patients], period: 2017–2019, hospitals A and B) for developing a scaphoid segmentation CNN, a dataset of 3000 radiographs (1846 patients [mean age, 42 years ± 22; 937 female patients], period: 2003–2019, hospital B) for developing a scaphoid fracture detection CNN, and a dataset of 190 radiographs (190 patients [mean age, 43 years ± 20; 77 female patients], period: 2011–2020, hospital A) for testing the complete fracture detection system. Both CNNs were applied consecutively: The segmentation CNN localized the scaphoid and then passed the relevant region to the detection CNN for fracture detection. In an observer study, the performance of the system was compared with that of 11 radiologists. Evaluation metrics included the Dice similarity coefficient (DSC), Hausdorff distance (HD), sensitivity, specificity, positive predictive value (PPV), and area under the receiver operating characteristic curve (AUC). Results: The segmentation CNN achieved a DSC of 97.4% ± 1.4 with an HD of 1.31 mm ± 1.03. The detection CNN had sensitivity of 78% (95% CI: 70, 86), specificity of 84% (95% CI: 77, 92), PPV of 83% (95% CI: 77, 90), and AUC of 0.87 (95% CI: 0.81, 0.91). There was no difference between the AUC of the CNN and that of the radiologists (0.87 [95% CI: 0.81, 0.91] vs 0.83 [radiologist range: 0.79–0.85]; P = .09). Conclusion: The developed CNN achieved radiologist-level performance in detecting scaphoid bone fractures on conventional radiographs of the hand, wrist, and scaphoid.
KW - Computer-aided diagnosis
KW - Convolutional neural network (CNN)
KW - Deep learning algorithms
KW - Feature detection-vision-application domain
KW - Machine learning algorithms
U2 - 10.1148/ryai.2021200260
DO - 10.1148/ryai.2021200260
M3 - Article
C2 - 34350413
SN - 2638-6100
VL - 3
SP - e200260
JO - Radiology: Artificial Intelligence
JF - Radiology: Artificial Intelligence
IS - 4
M1 - e200260
ER -