Hierarchical Severity Staging of Anterior Cruciate Ligament Injuries using Deep Learning with MRI Images
Authors:
Nikan K. Namiri,
Io Flament,
Bruno Astuto,
Rutwik Shah,
Radhika Tibrewala,
Francesco Caliva,
Thomas M. Link,
Valentina Pedoia,
Sharmila Majumdar
Abstract:
Purpose: To evaluate the diagnostic utility of two convolutional neural networks (CNNs) for severity staging of anterior cruciate ligament (ACL) injuries.
Materials and Methods: This retrospective analysis was conducted on 1243 knee MR images (1008 intact, 18 partially torn, 77 fully torn, and 140 reconstructed ACLs) from 224 patients (age 47 +/- 14 years, 54% women) acquired between 2011 and 20…
▽ More
Purpose: To evaluate the diagnostic utility of two convolutional neural networks (CNNs) for severity staging of anterior cruciate ligament (ACL) injuries.
Materials and Methods: This retrospective analysis was conducted on 1243 knee MR images (1008 intact, 18 partially torn, 77 fully torn, and 140 reconstructed ACLs) from 224 patients (age 47 +/- 14 years, 54% women) acquired between 2011 and 2014. The radiologists used a modified scoring metric. To classify ACL injuries with deep learning, two types of CNNs were used, one with three-dimensional (3D) and the other with two-dimensional (2D) convolutional kernels. Performance metrics included sensitivity, specificity, weighted Cohen's kappa, and overall accuracy, followed by McNemar's test to compare the CNNs performance.
Results: The overall accuracy and weighted Cohen's kappa reported for ACL injury classification were higher using the 2D CNN (accuracy: 92% (233/254) and kappa: 0.83) than the 3D CNN (accuracy: 89% (225/254) and kappa: 0.83) (P = .27). The 2D CNN and 3D CNN performed similarly in classifying intact ACLs (2D CNN: 93% (188/203) sensitivity and 90% (46/51) specificity; 3D CNN: 89% (180/203) sensitivity and 88% (45/51) specificity). Classification of full tears by both networks were also comparable (2D CNN: 82% (14/17) sensitivity and 94% (222/237) specificity; 3D CNN: 76% (13/17) sensitivity and 100% (236/237) specificity). The 2D CNN classified all reconstructed ACLs correctly.
Conclusion: 2D and 3D CNNs applied to ACL lesion classification had high sensitivity and specificity, suggesting that these networks could be used to help grade ACL injuries by non-experts.
△ Less
Submitted 13 April, 2020; v1 submitted 19 March, 2020;
originally announced March 2020.
