End-to-end deep learning model for segmentation and severity staging of anterior cruciate ligament injuries from MRI - 01/11/22
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder
An end-to-end deep learning segmentation and classification model was built to identify anterior cruciate ligament tears on MRI.
The deep learning model (double-linear layers U-Net) combined with radiomic features yields high accuracy in classifying anterior cruciate ligament as intact (90% accuracy), partially torn (82% accuracy) or fully ruptured (92% accuracy).
The final model using pseudo data yields mean Dice similarity coefficients of 0.84 (intact anterior cruciate ligament [ACL]), 0.75 (partially torn ACL) and 0.73 (fully ruptured ACL), and root mean square errors of 0.06 (intact ACL), 0.05 (partially torn ACL) and 0.05 (fully ruptured ACL) in segmentation.
The purpose of this study was to develop a semi-supervised segmentation and classification deep learning model for the diagnosis of anterior cruciate ligament (ACL) tears on MRI based on a semi-supervised framework, double-linear layers U-Net (DCLU-Net).
Materials and methods
A total of 297 participants who underwent of total of 303 MRI examination of the knee with fat-saturated proton density (PD) fast spin-echo (FSE) sequence in the sagittal plane were included. There were 214 men and 83 women, with a mean age of 37.46 ± 1.40 (standard deviation) years (range: 29–44 years). Of these, 107 participants had intact ACL (36%), 98 had partially torn ACL (33%), and 92 had fully ruptured ACL (31%). The DCLU-Net was combined with radiomic features for enhancing performances in the diagnosis of ACL tear. The different evaluation metrics for both classification (accuracy, sensitivity, accuracy) and segmentation (mean Dice similarity coefficient and root mean square error) were compared individually for each image class across the three phases of the model, with each value being compared to its respective value from the previous phase. Findings at arthroscopic knee surgery were used as the standard of reference.
With the addition of radiomic features, the final model yielded accuracies of 90% (95% CI: 83–92), 82% (95% CI: 73–86), and 92% (95% CI: 87–94) for classifying ACL as intact, partially torn and fully ruptured, respectively. The DCLU-Net achieved mean Dice similarity coefficient and root mean square error of 0.78 (95% CI: 0.71–0.80) and 0.05 (95% CI: 0.06–0.07), respectively, when segmenting the three ACL conditions with pseudo data (P < 0.001).
A dual-modules deep learning model with segmentation and classification capabilities was successfully developed. In addition, the use of semi-supervised techniques significantly reduced the amount of manual segmentation data without compromising performance.Le texte complet de cet article est disponible en PDF.
Keywords : Anterior cruciate ligament tears, Classification, Knee injuries, Magnetic resonance imaging, Semi-supervised learning, Segmentation
Abbreviations : 3D, ACL, AUROC, BMI, CI, DCLU-Net, DL, FSE, MRI, PD, SD
Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?