Deep Learning Approach for Evaluating Knee MR Images: Achieving High Diagnostic Performance for Cartilage Lesion Detection

Abstract

Purpose To determine the feasibility of using a deep learning approach to detect cartilage lesions (including cartilage softening, fibrillation, fissuring, focal defects, diffuse thinning due to cartilage degeneration, and acute cartilage injury) within the knee joint on MR images. Materials and Methods A fully automated deep learning-based cartilage lesion detection system was developed by using segmentation and classification convolutional neural networks (CNNs). Fat-suppressed T2-weighted fast spin-echo MRI data sets of the knee of 175 patients with knee pain were retrospectively analyzed by using the deep learning method. The reference standard for training the CNN classification was the interpretation provided by a fellowship-trained musculoskeletal radiologist of the presence or absence of a cartilage lesion within 17 395 small image patches placed on the articular surfaces of the femur and tibia. Receiver operating curve (ROC) analysis and the κ statistic were used to assess diagnostic performance and intraobserver agreement for detecting cartilage lesions for two individual evaluations performed by the cartilage lesion detection system. Results The sensitivity and specificity of the cartilage lesion detection system at the optimal threshold according to the Youden index were 84.1% and 85.2%, respectively, for evaluation 1 and 80.5% and 87.9%, respectively, for evaluation 2. Areas under the ROC curve were 0.917 and 0.914 for evaluations 1 and 2, respectively, indicating high overall diagnostic accuracy for detecting cartilage lesions. There was good intraobserver agreement between the two individual evaluations, with a κ of 0.76. Conclusion This study demonstrated the feasibility of using a fully automated deep learning-based cartilage lesion detection system to evaluate the articular cartilage of the knee joint with high diagnostic performance and good intraobserver agreement for detecting cartilage degeneration and acute cartilage injury. © RSNA, 2018 Online supplemental material is available for this article .

Deep convolutional neural networks with multiplane consensus labeling for lung function quantification using UTE proton MRI

“…The DL framework for lung segmentation (Fig. a) used a 2D convolutional encoder‐decoder (CED) architecture, which has been successfully applied for cartilage and brain tissue segmentation . The encoder network uses the same 13 Visual Geometry Group 16 convolutional layers and the decoder uses a mirrored structure of the encoder network with max‐pooling replaced by an upsampling process.…”

“…successfully applied for cartilage and brain tissue segmentation. 23,26,35,36 The encoder network uses the same 13 Visual Geometry Group 16 convolutional layers and the decoder uses a mirrored structure of the encoder network with max-pooling replaced by an upsampling process. A symmetric shortcut connection (SC) between the encoder and decoder network is added to enhance the segmentation performance.…”

SUSAN: segment unannotated image structure using adversarial network

“…7 Deep CNN-based methods have achieved state-of-the-art performance in many medical image segmentation tasks including segmenting brain tumors, 8,9 tissues, 10,11 and multiple sclerosis lesions, 12 cardiac, 13,14 liver, 15 and lung 16 tissues, and musculoskeletal tissues such as bone and cartilage. [17][18][19] On the other hand, medical image segmentation is typically seen as a multiclass labeling problem which is closely related to the supervised semantic segmentation described in most segmentation CNN studies. In particular, convolutional encoder-decoder (CED) networks have proven to be highly efficient in the medical image domain.…”

Translation of morphological and functional musculoskeletal imaging

“…The best performance for staging lesion severity was obtained by including demographic factors, achieving accuracies of 80.74%, 78.02%, and 75.00% for normal, small and complex large lesions, respectively. Another study aimed to classify cartilage lesions using similar MRI knee data and deep learning . This method was tested on a smaller dataset of 175 subjects with a 2D patch‐based approach and “hard supervision”, with two radiologists (readers) identifying of the presence or absence of cartilage lesions such for 2D image patches (64 × 64) localized to the cartilage region.…”

“…Another study aimed to classify cartilage lesions using similar MRI knee data and deep learning. 33 This method was tested on a smaller dataset of 175 subjects with a 2D patch-based approach and "hard supervision", with two radiologists (readers) identifying of the presence or absence of cartilage lesions such for 2D image patches (64 Â 64) localized to the cartilage region. In this study, the accuracy in binary lesion detection was comparable to, 32 sensitivity and specificity equal to 84.1% and 85.2% respectively for evaluation from the first reader and 80.5% and 87.9%, respectively, for evaluation from the second reader.…”