Automated rotator cuff tear classification using 3D convolutional neural network.

Rotator cuff tear (RCT) is one of the most common shoulder injuries. When diagnosing RCT, skilled orthopedists visually interpret magnetic resonance imaging (MRI) scan data. For automated and accurate diagnosis of RCT, we propose a full 3D convolutional neural network (CNN) based method using deep learning. This 3D CNN automatically diagnoses the presence or absence of an RCT, classifies the tear size, and provides 3D visualization of the tear location. To train the 3D CNN, the Voxception-ResNet (VRN) structure was used. This architecture uses 3D convolution filters, so it is advantageous in extracting information from 3D data compared with 2D-based CNNs or traditional diagnosis methods. MRI data from 2,124 patients were used to train and test the VRN-based 3D CNN. The network is trained to classify RCT into five classes (None, Partial, Small, Medium, Large-to-Massive). A 3D class activation map (CAM) was visualized by volume rendering to show the localization and size information of RCT in 3D. A comparative experiment was performed for the proposed method and clinical experts by using randomly selected 200 test set data, which had been separated from training set. The VRN-based 3D CNN outperformed orthopedists specialized in shoulder and general orthopedists in binary accuracy (92.5% vs. 76.4% and 68.2%), top-1 accuracy (69.0% vs. 45.8% and 30.5%), top-1±1 accuracy (87.5% vs. 79.8% and 71.0%), sensitivity (0.94 vs. 0.86 and 0.90), and specificity (0.90 vs. 0.58 and 0.29). The generated 3D CAM provided effective information regarding the 3D location and size of the tear. Given these results, the proposed method demonstrates the feasibility of artificial intelligence that can assist in clinical RCT diagnosis.

Web-based fully automated cephalometric analysis by deep learning.

BACKGROUND AND OBJECTIVE: An accurate lateral cephalometric analysis is vital in orthodontic diagnosis. Identification of anatomic landmarks on lateral cephalograms is tedious, and errors may occur depending on the doctor's experience. Several attempts have been made to reduce this time-consuming process by automating the process through machine learning; however, they only dealt with a small amount of data from one institute. This study aims to develop a fully automated cephalometric analysis method using deep learning and a corresponding web-based application that can be used without high-specification hardware. METHODS: We built our own dataset comprising 2,075 lateral cephalograms and ground truth positions of 23 landmarks from two institutes and trained a two-stage automated algorithm with a stacked hourglass deep learning model specialized for detecting landmarks in images. Additionally, a web-based application with the proposed algorithm for fully automated cephalometric analysis was developed for better accessibility regardless of the user's computer hardware, which is essential for a deep learning-based method. RESULTS: The algorithm was evaluated with datasets from various devices and institutes, including a widely used open dataset and achieved 1.37 ± 1.79 mm of point-to-point errors with ground truth positions for 23 cephalometric landmarks. Based on the predicted positions, anatomical types of the subjects were automatically classified and compared with the ground truth, and the automated algorithm achieved a successful classification rate of 88.43%. CONCLUSIONS: We expect that this fully automated cephalometric analysis algorithm and the web-based application can be widely used in various medical environments to save time and effort for manual marking and diagnosis.

Serial Changes in 3-Dimensional Supraspinatus Muscle Volume After Rotator Cuff Repair.

BACKGROUND: There is considerable debate on the recovery of rotator cuff muscle atrophy after rotator cuff repair. PURPOSE: To evaluate the serial changes in supraspinatus muscle volume after rotator cuff repair by using semiautomatic segmentation software and to determine the relationship with functional outcomes. STUDY DESIGN: Case series; Level of evidence, 4. METHODS: Seventy-four patients (mean age, 62.8 ± 8.8 years) who underwent arthroscopic rotator cuff repair and obtained 3 consecutive (preoperatively, immediately postoperatively, and later postoperatively [≥1 year postoperatively]) magnetic resonance imaging (MRI) scans having complete Y-views were included. We generated a 3-dimensional (3D) reconstructed model of the supraspinatus muscle by using in-house semiautomatic segmentation software (ITK-SNAP) and calculated both the 2-dimensional (2D) cross-sectional area and 3D volume of the muscle in 3 different views (Y-view, 1 cm medial to the Y-view [Y+1 view], and 2 cm medial to the Y-view [Y+2 view]) at the 3 time points. The area and volume changes at each time point were evaluated according to repair integrity. Later postoperative volumes were compared with immediately postoperative volumes, and their relationship with various clinical factors and the effect of higher volume increases on range of motion, muscle power, and visual analog scale pain and American Shoulder and Elbow Surgeons scores were evaluated. RESULTS: The interrater reliabilities were excellent for all measurements. Areas and volumes increased immediately postoperatively as compared with preoperatively; however, only volumes on the Y+1 view and Y+2 view significantly increased later postoperatively as compared with immediately postoperatively ( P < .05). There were 9 patients with healing failure, and area and volume changes were significantly less later postoperatively compared with immediately postoperatively at all measurement points in these patients ( P < .05). After omitting the patients with healing failure, volume increases later postoperatively became more prominent ( P < .05) in the order of the Y+2 view, Y+1 view, and Y-view. Volume increases were higher in patients who healed successfully with larger tears ( P = .040). Higher volume increases were associated only with an increase in abduction power ( P = .029) and not with other outcomes. CONCLUSION: The supraspinatus muscle volume increased immediately postoperatively and continuously for at least 1 year after surgery. The increase was evident in patients who had larger tears and healed successfully and when measured toward the more medial portion of the supraspinatus muscle. The volume increases were associated with an increase in shoulder abduction power.