Achieve accurate results. In this study, we investigated the clinical application of this system using a corrective osteotomy of malunited fractures of the upper extremity....

Background: Although the automatic diagnosis of fractures using artificial intelligence (AI) has recently been reported to be more accurate than those by orthopedics specialists, big data with at least 1000 images or more are required for deep learning of the convolutional neural network (CNN) to improve diagnostic accuracy. The aim of this study was to develop an AI system capable of diagnosing distal radius fractures with high accuracy even when learning with relatively small data by learning to use bi-planar X-rays images. Methods: VGG16, a learned image recognition model, was used as the CNN. It was modified into a network with two output layers to identify the fractures in plain X-ray images. We augmented 369 plain X-ray anteroposterior images and 360 lateral images of distal radius fractures, as well as 129 anteroposterior images and 125 lateral images of normal wrists to conduct training and diagnostic tests. Similarly, diagnostic tests for fractures of the styloid process of the ulna were conducted using 189 plain X-ray anteroposterior images of fractures and 302 images of the normal styloid process. The distal radius fracture is determined by entering an anteroposterior image of the wrist for testing into the trained AI. If it identifies a fracture, it is diagnosed as the same. However, if the anteroposterior image is determined as normal, the lateral image of the same patient is entered. If a fracture is identified, the final diagnosis is fracture; if the lateral image is identified as normal, the final diagnosis is normal. Results: The diagnostic accuracy of distal radius fractures and fractures of the styloid process of the ulna were 98.0 ± 1.6% and 91.1 ± 2.5%, respectively. The areas under the receiver operating characteristic curve were 0.991 {n = 540; 95% confidence interval (CI), 0.984–0.999} and 0.956 (n = 450; 95% CI 0.938–0.973). Conclusions: Our method resulted in a good diagnostic rate, even when using a relatively small amount of data.

Background: Medical image processing has facilitated simulation of 3-dimensional (3-D) corrective osteotomy, and 3-D rapid prototyping technology has further enabled the manufacturing of patient-matched surgical guides and implants (patient-matched instruments, or PMIs). However, 3-D corrective osteotomy using these technologies has not been the standard procedure. We aimed to prospectively verify the efficacy and safety of PMIs in corrective osteotomy for deformities of the upper extremity. Methods: We enrolled 16 patients with a total of 17 bone deformities in the upper extremity. Eight patients had distal radial malunion; 5, distal humeral malunion; and 3, forearm diaphyseal malunion. All cases underwent 3-D corrective osteotomy with PMIs. The primary end point was the residual maximum deformity angle (MDA), which was calculated from 2 deformity angles—1 on the anteroposterior and 1 on the lateral postoperative radiograph. Secondary end points included the deformity angle on radiographs, 3-D error between the preoperative planning model and the postoperative result, range of motion, grip strength, pain measured with a visual analog scale (VAS), patient satisfaction, and Disabilities of the Arm, Shoulder and Hand (DASH) score. Results: The average MDA significantly improved from 25.5° preoperatively to 3.3° at the final follow-up (p < 0.001). The angular deformity was within 5° in all cases, except for 1 with distal radial malunion who had a higher angle on the anteroposterior radiograph. The error between the correction seen on the postoperative 3-D bone model and the planned correction was <1° and <1 mm. Flexion and extension of the wrist and pronation of the forearm of the patients treated for distal radial malunion improved significantly, and pronation improved for those treated for forearm diaphyseal malunion. The average VAS score, grip strength, and DASH score significantly improved as well. Of the 16 patients, 15 were very satisfied or satisfied with the outcomes. Conclusions: Corrective osteotomy using PMIs achieved accurate correction and good functional recovery in the upper extremity. Although our study was limited to cases without any deformity on the contralateral side, 3-D corrective osteotomy using PMIs resolved treatment challenges for complex deformities in upper extremities. Level of Evidence: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.

Treatment of chronic radial head dislocation is controversial, considering whether to reduce and reconstruct the proximal radioulnar joint. The anatomic alteration that influences the decision to reduce the dislocation is not completely understood. We attempted to clarify the changes of the proximal radioulnar joint that occur in chronic radial head dislocations to clarify how they might influence the decision to perform repair. We evaluated 15 patients with chronic radial head dislocations categorized by duration of “early” (< 3 years) (n = 8) and “longstanding” (> 3 years) (n = 7) groups. We measured the angle and depth of the radial notch of the proximal ulna and evaluated radial head deformity using 3-D bone models created from CT data. For the early group, no differences were observed in the shape of the radial notch between affected and normal sides. For the longstanding group, the radial notch angle was greater on the affected side (mean ± SD, 45.5° ± 9.7°) than on the normal side (29.7° ± 6.3°), and the radial notch depth was smaller on the affected side (0.2 ± 1.6 mm) than on the normal side (2.3 ± 1.3 mm). The shape of the radial head was nearly normal in the early group, whereas the longstanding group had a dome-shaped deformity. In longstanding chronic radial head dislocation, deformation develops in the radial head and radial notch of the ulna, which is remodeled in a manner corresponding to the dislocated position of the radial head. Level III, prognostic study. See Guidelines for Authors for a complete description of levels of evidence.