An accurate pediatric bone age prediction model using deep learning and contrast conversion / 이화의대지

Objectives@#This study aimed to develop an accurate pediatric bone age prediction model by utilizing deep learning models and contrast conversion techniques, in order to improve growth assessment and clinical decision-making in clinical practice. @*Methods@#The study employed a variety of deep learning models and contrast conversion techniques to predict bone age. The training dataset consisted of pediatric left-hand X-ray images, each annotated with bone age and sex information. Deep learning models, including a convolutional neural network , Residual Network 50 , Visual Geometry Group 19, Inception V3, and Xception were trained and assessed using the mean absolute error (MAE). For the test data, contrast conversion techniques including fuzzy contrast enhancement, contrast limited adaptive histogram equalization (HE) , and HE were implemented. The quality of the images was evaluated using peak signal-to-noise ratio (SNR), mean squared error, SNR, coefficient of variation, and contrast-to-noise ratio metrics. The bone age prediction results using the test data were evaluated based on the MAE and root mean square error, and the t-test was performed. @*Results@#The Xception model showed the best performance (MAE=41.12). HE exhibited superior image quality, with higher SNR and coefficient of variation values than other methods. Additionally, HE demonstrated the highest contrast among the techniques assessed, with a contrast-to-noise ratio value of 1.29. Improvements in bone age prediction resulted in a decline in MAE from 2.11 to 0.24, along with a decrease in root mean square error from 0.21 to 0.02. @*Conclusion@#This study demonstrates that preprocessing the data before model training does not significantly affect the performance of bone age prediction when comparing contrast-converted images with original images.

Fixation of the Greater Trochanter in Arthroplasty for Unstable Intertrochnateric Fracture

PURPOSE: We classified fractures of the greater trochanter (GT) and evaluated fracture fragment stability according to GT type. MATERIALS AND METHODS: A total of 43 patients with an unstable intertrochanteric fracture treated between January 2007 and July 2009 with bipolar hemiarthroplasty were included in this study. GT fractures were classified as type A, B, C, or D and fixed using either cerclage wiring alone, cerclage wiring and non-absorbable suture or a greater trochanteric reattachment (GTR) plate. RESULTS: Type A fractures were fixed using cerclage wiring with non-absorbable suture in two cases, cerclage wiring in six cases and GTR plate in seven cases. Failure occurred in three cases of type A fractures treated with cerclage wiring alone. A total of 11 type B fractures were fixed with cerclage wiring (7), cerclage wiring and non-absorbable suture (3) and GTR plate (1). There was no failure of type B fractures. Type C fractures were fixed using cerclage wiring with non-absorbable suture in one case and GTR plate in three. There was no fixation in three cases. Of 10 type D fractures, six were treated with cerclage wiring and one with GTR plate. Fixation was not performed in three patients. There was no failure in type C and D type fractures. CONCLUSION: Fracture fragment stability differed according to fracture types. Cerclage wiring alone was insufficient to fix type A fractures, so type A fracture required a stronger fixation method.

Two-Incision Bipolar Hemiarthroplasty for Treating Femoral Neck Fracture: Analysis of 15 Cases

INTRODUCTION: We wanted to evaluate the efficacy of bipolar hemiarthroplasty using Berger's two-incision surgical technique for patients with muscular weakness around the hip joint and/or a high risk of dislocation. MATERIALS AND METHODS: We performed bipolar hemiarthroplasty for 15 femoral neck fractures using Berger's two-incision technique between December 2005 and July 2007. The mean age of the patients was 75.2 years old. Four of them had difficulty in walking due to the sequalae after stroke and five have been treated for psychoneurologic disorders. We investigated the operation time, the length of the anterior and posterior incisions, the amount of bleeding, the time untill walking after the operation, the total hospital stay, the recovery to activities of daily living and the complications such as dislocation. RESULTS: The mean operating time was 93 minutes. The average anterior and posterior skin incision length was 6.4 cm and 7.2 cm, respectively. The mean amount of bleeding was 420 cc at the time of surgery and 230 cc postoperatively through a drain. The patients started walking at a mean of 3.3 (1 to 5) days after the operation and the mean hospitalization was 24.3 days. Fourteen patients went back to their pre-injured activities of daily living, except one case with an intraoperative periprosthetic fracture. As for complications, two cases (13.3%) of femoral fracture were intraoperatively observed and one case of skin necrosis on the anterior incision site occurred. There were not any cases of dislocation or infection. CONCLUSION: Two-incision bipolar hemiarthroplasty had advantages for rehabilitation in elderly patients who have a high risk of dislocation, as well as in the patients with muscle weakness. But the operation took a long time and it had a high complication rate.