Objective To improve the efficiency of skin soft tissueexpansion with the overlapping tissue expansion techniques. Methods From June 2003 to March 2005, 5 cases of skin soft tissue defects were treated with the overlapping tissue expansion techniquetwo overlapped expanders in one soft tissue pocket, which was different from the traditional technique——one expander in one soft tissue pocket. Five patients included 3 males and 2 females, aging from 11 to 28 years. The defect was caused by scar of forearm in 2 cases, by melanotic nevus in 1 caseand by cicatricial baldness in 2 cases. The disease course was 1.5 to 24 years. Thedefect size ranged from 12 cm×5 cm to 13 cm×12 cm. Results Skin expansion process was satisfactory and skin defect was completely repaired with the expanded skin tissue in one operation in 5 cases. After operation, the wound of donor-recipient site healed by first intention. All patients were followed up from 3 to 15months, no contracture, pigmentation and scar occurred at the expanded skin area. The long-term appearances were satisfactory. Conclusion Compare with the traditional tissue expansion techniques, the new overlapping tissue expansion techniques can apparently improve the efficiency of skin soft tissue expansion. Itis suitable for the patients whose expandable skin is limited or no more skin tissue can be dissected near the skin defect and who need more expandable skin torepair skin defect.
ObjectiveTo realize automatic risk bias assessment for the randomized controlled trial (RCT) literature using BERT (Bidirectional Encoder Representations from Transformers) as an approach for feature representation and text classification.MethodsWe first searched The Cochrane Library to obtain risk bias assessment data and detailed information on RCTs, and constructed data sets for text classification. We assigned 80% of the data set as the training set, 10% as the test set, and 10% as the validation set. Then, we used BERT to extract features, construct text classification model, and evaluate the seven types of risk bias values (high and low). The results were compared with those from traditional machine learning methods using a combination of n-gram and TF-IDF as well as the Linear SVM classifier. The accuracy rate (P value), recall rate (R value) and F1 value were used to evaluate the performance of the models.ResultsOur BERT-based model achieved F1 values of 78.5% to 95.2% for the seven types of risk bias assessment tasks, which was 14.7% higher than the traditional machine learning method. F1 values of 85.7% to 92.8% were obtained in the extraction task of the other six types of biased descriptors except "other sources of bias", which was 18.2% higher than the traditional machine learning method.ConclusionsThe BERT-based automatic risk bias assessment model can realize higher accuracy in risk of bias assessment for RCT literature, and improve the efficiency of assessment.