Establishment and Reliability Evaluation of Prognostic Models in Diabetic Foot.

Purpose: To study the risk factors affecting amputation and survival in patients with diabetic foot (DF) and to construct a predictive model using the machine learning technique for DF foot amputation and survival and evaluate its effectiveness. Materials and Methods: A total of 200 patients with DF hospitalized in the First Affiliated Hospital of Shantou University Medical College in China were selected via cluster analysis screening, Kaplan-Meier survival calculation, amputation rate and Cox proportional hazards model investigation of risk factors associated with amputation and death. In addition, we constructed various models, including Cox proportional hazards regression analysis, the deep learning method convolution neural network (CNN) model, backpropagation (BP) neural network model, and backpropagation neural network prediction model after optimizing the genetic algorithm. The accuracy of the 4 prediction models for survival and amputation was assessed, and we evaluated the reliability of these computational models based on the size of the area under the ROC curve (AUC), sensitivity and specificity. Results: We found that the 1-year survival rate in patients with DF was 88.5%, and the 1-year amputation rate was 12.5%. Wagner's Classification of Diabetic Foot Ulcers grade, ankle-brachial index (ABI), low-density lipoprotein (LDL), and percutaneous oxygen partial pressure (TcPO2) were independent risk factors for amputation in patients with DF, while cerebrovascular disease, Sudoscan sweat gland function score, glycated hemoglobin (HbA1c) and peripheral artery disease (PAD) were independent risk factors for death in patients with DF. In addition, our results showed that in the case of amputation, the COX regression predictive model revealed an AUC of 0.788, sensitivity of 74.1% and specificity of 83.6%. The BP neural network predictive model identified an AUC of 0.874, sensitivity of 87.0% and specificity of 87.7%. An AUC of 0.909, sensitivity of 90.7% and specificity of 91.1% were found after optimizing the BP neural network prediction model via genetic algorithm. In the deep learning CNN model, the AUC, sensitivity and specificity were 0.939, 92.6%, and 95.2%, respectively. In the analysis of risk factors for death, the COX regression predictive model identified the AUC, sensitivity and specificity as 0.800, 74.1% and 85.9%, respectively. The BP neural network predictive model revealed an AUC, sensitivity and specificity of 0.937, 93.1% and 94.4%, respectively. Genetic algorithm-based optimization of the BP neural network predictive model identified an AUC, sensitivity and specificity of 0.932, 91.4% and 95.1%, respectively. The deep learning CNN model found the AUC, sensitivity and specificity to be 0.861, 82.8% and 89.4%, respectively. Conclusion: To identify risk factors for death, the BP neural network predictive model and genetic algorithm-based optimizing BP neural network predictive model have higher sensitivity and specificity than the deep learning method CNN predictive model and COX regression analysis.

Ethnicity modifies the association between functional microRNA polymorphisms and breast cancer risk: a HuGE meta-analysis.

Common functional polymorphisms in the promoter region of microRNAs (miRNAs), based on multiple lines of evidence, might participate in transcriptional regulation and other biological processes, which interact to increase the risk of developing breast cancer. Since 2005, many studies have investigated the association between breast cancer risk and common single nucleotide polymorphisms (SNPs) in miRNAs. However, the findings of several meta-analyses are inconclusive or ambiguous. The aim of this Human Genome Epidemiology meta-analysis was to determine more precisely the relationship between common miRNA polymorphisms and breast cancer risk. Twelve case-control studies with a total of 7,170 breast cancer patients and 8,783 healthy controls were included. Eight SNPs in miRNA genes were examined. When all eligible studies were pooled in the meta-analysis, the miR-196a-2 rs11614913*T, miR-499 rs3746444*T, and miR-605 rs2043556*A alleles predicted a decreased risk of breast cancer among Asians, while not Caucasians. In addition, the miR-27a rs895919*C allele might be a protective factor for breast cancer among Caucasians. However, for the miR-146a rs2910164 (G>C), miR-149 rs2292832 (G>T), miR-373 rs12983273 (C>T), and miR-423 rs6505162 (C>A) polymorphisms, we failed to find any significant association with the risk of breast cancer in any genetic model. In conclusion, the current meta-analysis supports that the miR-196a-2 rs11614913*T, miR-499 rs3746444*T, miR-605 rs2043556*A, and miR-27a rs895919*C alleles might be protective factors for breast cancer.

Upper gastrointestinal tract heterotopic pancreas: findings from CT and endoscopic imaging with histopathologic correlation.

OBJECTIVE: To evaluate and describe computed tomographic (CT) and endoscopic (ES) imaging findings in patients with pathologically confirmed upper gastrointestinal (GI) tract heterotopic pancreas (HP). METHODS: Findings from imaging examinations in 11 patients with pathologically confirmed HP were retrospectively reviewed (CT images obtained from 11 patients and ES images from 6 patients were available for review). Two radiologists evaluated lesion location, size, shape and border as well as growth pattern, enhancement pattern, enhancement grade and number of tumors. The presence of surface dimpling, prominent enhancement of overlying mucosa, and low intralesional attenuation were also evaluated. RESULTS: HP in the upper GI tract showed typical features in CT imaging: submucosal masses, ill-defined borders, endoluminal growth patterns, bright enhancement similar to the normal pancreas, surface dimpling and low intralesional attenuation. Endoscopic photographs manifested an endoluminal, ill-defined, submucosal mass in the upper GI tract wall, typically with central umbilication. The LD (long diameter)/SD (short diameter) ratios were found to be significantly different between HP in the stomach and HP in the duodenum (P<.05 for each finding). In addition, HP in the duodenum tended to be small and round. CONCLUSIONS: HP exhibits typical pancreatic pathologic features. Images showed characteristic features in CT imaging: submucosal masses, ill-defined lesions with an endoluminal growth pattern, bright enhancement similar to the normal pancreas, surface dimpling and low intralesional attenuation. ES imaging showed an endoluminal, ill-defined, submucosal mass, typically with central umbilication.