Search

Objective To explore the application of deep learning in automatic classification of anemia with conjunctival images as input. Methods The conjunctival images of 284 patients undergoing elective surgery in the Department of Anesthesiology of the First Affiliated Hospital of Army Medical University from March 18 to April 26, 2021 were collected and analyzed prospectively. The images divided into 2 types: normal and anemia according to the corresponding hemoglobin concentration. Four deep learning algorithms, including InceptionV3, ResNet50V2, EfficientNetV2B0 and DenseNet121, were used to construct a prediction model for anemia. The performance of the model was evaluated by receiver operating characteristic (ROC) curve with accuracy, sensitivity, specificity, positive predictive value and negative predictive value. Results The area under ROC curve (AUC) was 0.709 (95%CI: 0.643~0.769), 0.661 (95%CI: 0.594~0.725), 0.670 (95%CI: 0.603~0.733), and 0.695 (95%CI: 0.628~0.756), respectively for the 4 deep learning algorithms. The InceptionV3 model showed superior predictive performance on the test set, with an AUC value of 0.709 (95%CI: 0.643~0.769), an accuracy of 0.695, a sensitivity of 0.750, a specificity of 0.412, a positive predictive value of 0.707 and a negative predictive value of 0.629. Based on the optimal algorithm, a network service application which can be used for online prediction of anemia was developed (http://150.158.58.4). Conclusion Our model, which is established based on deep learning algorithm with conjunctiva image as input, has a good performance on fast and automatic prediction for anemia. The InceptionV3model has better comprehensive prediction performance.

Elevated expression of long non-coding RNA homeobox A cluster antisense RNA 2 (lncRNA HOXA-AS2) is known to have prognostic value in various solid tumors. The present meta-analysis aimed to comprehensively quantify its prognostic significance across a wider spectrum of malignancies and to provide an updated synthesis of evidence that could refine prognostic models. To achieve this aim, multiple databases were carefully searched for lncRNA HOXA-AS2-related articles published in the past 10 years. Hazard ratios (HRs) or odds ratios (ORs) with 95% confidence intervals (CIs) were calculated to demonstrate the prognostic value of lncRNA HOXA-AS2 using Stata 15.0 software. The function of lncRNA HOXA-AS2 was inferred from its associations with key clinical outcomes such as lymph node metastasis, distant metastasis, tumor stage and tumor size, which may reflect its role in tumor biology. In the present systematic review and meta-analysis of 454 patients across 7 studies, it was found that high lncRNA HOXA-AS2 expression was significantly associated with a shorter overall survival (OS) time in patients with cancer (HR=2.14; 95% CI, 1.40–3.27; P<0.001). High lncRNA HOXA-AS2 expression was also associated with lymph node metastasis [odds ratio (OR)=2.06; 95% CI, 1.07–3.99; P=0.032], distant metastasis (OR=2.11; 95% CI, 1.15–3.88; P=0.016), advanced tumor stage (OR=2.71; 95% CI, 1.50–4.89; P=0.001) and larger tumor size (OR=2.02; 95% CI, 0.86–4.78; P=0.006). However, no significant association was observed with age (OR=1.00; 95% CI, 0.63–1.59; P=0.991) or sex (OR=1.55; 95% CI, 0.72–3.34; P=0.258). In conclusion, elevated expression of lncRNA HOXA-AS2 was significantly related to poor clinical outcomes in various cancer types, such as osteosarcoma, non-small cell lung cancer and papillary thyroid carcinoma, a finding that was further confirmed by the present study. Specifically, the potential of lncRNAHOXA-AS2 as a biomarker in assessing tumor stage, metastasis risk and OS in patients was demonstrated. However, the results of the present study also indicated that the expression of lncRNA HOXA-AS2 was not significantly associated with age or sex, suggesting its role in cancer progression might be independent of these factors. This insight may direct future research to place more focus on the relationship between lncRNA HOXA-AS2 and specific cancer types and clinical characteristics. [ABSTRACT FROM AUTHOR]