Deep learning method can be used to automatically analyze electrocardiogram (ECG) data and rapidly implement arrhythmia classification, which provides significant clinical value for the early screening of arrhythmias. How to select arrhythmia features effectively under limited abnormal sample supervision is an urgent issue to address. This paper proposed an arrhythmia classification algorithm based on an adaptive multi-feature fusion network. The algorithm extracted RR interval features from ECG signals, employed one-dimensional convolutional neural network (1D-CNN) to extract time-domain deep features, employed Mel frequency cepstral coefficients (MFCC) and two-dimensional convolutional neural network (2D-CNN) to extract frequency-domain deep features. The features were fused using adaptive weighting strategy for arrhythmia classification. The paper used the arrhythmia database jointly developed by the Massachusetts Institute of Technology and Beth Israel Hospital (MIT-BIH) and evaluated the algorithm under the inter-patient paradigm. Experimental results demonstrated that the proposed algorithm achieved an average precision of 75.2%, an average recall of 70.1% and an average F ₁-score of 71.3%, demonstrating high classification accuracy and being able to provide algorithmic support for arrhythmia classification in wearable devices.
Humans ; Arrhythmias, Cardiac/diagnosis* ; Algorithms ; Electrocardiography/methods* ; Neural Networks, Computer ; Signal Processing, Computer-Assisted ; Deep Learning ; Classification Algorithms

Objective To address the challenges of poor performance and lack of interpretability in existing models,the SHAP algorithm is used to develop an interpretable machine learning model that offers a novel approach to wound age estimation,Methods Based on the previous discovery of the expression of 35 wound age healing-related genes in contused skeletal muscle,the woun age estimaton model was constructed using four algorithms,namly,Multilayer Perceptron(MLP),Random Forest(RF),LightGBM(LGBM),and Support Vector Machine(SVM).The SHAP(Shapley Additive Explanation)algorithm was used to rank the importance of genetic features,eliminate redundant attributes,and optimize the model for accurate wound age estimation.the genetic features of the optimal model were analyzed using SHAP's local interpretation capabilities.Results The best results were obtained using model of MLP(area under the curve(AUC)=0.99)The wound ages were classified into four categories:4～12 h,16～24 h,28～36 h,and 40～48 h,using only 15 gene features.According to SHAP analysis,Fam210a was identified as the most relevant gene.Local analysis revealed that high expression of Fam210a contributed to an increase in the predicted probability of 4 h～12 h,while high expression of Rae1 contributed to an increase in the predicted probability of 16 h～24 h.Additionally,low expression of Tbx18 contributed to an increase in the predicted probability of 28 h～36 h,whereas high expression of Tbx18 contributed to an increase in the predicted probability of 40 h～48 h.Conclusions The combined MLP and SHAP model can be used to predict wound age.Using the SHAP interpreter can better understand the degree of contribution of feature genes to the model prediction,and lay the foundation for further in-depth study of wound age estimation.

Objective:To investigate the predictive value of neutrophil percentage-to-albumin ratio (NPAR) on the outcome after intravenous thrombolysis (IVT) in elderly patients with acute ischemic stroke (AIS).Methods:Elderly patients with AIS who received IVT in Fuyang People's Hospital from October 2021 to September 2023 were retrospectively included. Clinical outcome were assessed by the modified Rankin Scale at 90 days after onset, with a score of >2 defined as poor outcome. Multivariate logistic regression analysis was used to determine the association between NPAR and poor clinical outcome after IVT in elderly AIS patients. Receiver operating characteristic (ROC) curves were used to evaluate the predictive value of NPAR for poor outcome. Results:A total of 148 patients were included, including 86 males (58.1%), aged (74.11±6.17) years. The median baseline National Institutes of Health Stroke Scale (NIHSS) score was 5 (interquartile range: 3-8), and the NPAR was 1.58±0.30. The neutrophil count, neutrophil percentage, NPAR, fasting blood glucose and baseline NIHSS score in the poor outcome group were significantly higher than those in the good outcome group (all P<0.05). Multivariate logistic regression analysis showed that higher baseline NPAR (odds ratio [ OR] 2.659, 95% confidence interval [ CI] 1.117-5.324; P<0.001), NIHSS score ( OR 1.191, 95% CI 1.083-1.309; P<0.001) and fasting blood glucose ( OR 1.224, 95% CI 1.013-1.479; P=0.037) were independent risk factors for poor outcome. ROC curve analysis showed that the area under the curve for NPAR to predict poor outcome was 0.712 (95% CI 0.613-0.812; P<0.001), the optimal cut-off value was 1.728, and the predictive sensitivity and specificity were 65.1% and 75.2%, respectively. Conclusion:Higher baseline NPAR may be a predictor of poor outcome after IVT in elderly AIS patients.

Objective To construct and validate a risk prediction model for delayed healing of venous leg ulcer(VLU),so as to provide a reference basis for early identification of people at high risk of delayed healing.Methods Using a convenience sampling method,331 VLU patients attending vascular surgery departments in 2 tertiary A hospitals in Sichuan Province from January 2018 to December 2022 were selected as a modeling group and an internal validation group,and 112 patients admitted to another tertiary A hospital were selected as an external validation group.Risk factors for delayed healing in VLU patients were screened using univariate analysis,LASSO regression,and multivariate logistic regression analysis,and a risk prediction model was constructed using R software,and the predictive effects of the models were examined using the area under the receiver operating characteristic curve,the Hosmer-Lemeshow test,decision curve,and the bootstrap resampling for internal validation and spatial external validation were performed,respectively.Results The predictors that ultimately entered the prediction model were diabetes(OR=4.752),deep vein thrombosis(OR=4.104),lipodermatosclerosis(OR=5.405),ulcer recurrence(OR=3.239),and ankle mobility(OR=5.520).The model had good discrimination(AUC:0.819 for internal validation and 0.858 for external validation),calibration(Hosmer-Lemeshow test:χ2=13.517,P=0.095 for internal validation and χ2=3.375,P=0.909 for external validation)and clinical validity.Conclusion The model constructed in this study has good differentiation and calibration,and it can effectively predict people at high risk of delayed healing of VLU,which facilitates targeted clinical interventions to improve ulcer outcomes and reduce the risk of delayed ulcer healing.

Objective:To validate the efficacy of chimeric antigen receptor T cells targeting CD7 (CD7 CAR-T cells) modified with protein blocking technology and analyze whether pretreatment with dasatinib can enhance CD7 CAR-T killing ability or reverse the depletion phenotype.Methods:Green fluorescent protein (GFP)-labeled tumor cells were co-incubated with CD7 CAR-T cells or T cells at different potency-to-target ratios, but the culture volume and the numbers of CAR-T/T cells were same. The number of tumor cells was detected using flow cytometry. The killing effect of CAR-T cells on tumor cells was evaluated. A mouse model of acute T-lymphoblastic leukemia (T-ALL) was constructed by injecting 1×10 6 luciferase-expressing CCRF-CEM cells into the mouse tail vein to evaluate the therapeutic effect of CD7 CAR-T cells. Results:CD7 CAR-T cells had a significant killing effect on CCRF-CEM and Jurkat cells, but not on CD7-negative NALM6 cells. The mice in the group receiving CD7 CAR-T cells had a significantly reduced in vivo tumor load and a significantly prolonged survival time as compared with the mice in the group receiving untransduced T cells ( P<0.05). Dasatinib pretreatment significantly reversed the depletion phenotype of CD7 CAR-T cells ( P<0.05) and had no adverse effects on the killing effect and the proliferation of the cells. Conclusions:Protein-blocking technology-modified CD7 CAR-T cells are protected from killing each other, and pretreatment with dasatinib is expected to improve the efficacy and durability of CD7 CAR-T cells.