ObjectiveThe aim of this study was to investigate the value of Artificial Neural Networks (ANNs) in predicting the occurrence of Venous Thromboembolism (VTE) in patients with Obstructive Sleep Apnea (OSA), and to compare it with traditional Logistic regression models to assess its predictive efficacy, providing theoretical basis for the prediction of VTE risk in OSA patients. MethodsA retrospective analysis was conducted on patients diagnosed with OSA and hospitalized in the Department of Respiratory and Critical Care Medicine, Second Affiliated Hospital of Kunming Medical University, from January 2018 to August 2023. Patients were divided into OSA combined with VTE group (n=128) and pure OSA control group (n=680). The dataset was randomly divided into a training set (n=646) and an independent validation set (n=162). The Synthetic Minority Oversampling Technique (SMOTE) was employed to address the issue of data imbalance. Artificial Neural Networks and Logistic regression models were then built on training sets with and without SMOTE. Finally, the performance of each model was evaluated using accuracy, sensitivity, specificity, Youden's index, and Area Under the Receiver Operating Characteristic Curve (AUC). Results When oversampling was conducted using SMOTE on the training set, both the Artificial Neural Network and Logistic regression models showed improved AUC. The Artificial Neural Network model with SMOTE performed the best with an AUC value of 0.935 (95%CI: 0.898–0.961), achieving an accuracy of 90.15%, specificity of 87.32%, sensitivity of 93.44%, and Youden’s index of 0.808 at the optimal cutoff point. The Logistic regression model with SMOTE yielded an AUC value of 0.817 (95%CI: 0.765–0.861), with an accuracy of 77.27%, specificity of 83.80%, sensitivity of 69.67%, and Youden's index of 0.535. The difference in AUC between the Artificial Neural Network model and Logistic regression model was statistically significant after employing SMOTE (P<0.05). Conclusions The Artificial Neural Network model demonstrates high effectiveness in predicting VTE formation in OSA patients, particularly with the further improvement in predictive performance when utilizing SMOTE oversampling technique, rendering it more accurate and stable compared to the traditional Logistic regression model.
This review systematically analyzes recent research progress in multimodal fusion techniques for medical imaging classification, focusing on various fusion strategies and their effectiveness in classification tasks. Studies indicate that multimodal fusion methods significantly enhance classification performance and demonstrate potential in clinical decision support. However, challenges remain, including insufficient dataset sharing, limited utilization of text modalities, and inadequate integration of fusion strategies with medical knowledge. Future efforts should focus on developing large-scale public datasets and optimizing deep fusion strategies for image and text modalities to promote broader application in medical scenarios.
Objective To analyze the preoperative risk factors of atrial fibrillation (AF) in patients with coronary artery disease after coronary artery bypass grafting (CABG). Methods From September 2007 to April 2008, the clinical information of 226 patients who underwent onpump coronary artery bypass grafting(CABG)or offpump coronary artery bypass grafting(OPCAB) was collected. The patients were divided into nonAF group and AF group according to whether AF lasted more than 5 mins in 3 days after operation. Ultrasonic cardiography (UCG) and clinical information of preoperation in two groups were analyzed. Results Twentyfour（10.6%） patients had AF after operation. There were more patients whose left atrial diameter gt;35 mm in AF group than that in nonAF group ［41.7%（10）vs. 22.3% (45),χ2=4.380, P=0.036）］, more patients had mitral regurgitation in AF group than that in nonAF group ［37.5%（9） vs. 17.3% (35),χ2=5.568, P=0.018）］, more patients had left main coronary artery involvement in AF group than that in nonAF group ［33.3% (8) vs.12.4% (25),χ2=7.560,P=0.006］, and patients in AF group were older than those in nonAF group ［65.7±9.5 years vs. 60.1±10.1 years,t=-2.724,P=0.010］. In univariate analysis, in terms of preoperative clinical indexs such as the aged, mitral regurgitation, left atrial diameter, left mainm coronary artery involvement, and postoperative clinical indexs such as ventilatory time （χ2=4.190，P=0.040）, electrocardiogram (ECG) monitoring time（χ2=5.948，P=0.015）, hospitalization expense（χ2=4.110，P=0.043）, there were significant differences between 2 groups. Conclusion Risk factors such as the aged, mitral regurgitation, left atrial diameter and left main coronary artery involvement are related to AF after CABG. Clinical index, ECG and echocardiography are helpful to predict AF, and can provide better prevention and treatment, and reduce the rate of AF.
ObjectiveTo analyze the clinical outcomes of cardiac surgery in patients after kidney transplantation underwent immunosuppression therapy.
Methods Clinical data of 6 patients (including 4 males and 2 females aged from 27 to 66 years) undergoing cardiac surgery after kidney transplantation with immunosuppression between January 2011 and October 2013 in Beijing Anzhen Hospital were retrospectively analyzed. Two patients underwent off pump coronary artery bypass grafting, 3 patients underwent aortic valve replacement and 1 patient underwent mitral valve replacement. The interval between kidney transplantation and cardiac operation was 8.5±1.7 years (range, 6-10 years).
ResultsAll the operations were performed successfully without in-hospital death and complications. Operation time was 237.0±93.0 min. Cardiopulmonary bypass (CPB) time was 101.3±16.7 min. Aortic cross-clamping time was 75.6±9.7 min. in-hospital stay was 17±6 d. There was a statistical difference in preoperative and postoperative left ventricle ejection fraction (63.5%±4.5% vs. 56.5%±5.8%, P < 0.05), as well as preoperative and postoperative left ventricular enddiastolic diameter (54.5±8.5 mm vs. 43.7±6.8 mm, P < 0.05), but there was no statistical difference in preoperative and postoperative serum creatinine levels (103.7±15.1μmol/L vs. 106.6±34.7μmol/L, P > 0.05). The mean follow-up time of the 6 patients was 4-15 months. The follow-up rate was 100%. All the patients were in cardiac function NYHA classⅠ-Ⅱ. And the quality of life was improved.
ConclusionCardiac operations for kidney transplant recipients undergoing immunosuppressive treatment are effective and safe as long as appropriate perioperative treatment are taken.
ObjectiveTo provide experimental data and theoretical support for further studying the maturity of cardiac patches in other in vitro experiments and the safety in other in vivo animal experiments, through standard chemically defined and small molecule-based induction protocol (CDM3) for promoting the differentiation of human induced pluripotent stem cells (hiPSCs) into myocardium, and preliminarily preparing cardiac patches. MethodsAfter resuscitation, culture and identification of hiPSCs, they were inoculated on the matrigel-coated polycaprolactone (PCL). After 24 hours, the cell growth was observed by DAPI fluorescence under a fluorescence microscope, and the stemness of hiPSCs was identified by OCT4 fluorescence. After fixation, electron microscope scanning was performed to observe the cell morphology on the surface of the patch. On the 1st, 3rd, 5th, and 7th days of culture, the cell viability was determined by CCK-8 method, and the growth curve was drawn to observe the cell growth and proliferation. After co-cultured with matrigel-coated PCL for 24 hours, hiPSCs were divided into a control group and a CDM3 group, and continued to culture for 6 days. On the 8th day, the cell growth was observed by DAPI fluorescence under a fluorescence microscope, and hiPSCs stemness was identified by OCT4 fluorescence, and cTnT and α-actin for cardiomyocyte marker identification. ResultsImmunofluorescence of hiPSCs co-cultured with matrigel-coated PCL for 24 hours showed that OCT4 emitted green fluorescence, and hiPSCs remained stemness on matrigel-coated PCL scaffolds. DAPI emitted blue fluorescence: cells grew clonally with uniform cell morphology. Scanning electron microscope showed that hiPSCs adhered and grew on matrigel-coated PCL, the cell outline was clearly visible, and the morphology was normal. The cell viability assay by CCK-8 method showed that hiPSCs proliferated and grew on PCL scaffolds coated with matrigel. After 6 days of culture in the control group and the CDM3 group, immunofluorescence showed that the hiPSCs in the control group highly expressed the stem cell stemness marker OCT4, but did not express the cardiac markers cTnT and α-actin. The CDM3 group obviously expressed the cardiac markers cTnT and α-actin, but did not express the stem cell stemness marker OCT4. ConclusionhiPSCs can proliferate and grow on matrigel-coated PCL. Under the influence of CDM3, hiPSCs can be differentiated into cardiomyocyte-like cells, and the preliminary preparation of cardiac patch can provide a better treatment method for further clinical treatment of cardiac infarction.
