In order to realize the quantitative assessment of muscle strength in hand function rehabilitation and then formulate scientific and effective rehabilitation training strategies, this paper constructs a multi-scale convolutional neural network (MSCNN) - convolutional block attention module (CBAM) - bidirectional long short-term memory network (BiLSTM) muscle strength prediction model to fully explore the spatial and temporal features of the data and simultaneously suppress useless features, and finally achieve the improvement of the accuracy of the muscle strength prediction model. To verify the effectiveness of the model proposed in this paper, the model in this paper is compared with traditional models such as support vector machine (SVM), random forest (RF), convolutional neural network (CNN), CNN - squeeze excitation network (SENet), MSCNN-CBAM and MSCNN-BiLSTM, and the effect of muscle strength prediction by each model is investigated when the hand force application changes from 40% of the maximum voluntary contraction force (MVC) to 60% of the MVC. The research results show that as the hand force application increases, the effect of the muscle strength prediction model becomes worse. Then the ablation experiment is used to analyze the influence degree of each module on the muscle strength prediction result, and it is found that the CBAM module plays a key role in the model. Therefore, by using the model in this article, the accuracy of muscle strength prediction can be effectively improved, and the characteristics and laws of hand muscle activities can be deeply understood, providing assistance for further exploring the mechanism of hand functions.
ObjectiveTo explore the independent risk factors for benign and malignant subsolid pulmonary nodules and establish a malignant probability prediction model.MethodsA retrospective analysis was performed in 443 patients with subsolid pulmonary nodules admitted to Subei People's Hospital of Jiangsu Province from 2014 to 2018 with definite pathological findings. The patients were randomly divided into a modeling group and a validation group. There were 296 patients in the modeling group, including 125 males and 171 females, with an average age of 55.9±11.1 years. There were 147 patients in the verification group, including 68 males and 79 females, with an average age of 56.9±11.6 years. Univariate and multivariate analysis was used to screen the independent risk factors for benign and malignant lesions of subsolid pulmonary nodules, and then a prediction model was established. Based on the validation data, the model of this study was compared and validated with Mayo, VA, Brock and PKUPH models.ResultsUnivariate and multivariate analysis showed that gender, consolidation/tumor ratio (CTR), boundary, spiculation, lobulation and carcinoembryonic antigen (CEA) were independent risk factors for the diagnosis of benign and malignant subsolid pulmonary nodules. The prediction model formula for malignant probability was: P=ex/(1+ex). X=0.018+(1.436×gender)+(2.068×CTR)+(−1.976×boundary)+ (2.082×spiculation)+(1.277×lobulation)+(2.296×CEA). In this study, the area under the curve was 0.856, the sensitivity was 81.6%, the specificity was 75.6%, the positive predictive value was 95.4%, and the negative predictive value was 39.8%. Compared with the traditional model, the predictive value of this model was significantly better than that of Mayo, VA, Brock and PKUPH models.ConclusionCompared with Mayo, VA, Brock and PKUPH models, the predictive value of the model is more ideal and has greater clinical application value, which can be used for early screening of subsolid nodules.
ObjectiveTo explore the combined application of neutrophil to lymphocyte ratio (NLR) and systemic immune inflammation index (SII) on the prognosis of hepatitis B-related hepatocellular carcinoma after resection.MethodsRetrospectively collected data of 180 patients with hepatitis B-related hepatocellular carcinoma who were hospitalized in the Department of Infectious Diseases and Hepatobiliary Surgery of the Affiliated Hospital of Southwest Medical University and received surgical treatment from January 2013 to December 2019, including general information, laboratory examination and abdominal CT or MRI results. NLR and SII values were measured at one week before operation, and their critical values of NLR and SII were determined by ROC curve analysis. Univariate and multivariate analysis were performed to determine the risk factors to predict the survival status of patients with hepatitis B-related hepatocellular carcinoma after hepatectomy.ResultsUnivariate analysis showed that AFP, platelets, TNM staging, portal vein tumor thrombus, tumor differentiation, NLR, SII, and NLR+SII combined score were significantly correlated with the prognosis of patients with hepatitis B-related hepatocellular carcinoma (P<0.05). Multivariate analysis showed that PLT [HR=1.791, 95%CI (1.124, 2.854), P=0.014], NLR [HR=4.289, 95%CI (2.571, 7.156), P<0.001], SII [HR=5.317, 95%CI (3.016, 9.374), P<0.001], and NLR+SII combined score [HR=7.901, 95%CI (4.124, 15.138), P<0.001] were independently correlated with the survival of patients with hepatitis B-related hepatocellular carcinoma.ConclusionsThe preoperative NLR+SII combined score can be used to evaluate the postoperative prognosis of patients with hepatitis B-related hepatocellular carcinoma. The higher the score, the lower the postoperative survival rate.
ObjectiveTo explore the risk factors affecting the prognosis of patients with metastatic breast cancer (MBC) and construct a nomogram survival prediction model.MethodsThe patients with MBC from 2010 to 2013 were collected from surveillance, epidemiology, and end results (SEER) database, then were randomly divided into training group and validation group by R software. SPSS software was used to compare the survival and prognosis of MBC patients with different metastatic sites in the training group by log-rank method and construct the Kaplan-Meier survival curve. The Cox proportional hazards model was used to analyze the factors of 3-year overall survival, then construct a nomogram survival prediction model by the independent prognostic factors. The C-index was used to evaluate its predictive value and the calibration curve was used to verify the nomogram survival prediction model by internal and external calibration graph.ResultsA total of 3 288 patients with MBC were collected, including 2 304 cases in the training group and 984 cases in the validation group. The data of the two groups were comparable. The median follow-up time of training group and validation group was 34 months and 34 months, respectively. In the training group, the results of Cox proportional hazards model showed that the older, black race, higher histological grading, without operation, ER (–), PR (–), HER-2 (–), and metastases of bone, brain, liver and lung were the risk factors of survival prognosis (P<0.05) and constructed the nomogram survival prediction model with these independent prognostic factors. The nomogram survival prediction showed a good accuracy with C-index of 0.704 [95%CI (0.691, 0.717)] in internal validation (training group) and C-index of 0.691 [95%CI (0.671, 0.711)] in external validation (validation group) in predicting 3-year overall survival. All calibration curves showed excellent consistency.ConclusionNomogram for predicting 3-year overall survival of patients with MBC in this study has a good predictive capability, and it is conducive to development of individualized clinical treatment.
In intensive care units (ICU), the occurrence of acute hypotensive episodes (AHE) is the key problem for the clinical research and it is meaningful for clinical care if we can use appropriate computational technologies to predict the AHE. In this study, based on the records of patients in ICU from the MIMICⅡclinical data, the chaos signal analysis method was applied to the time series of mean artery pressure, and then the patient's Lyapunov exponent curve was drawn ultimately. The research showed that a curve mutation appeared before AHE symptoms took place. This is powerful and clear basis for AHE determination. It is also expected that this study may offer a reference to research of AHE theory and clinical application.
Pathological images of gastric cancer serve as the gold standard for diagnosing this malignancy. However, the recurrence prediction task often encounters challenges such as insignificant morphological features of the lesions, insufficient fusion of multi-resolution features, and inability to leverage contextual information effectively. To address these issues, a three-stage recurrence prediction method based on pathological images of gastric cancer is proposed. In the first stage, the self-supervised learning framework SimCLR was adopted to train low-resolution patch images, aiming to diminish the interdependence among diverse tissue images and yield decoupled enhanced features. In the second stage, the obtained low-resolution enhanced features were fused with the corresponding high-resolution unenhanced features to achieve feature complementation across multiple resolutions. In the third stage, to address the position encoding difficulty caused by the large difference in the number of patch images, we performed position encoding based on multi-scale local neighborhoods and employed self-attention mechanism to obtain features with contextual information. The resulting contextual features were further combined with the local features extracted by the convolutional neural network. The evaluation results on clinically collected data showed that, compared with the best performance of traditional methods, the proposed network provided the best accuracy and area under curve (AUC), which were improved by 7.63% and 4.51%, respectively. These results have effectively validated the usefulness of this method in predicting gastric cancer recurrence.
ObjectiveTo reveal the scientific output and trends in pulmonary nodules/early-stage lung cancer prediction models. MethodsPublications on predictive models of pulmonary nodules/early lung cancer between January 1, 2002 and June 3, 2023 were retrieved and extracted from CNKI, Wanfang, VIP and Web of Science database. CiteSpace 6.1.R3 and VOSviewer 1.6.18 were used to analyze the hotspots and theme trends. ResultsA marked increase in the number of publications related to pulmonary nodules/early-stage lung cancer prediction models was observed. A total of 12581 authors from 2711 institutions in 64 countries/regions published 2139 documents in 566 academic journals in English. A total of 282 articles from 1256 authors were published in 176 journals in Chinese. The Chinese and English journals which published the most pulmonary nodules/early-stage lung cancer prediction model-related papers were Journal of Clinical Radiology and Frontiers in Oncology, respectively. Chest was the most frequently cited journal. China and the United States were the leading countries in the field of pulmonary nodules/early-stage lung cancer prediction models. The institutions represented by Fudan University had significant academic influence in the field. Analysis of keywords revealed that multi-omics, nomogram, machine learning and artificial intelligence were the current focus of research. ConclusionOver the last two decades, research on risk-prediction models for pulmonary nodules/early-stage lung cancer has attracted increasing attention. Prognosis, machine learning, artificial intelligence, nomogram, and multi-omics technologies are both current hotspots and future trends in this field. In the future, in-depth explorations using different omics should increase the sensitivity and accuracy of pulmonary nodules/early-stage lung cancer prediction models. More high-quality future studies should be conducted to validate the efficacy and safety of pulmonary nodules/early-stage lung cancer prediction models further and reduce the global burden of lung cancer.
This study introduced the construction of individualized risk assessment model based on Bayesian networks, comparing with traditional regression-based logistic models using practical examples. It evaluates the model's performance and demonstrates its implementation in the R software, serving as a valuable reference for researchers seeking to understand and utilize Bayesian network models.
Keloids are benign skin tumors resulting from the excessive proliferation of connective tissue in wound skin. Precise prediction of keloid risk in trauma patients and timely early diagnosis are of paramount importance for in-depth keloid management and control of its progression. This study analyzed four keloid datasets in the high-throughput gene expression omnibus (GEO) database, identified diagnostic markers for keloids, and established a nomogram prediction model. Initially, 37 core protein-encoding genes were selected through weighted gene co-expression network analysis (WGCNA), differential expression analysis, and the centrality algorithm of the protein-protein interaction network. Subsequently, two machine learning algorithms including the least absolute shrinkage and selection operator (LASSO) and the support vector machine-recursive feature elimination (SVM-RFE) were used to further screen out four diagnostic markers with the highest predictive power for keloids, which included hepatocyte growth factor (HGF), syndecan-4 (SDC4), ectonucleotide pyrophosphatase/phosphodiesterase 2 (ENPP2), and Rho family guanosine triphophatase 3 (RND3). Potential biological pathways involved were explored through gene set enrichment analysis (GSEA) of single-gene. Finally, univariate and multivariate logistic regression analyses of diagnostic markers were performed, and a nomogram prediction model was constructed. Internal and external validations revealed that the calibration curve of this model closely approximates the ideal curve, the decision curve is superior to other strategies, and the area under the receiver operating characteristic curve is higher than the control model (with optimal cutoff value of 0.588). This indicates that the model possesses high calibration, clinical benefit rate, and predictive power, and is promising to provide effective early means for clinical diagnosis.
Objective To investigate the correlation between monocyte-lymphocyte ratio (MLR) and intensive care unit (ICU) results in ICU hospitalized patients. Methods Clinical data were extracted from Medical Information Mart for Intensive Care Ⅲ database, which contained health data of more than 50000 patients. The main result was 30-day mortality, and the secondary result was 90-day mortality. The Cox proportional hazards model was used to reveal the association between MLR and ICU results. Multivariable analyses were used to control for confounders. Results A total of 7295 ICU patients were included. For the 30-day mortality, the hazard ratio (HR) and 95% confidence interval (CI) of the second (0.23≤MLR<0.47) and the third (MLR≥0.47) groups were 1.28 (1.01, 1.61) and 2.70 (2.20, 3.31), respectively, compared to the first group (MLR<0.23). The HR and 95%CI of the third group were still significant after being adjusted by the two different models [2.26 (1.84, 2.77), adjusted by model 1; 2.05 (1.67, 2.52), adjusted by model 2]. A similar trend was observed in the 90-day mortality. Patients with a history of coronary and stroke of the third group had a significant higher 30-day mortality risk [HR and 95%CI were 3.28 (1.99, 5.40) and 3.20 (1.56, 6.56), respectively]. Conclusion MLR is a promising clinical biomarker, which has certain predictive value for the 30-day and 90-day mortality of patients in ICU.