Search

BackgroundPredictions of cardiovascular disease risks based on health records have long attracted broad research interests. Despite extensive efforts, the prediction accuracy has remained unsatisfactory. This raises the question as to whether the data insufficiency, statistical and machine-learning methods, or intrinsic noise have hindered the performance of previous approaches, and how these issues can be alleviated. ObjectiveBased on a large population of patients with hypertension in Shenzhen, China, we aimed to establish a high-precision coronary heart disease (CHD) prediction model through big data and machine-learning MethodsData from a large cohort of 42,676 patients with hypertension, including 20,156 patients with CHD onset, were investigated from electronic health records (EHRs) 1-3 years prior to CHD onset (for CHD-positive cases) or during a disease-free follow-up period of more than 3 years (for CHD-negative cases). The population was divided evenly into independent training and test datasets. Various machine-learning methods were adopted on the training set to achieve high-accuracy prediction models and the results were compared with traditional statistical methods and well-known risk scales. Comparison analyses were performed to investigate the effects of training sample size, factor sets, and modeling approaches on the prediction performance. ResultsAn ensemble method, XGBoost, achieved high accuracy in predicting 3-year CHD onset for the independent test dataset with an area under the receiver operating characteristic curve (AUC) value of 0.943. Comparison analysis showed that nonlinear models (K-nearest neighbor AUC 0.908, random forest AUC 0.938) outperform linear models (logistic regression AUC 0.865) on the same datasets, and machine-learning methods significantly surpassed traditional risk scales or fixed models (eg, Framingham cardiovascular disease risk models). Further analyses revealed that using time-dependent features obtained from multiple records, including both statistical variables and changing-trend variables, helped to improve the performance compared to using only static features. Subpopulation analysis showed that the impact of feature design had a more significant effect on model accuracy than the population size. Marginal effect analysis showed that both traditional and EHR factors exhibited highly nonlinear characteristics with respect to the risk scores. ConclusionsWe demonstrated that accurate risk prediction of CHD from EHRs is possible given a sufficiently large population of training data. Sophisticated machine-learning methods played an important role in tackling the heterogeneity and nonlinear nature of disease prediction. Moreover, accumulated EHR data over multiple time points provided additional features that were valuable for risk prediction. Our study highlights the importance of accumulating big data from EHRs for accurate disease predictions.

The adenosine monophosphate (AMP)-activated protein kinase (AMPK) sits at a central node in the regulation of energy metabolism and tumor progression. AMPK is best known to sense high cellular ADP or AMP levels, which indicate the depletion of energy stores. Previous studies have shown that the low expression of phosphorylated AMPK is associated with a poor prognosis of pancreatic cancer. In this study, we report that AMPK is also highly sensitive to extracellular matrix (ECM) stiffness. We found that AMPK is activated in cells when cultured under low ECM stiffness conditions and is functionally required for the metabolic switch induced by ECM stiffness. This regulation of AMPK requires the Hippo kinases but not LKB1/CaMKKβ. Hippo kinases directly phosphorylate AMPKα at Thr172 to activate AMPK at low ECM stiffness. Furthermore, we found AMPK activity is inhibited in patients with pancreatic ductal adenocarcinoma (PDAC) with high ECM stiffness and is associated with a poor survival outcome. The activation of Hippo kinases by ROCK inhibitor Y-27632 in combination with the mitochondrial inhibitor metformin synergistically activates AMPK and dramatically inhibits PDAC growth. Together, these findings establish a novel model for AMPK regulation by the mechanical properties of ECMs and provide a rationale for simultaneously targeting the ECM stiffness-Hippo kinases-AMPK signaling and low glucose-LKB1-AMPK signaling pathways as an effective therapeutic strategy against PDAC.

Wastewater-based epidemiology (WBE) has become a valuable tool for monitoring the prevalence of SARS-CoV-2 on university campuses. However, concerns about effectiveness of raw sewage as a COVID-19 early warning system still exist, and its not clear how useful normalization by simultaneous comparison of Pepper Mild Mottle Virus (PMMoV) is in addressing variations resulting from fecal discharge dilution. This study aims to contribute insights into these aspects by conducting an academic-year field trial at the student residences on the University of Tennessee, Knoxville campus, raw sewage. This was done to investigate the correlations between SARS-CoV-2 RNA load, both with and without PMMoV normalization, and various parameters, including active COVID-19 cases, self-isolations, and their combination among all student residents. Significant positive correlations between SARS-CoV-2 RNA load a week prior, during the monitoring week, and the subsequent week with active cases. Despite these correlations, normalization by PMMoV does not enhance these associations. These findings suggest the potential utility of SARS-CoV-2 RNA load as an early warning indicator and provide valuable insights into the application and limitations of WBE for COVID-19 surveillance specifically within the context of raw sewage on university campuses.

Abstract Objective To develop and validate a machine learning model for predicting mortality-associated prognostic factors in order to reduce in-hospital mortality rates among HIV/AIDS patients with Cryptococcus infection in Guangxi, China. Methods This retrospective prognostic study included HIV/AIDS patients with cryptococcosis in the Fourth People’s Hospital of Nanning from October 2011 to June 2019. Clinical features were extracted and used to train ten machine learning models, including Logistic Regression, KNN, DT, RF, Adaboost, Xgboost, LightGBM, Catboost, SVM, and NBM, to predict the outcome of HIV patients with cryptococcosis infection. The sensitivity, specificity, AUC, and F1 value were applied to assess model performance in both the testing and training sets. The optimal model was selected and interpreted. Results A total of 396 patients were included in the study. The average in-hospital mortality of HIV/AIDS patients with cryptococcosis was 12.9% from 2012 to 2019. After feature screening, 20 clinical features were selected for model construction, accounting for 93.8%, including ART, Electrolyte disorder, Anemia, and 17 laboratory tests. The RF model (AUC 0.9787, Sensitivity 0.9535, Specificity 0.8889, F1 0.7455) and the SVM model (AUC 0.9286, Sensitivity 0.7907, Specificity 0.9786, F1 0.8293) had excellent performance. The SHAP analysis showed that the primary risk factors for prognosis prediction were identified as BUN/CREA, Electrolyte disorder, NEUT%, Urea, and IBIL. Conclusions RF and SVM machine learning models have shown promising predictive abilities for the prognosis of hospitalized HIV/AIDS patients with cryptococcosis, which can aid clinical assessment and treatment decisions for patient prognosis.

Abstract Purpose The essence of this scholarly work was to carefully outline the key factors intensifying the virulence and protracted contagion of COVID-19, particularly among individuals afflicted with hematologic malignancies (HM), in an epoch predominantly governed by the Omicron variant. Methods Adults with HM diagnosed with COVID-19 from November 2022 to February 2023 were monitored in this retrospective study. Patient blood samples yielded biochemical data, and COVID-19 was confirmed through RNA or antigen testing. The factors affecting severity and infection duration were examined using both univariate and multivariate logistic regression analyses. For calculating the overall survival probabilities, the Kaplan–Meier product limit approach was employed. Results In the examined cohort, 133 individuals diagnosed with HM and concomitantly infected with COVID-19 were scrutinized. Of the participants, 29.3% (39 patients) were classified as Severe/Critical, while the other 70.7% (94 patients) were categorized as Non-severe. A significant difference was observed in vaccination status: 61.7% of patients in the Non-severe group had received at least a two-dose vaccine regimen, whereas 61.5% of the Severe/Critical group had either minimal or only one dose of vaccination. The data analysis revealed that elevated C-reactive protein levels (≥ 100 mg/L) significantly raised the risk of severe/critical conditions in HM patients with COVID-19, as determined by advanced multivariate logistic regression. The odds ratio was 3.415 with a 95% confidence interval of 1.294–9.012 (p = 0.013). Patients who continued to have positive nucleic acid tests and ongoing symptoms beyond 30 days were categorized as having a persistent infection, whereas those who achieved infection control within this timeframe were categorized as having infection recovery. Of the HM cohort, 11 did not survive beyond 30 days after diagnosis. The results from a competing risk model revealed that increased interleukin-6 levels (HR: 2.626, 95% CI: 1.361–5.075; p = 0.004) was significantly associated with persistent infection. Conversely, receiving more than two vaccine doses (HR: 0.366, 95% CI: 0.158–0.846; p = 0.019), and having high IgG levels (≥ 1000 mg/dl) (HR: 0.364, 95% CI: 0.167–0.791; p = 0.011), were associated with infection recovery. There was a notable disparity in survival rates between patients with persistent infections and infection recovery, with those in the non-persistent group demonstrating superior survival outcomes (P

Abstract SHP-1, a nonreceptor protein tyrosine phosphatase encoded by ptpn6, has been regarded as a regulatory protein of hematopoietic cell biology for years. However, there is now increasing evidence to support its role in tumors. Thus, the role of ptpn6 for prognosis and immune regulation across 33 tumors was investigated, aiming to explore its functional heterogeneity and clinical significance in pan-cancer. Differential expression of ptpn6 was found between cancer and adjacent normal tissues, and its expression was significantly correlated with the prognosis of tumor patients. In most cancers, ptpn6 expression was significantly associated with immune infiltration. This was further confirmed by ptpn6-related genes/proteins enrichment analysis. Additionally, genetic alterations in ptpn6 was observed in most cancers. As for epigenetic changes, it’s phosphorylation levels significantly altered in 6 tumors, while methylation levels significantly altered in 12 tumors. Notably, the methylation levels of ptpn6 were significantly decreased in 11 tumors, accompanied by its increased expression in 8 of them, suggesting that the hypomethylation may be related to its increased expression. Our results show that ptpn6 plays a specific role in tumor immunity and exerts a pleiotropic effect in a variety of tumors. It can serve as a prognostic factor for some cancers. Especially in LGG, KIRC, UCS and TGCT, the increased expression of ptpn6 is associated with poor prognosis and high immune infiltration. This aids in understanding the role of ptpn6 in tumor biology, and can provide insight into presenting a potential biomarker for poor prognosis and immune infiltration in cancers.

Abstract Background Migraine and insomnia are prevalent conditions that often co-occur, each exacerbating the other and substantially impacting the quality of life. The locus coeruleus (LC), a brainstem region responsible for norepinephrine synthesis, participates in pain modulation, sleep/wake cycles, and emotional regulation, rendering it a potential nexus in the comorbidity of migraine and insomnia. Disruptions in the LC-noradrenergic system have been hypothesized to contribute to the comorbidities of migraine and insomnia, although neuroimaging evidence in humans remains scarce. In this study, we aimed to investigate the intrinsic functional connectivity (FC) network of the LC in patients with comorbid migraine and subjective chronic insomnia and patients with migraine with no insomnia (MnI) using resting-state functional magnetic resonance imaging (rs-fMRI) and seed-based FC analyses. Methods In this cross-sectional study, 30 patients with comorbid migraine and chronic insomnia (MI), 30 patients with MnI, and 30 healthy controls (HCs) were enrolled. Participants underwent neuropsychological testing and rs-fMRI. The LC-FC network was constructed using seed-based voxel-wise FC analysis. To identify group differences in LC-FC networks, voxel-wise covariance analysis was conducted with sex and age as covariates. Subsequently, a partial correlation analysis was conducted to probe the clinical relevance of aberrant LC-FC in patients with MI and MnI. Results Except for the insomnia score, no other significant difference was detected in demographic characteristics and behavioral performance between the MI and MnI groups. Compared with HCs, patients with MI exhibited altered LC-FC in several brain regions, including the dorsomedial prefrontal cortex (DMPFC), anterior cerebellum, dorsolateral prefrontal cortex (DLPFC), thalamus, and parahippocampal gyrus (PHG). Lower FC between the LC and DLPFC was associated with greater insomnia severity, whereas higher FC between the LC and DMPFC was linked to longer migraine attack duration in the MI group. Conclusion Our findings reveal the presence of aberrant LC-FC networks in patients with MI, providing neuroimaging evidence of the interplay between these conditions. The identified LC-FC alterations may serve as potential targets for therapeutic interventions and highlight the importance of considering the LC-noradrenergic system in the management of MI.

Abstract Background This study aims to verify the effectiveness of a deep neural network (DNN) in automatically identifying pulp calcification on cone beam computed tomography (CBCT) images. Methods This study retrospectively analysed 150 CBCT images. Pulp calcification was identified and manually annotated by three endodontists with 10 years of experience. A DNN model based on the U-Net architecture was constructed to identify pulp calcification, and ten rounds of fourfold cross-validation were conducted. The model performance was evaluated using sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (AUC). Results The model achieved a sensitivity of 75.91 ± 2.84% in automatically identifying pulp calcification, with a specificity of 68.88 ± 2.35%, an accuracy of 72.78 ± 2.13%, and an AUC of 73.68 ± 3.09%. According to the ranking for diagnostic tests, the proposed method achieved a “good” grade for sensitivity, accuracy, and AUC and a “fair” grade for specificity. Conclusions The results indicate that the proposed method shows promise for identifying pulp calcification on CBCT images. Future research aims to expand the dataset and refine the model, thereby enhancing its clinical applicability. The integration of artificial intelligence into diagnostic and treatment systems is anticipated to increase the efficiency of diagnosing pulp calcification and assist dentists in assessing the difficulty of root canal treatment cases preoperatively. Clinical registration Registration was performed on the Chinese Clinical Trial Registry ( https://www.chictr.org.cn/ ) (Registration number: ChiCTR2300077078, 27/10/2023) and National Medical Research Registry Information System ( https://www.medicalresearch.org.cn/ , 30/10/2023) (Number: MR-44–23-039664).

Abstract Background HIV/AIDS has emerged as a nationwide epidemic and has taken the forefront position as the primary infectious killer of adults in China. The control and prevention of the disease have been hampered by a weak link in the form of heterosexual transmission. However, conventional intervention measures have demonstrated suboptimal efficacy in reducing the incidence of new HIV infections. In light of the current epidemiological characteristics, we have developed and executed an innovative intervention model known as the Joint Prevention and Control Mechanism of the ‘CDC-Public Security Bureau-NGO’. The purpose of this research is to assess the impact of this model on the AIDS awareness, HIV infection rates, sexual behavior, and associated factors among female sex workers and elderly clients. Through the provision of robust evidence of the efficacy of this innovative model, we seek to advocate for its implementation in future interventions. Methods The research design of this study incorporates both a serial cross-sectional study and time-series analysis from 2014 to 2021, including a 4-year traditional intervention (2014–2017) and the 4-year ‘CDC-Public Security Bureau-NGO’ innovative intervention (2018–2021), was conducted to evaluate the effects of the new intervention. The GM(1, 1) model was performed to predict the proportion of HIV infection without implementing the innovative intervention in 2018–2021; P and C values were used to evaluate the performance of the model. Mann-Kendall test and descriptive methods were used to analyzed the trend of traditional and innovative interventions models on HIV positive detection rate in FSWs and elderly clients. Results The condom usage rates during the last commercial sexual encounter for FSWs and elderly clients improved from 74.9% and 9.1%, respectively, to 96.9% and 28.1%. (P

The pedagogical value and global relevance of heritage education regarding the safeguarding of global tradition and practice for future generations is indisputable. This research aims to investigate the efficacy of integrating a puzzle-based game, with scaffolding strategies, to enhance students' understanding and appreciation of cultural heritage content. In the activity, students are tasked with reconstructing art patterns from cultural heritage artifacts, a process which is designed to bolster their performance in recognizing and appreciating such patterns. A control group, utilizing a non-interactive instructional video with the same content, was also established to contrast and evaluate the benefits of the game-based approach. The evaluation of knowledge acquisition and transfer was conducted through pattern tests, with the participants' eye movements recorded during the testing process. The results show that the participants in the game group exhibited significant advantages in terms of efficiently recognizing fundamental content and effectively transferring pattern recognition. Furthermore, the examination of the visual behavior within the game reveals that the integration of the jigsaw puzzle (game mechanic) and scaffolding (instructional strategy) can enhance the efficacy and cognitive strategies employed by participants in comprehending heritage content. This study develops design methodologies for cultural heritage games that enhances learning and appreciation.

BACKGROUND: To evaluate the value of computed tomography (CT) radiomics in predicting the risk of developing epidermal growth factor receptor (EGFR) T790M resistance mutation for metastatic non-small lung cancer (NSCLC) patients before first-line EGFR-tyrosine kinase inhibitors (EGFR-TKIs) therapy. METHODS: A total of 162 metastatic NSCLC patients were recruited and split into training and testing cohort. Radiomics features were extracted from tumor lesions on nonenhanced CT (NECT) and contrast-enhanced CT (CECT). Radiomics score (rad-score) of two CT scans was calculated respectively. A nomogram combining two CT scans was developed to evaluate T790M resistance within up to 14 months. Patients were followed up to calculate the time of T790M occurrence. Models were evaluated by area under the curve at receiver operating characteristic analysis (ROC-AUC), calibration curve, and decision curve analysis (DCA). The association of the nomogram with the time of T790M occurrence was evaluated by Kaplan-Meier survival analysis. RESULTS: The nomogram constructed with the rad-score of NECT and CECT for predicting T790M resistance within 14 months achieved the highest ROC-AUCs of 0.828 and 0.853 in training and testing cohorts, respectively. The DCA showed that the nomogram was clinically useful. The Kaplan-Meier analysis showed that the occurrence time of T790M difference between the high- and low-risk groups distinguished by the rad-score was significant (p

ObjectivesMultidrug-resistant TB (MDR-TB) is a severe burden and public health threat worldwide. This study aimed to develop a radiomics model based on the tree-in-bud (TIB) sign and nodules and validate its predictive performance for MDR-TB.MethodsWe retrospectively recruited 454 patients with proven active TB from two hospitals and classified them into three training and testing cohorts: TIB (n = 295, 102), nodules (n = 302, 97), and their combination (n = 261, 81). Radiomics features relating to TIB and nodules were separately extracted. The maximal information coefficient and recursive feature elimination were used to select informative features per the two signs. Two radiomics models were constructed to predict MDR-TB using a random forest classifier. Then, a combined model was built incorporating radiomics features based on these two signs. The capability of the models in the combined training and testing cohorts was validated with ROC curves.ResultsSixteen features were extracted from TIB and 15 from nodules. The AUCs of the combined model were slightly higher than those of the TIB model in the combined training cohort (0.911 versus 0.877, p > 0.05) and testing cohort (0.820 versus 0.786, p  0.05) and testing cohort (0.820 versus 0.855, p > 0.05).ConclusionsThe CT-based radiomics models hold promise for use as a non-invasive tool in the prediction of MDR-TB.Clinical relevance statementOur study revealed that complementary information regarding MDR-TB can be provided by radiomics based on the TIB sign and nodules. The proposed radiomics models may be new markers to predict MDR in active TB patients.Key points• This is the first study to build, validate, and apply radiomics based on tree-in-bud sign and nodules for the prediction of MDR-TB. • The radiomics model showed a favorable performance for the identification of MDR-TB. • The combined model holds potential to be used as a diagnostic tool in routine clinical practice.