Health literacy in patients with gout: A latent profile analysis.

OBJECTIVES: Understanding the health literacy status of patients with gout diagnosis is essential for improving the health of this population. Our study aimed to investigate the latent profiles of health literacy in patients with gout and to analyze differences in characteristics across potential profiles. METHODS: This was a cross-sectional study. Eligible participants attended the Shandong Gout Medical Center, from March 2023 to May 2023 and self-reported gout diagnosis. We used the Health Literacy Scale for Patients with Gout designed and validated by our team. The scale had good reliability and validity among patients with gout. 243 patients completed the Demographic Information Questionnaire and the Health Literacy Scale for Patients with Gout. We used latent profile analysis to identify the latent profiles of gout patients' health literacy. We used Chi-square tests with Bonferroni correction to analyze differences in demographics and illness characteristics across identified profiles. RESULTS: Three profiles of patients with gout emerged (prevalence): the low literacy-low critical group (21.81%), the moderate literacy group (42.79%), and the high literacy-stable group (35.39%). The three groups differed in age, education level, monthly income, disease duration, and place of residence (P<0.01). CONCLUSIONS: The health literacy of patients with gout was heterogeneous. Healthcare professionals should adopt targeted interventions based on the characteristics of each latent health literacy profile to improve the health literacy level of patients with gout.

The disease recurrence perception scale for patients with inflammatory bowel disease: Instrument development and cross-sectional validation study.

Disease recurrence perception plays a key role in disease management and subsequent disease recurrence prevention. However, there are no specific tools for assessing disease recurrence perception in patients with inflammatory bowel disease (IBD) characterized by alternating remission and recurrence. To develop and validate an instrument for measuring disease recurrence perception of patients with IBD, the study was conducted in two steps: (1) instrument development and (2) psychometric tests. A total of 623 patients with IBD participated in the study. The common sense model of illness self-regulation (CSM) was used as a framework for instrument development. The administered version contained 48 items intended to be relevant to at least one of the six dimensions of the model. Based on preliminary analyzes, 12 items were deleted leaving 36 items for more detailed psychometric and factor analyzes. The Cronbach's alpha coefficient of the total 36-item instrument was 0.915. The content validity indexes at item and scale levels were satisfactory. The test-retest reliability of the total instrument was 0.870. Exploratory principal components analysis (n = 278) was used to identify six components congruent with intended CSM constructs that accounted for 62.6% of total item variance. Confirmatory factor analysis (n = 345) found acceptable fit for the six factor measurement model (χ2/df = 1.999, GFI = 0.846, NFI = 0.855, IFI = 0.922, TLI = 0.910, CFI = 0.921, RMSEA = 0.054). Overall, the DRPSIBD demonstrated satisfactory reliability and validity to warrant further development as a measure of disease recurrence perception of patients with IBD.

The efficacy of antiviral treatment in chronic hepatitis B patients with hepatic steatosis.

Background & aims: With a drastic increase in the number of chronic hepatitis B (CHB) patients with coexisting nonalcoholic fatty liver disease (NAFLD), there is an urgent need to evaluate antiviral treatment effects in this special population. Methods: CHB patients with hepatic steatosis (CHB + HS) were prospectively recruited with followed-up of 3 years. HS and liver fibrosis were assessed by transient elastography. HS was defined as controlled attenuation parameter (CAP) ≥248 dB/m, and fibrosis progression was defined with ≥1-stage fibrosis increment. Multivariate and propensity score matching (PSM) analysis were used to evaluate antiviral therapy effects on fibrosis progression. Results: In total 212 recruited CHB + HS patients (median age 36 years, median ALT 59 U/L), 49.1% (104/212) received antiviral therapy and 50.9% (108/212) did not. Among patients with antiviral therapy, rates of serum HBV DNA undetectable, HBeAg and HBsAg loss, and ALT normalization at year 3 were 88.5%, 31.0%, 8.7% and 70.2%, respectively. Patients with mild-moderate HS didn't differ patients with severe HS regarding biochemical and virological responses. Antiviral therapy was independently associated with a lower risk of fibrosis progression among the entire cohort (odds ratio 0.473, 95% CI 0.245-0.911, P = 0.025). This finding was further verified by PSM analysis. When stratified by the severity of HS, the antiviral therapy benefits in reducing fibrosis progression were mainly seen in patients with mild-moderate HS. Conclusions: Among CHB + HS patients, long-term antiviral treatment effectively inhibits HBV replication and reduces fibrosis progression. Our findings have implications for the optimal management of this population.

A Deep Multi-Task Network to Learn Tumor Pathological Representations for Lymph Node Metastasis Prediction.

Lymph node metastasis is of paramount importance for patient treatment decision-making, prognosis evaluation, and clinical trial enrollment. However, accurate preoperative diagnosis remains challenging. In this study, we proposed a multi-task network to learn the primary tumor pathological features using the pT stage prediction task and leverage these features to facilitate lymph node metastasis prediction. We conducted experiments using electronic medical record data from 681 patients with non-small cell lung cancer. The proposed method achieved a 0.768 area under the receiver operating characteristic curve (AUC) value with a 0.073 standard deviation (SD) and a 0.448 average precision (AP) value with a 0.113 SD for lymph node metastasis prediction, which significantly outperformed the baseline models. Based on the results, we can conclude that the proposed multi-task method can effectively learn representations about tumor pathological conditions to support lymph node metastasis prediction.

Percutaneous Coronary Intervention in Acute Coronary Syndrome with Mild-to-Moderate Thrombocytopenia.

BACKGROUND: Baseline thrombocytopenia is commonly observed in patients with acute coronary syndrome (ACS) requiring percutaneous coronary intervention (PCI). AIM: The purpose of this analysis was to investigate safety and effectiveness of PCI in ACS patients with baseline mild-to-moderate thrombocytopenia. METHODS: The data were collected from the Improving Care for Cardiovascular Disease in China-Acute Coronary Syndrome project. A total of 50,009 ACS patients were recruited between July 2017 and December 2019. Among them, there were 6,413 patients with mild-to-moderate thrombocytopenia, defined as a platelet count of ≥50 × 109/L and <150 × 109/L on admission. The primary outcome was in-hospital net adverse clinical events (NACE), consisting of major adverse cardiac events (MACE) and major bleeding events. The associations between PCI and in-hospital outcomes were analyzed by inverse probability treatment weighting (IPTW) method. RESULTS: PCI was performed in 4,023 of 6,413 patients (62.7%). The IPTW analysis showed that PCI was significantly associated with a reduced risk of in-hospital MACE (odd ratio [OR]: 0.45; 95% confidence interval [CI]: 0.31-0.67; p < 0.01) and NACE (OR: 0.59; 95% CI: 0.42-0.83; p < 0.01). PCI was also associated with an increased risk of any bleeding (OR: 1.56; 95% CI: 1.09-2.22; p = 0.01) and minor bleeding (OR: 1.52; 95% CI: 1.00-2.30; p = 0.05), but not major bleeding (OR: 1.51; 95% CI: 0.76-2.98; p = 0.24). CONCLUSION: Compared with medical therapy alone, PCI is associated with better in-hospital outcomes in ACS patients with mild-to-moderate thrombocytopenia. Further studies with long-term prognosis are needed.

Zero-shot information extraction from radiological reports using ChatGPT.

INTRODUCTION: Electronic health records contain an enormous amount of valuable information recorded in free text. Information extraction is the strategy to transform free text into structured data, but some of its components require annotated data to tune, which has become a bottleneck. Large language models achieve good performances on various downstream NLP tasks without parameter tuning, becoming a possible way to extract information in a zero-shot manner. METHODS: In this study, we aim to explore whether the most popular large language model, ChatGPT, can extract information from the radiological reports. We first design the prompt template for the interested information in the CT reports. Then, we generate the prompts by combining the prompt template with the CT reports as the inputs of ChatGPT to obtain the responses. A post-processing module is developed to transform the responses into structured extraction results. Besides, we add prior medical knowledge to the prompt template to reduce wrong extraction results. We also explore the consistency of the extraction results. RESULTS: We conducted the experiments with 847 real CT reports. The experimental results indicate that ChatGPT can achieve competitive performances for some extraction tasks like tumor location, tumor long and short diameters compared with the baseline information extraction system. By adding some prior medical knowledge to the prompt template, extraction tasks about tumor spiculations and lobulations obtain significant improvements but tasks about tumor density and lymph node status do not achieve better performances. CONCLUSION: ChatGPT can achieve competitive information extraction for radiological reports in a zero-shot manner. Adding prior medical knowledge as instructions can further improve performances for some extraction tasks but may lead to worse performances for some complex extraction tasks.

A Hierarchy-driven Multi-label Network with Label Constraints for Post-operative Complication Prediction of Lung Cancer.

Lung cancer is one of the most dangerous cancers all over the world. Surgical resection remains the only potentially curative option for patients with lung cancer. However, this invasive treatment often causes various complications, which seriously endanger patient health. In this study, we proposed a novel multi-label network, namely a hierarchy-driven multi-label network with label constraints (HDMN-LC), to predict the risk of complications of lung cancer patients. In this method, we first divided all complications into pulmonary and cardiovascular complication groups and employed the hierarchical cluster algorithm to analyze the hierarchies between these complications. After that, we employed the hierarchies to drive the network architecture design so that related complications could share more hidden features. And then, we combined all complications and employed an auxiliary task to predict whether any complications would occur to impose the bottom layer to learn general features. Finally, we proposed a regularization term to constrain the relationship between specific and combined complication labels to improve performance. We conducted extensive experiments on real clinical data of 593 patients. Experimental results indicate that the proposed method outperforms the single-label, multi-label baseline methods, with an average AUC value of 0.653. And the results also prove the effectiveness of hierarchy-driven network architecture and label constraints. We conclude that the proposed method can predict complications for lung cancer patients more effectively than the baseline methods.Clinical relevance-This study presents a novel multi-label network that can more accurately predict the risk of specific postoperative complications for lung cancer patients. The method can help clinicians identify high-risk patients more accurately and timely so that interventions can be implemented in advance to ensure patient safety.

Disease severity and antiviral response in patients with chronic hepatitis B with non-obese NAFLD.

BACKGROUND: The burden of nonalcoholic fatty liver disease (NAFLD) is growing in patients with chronic hepatitis B (CHB). NAFLD is typically associated with obesity, however, it is increasingly being identified in non-obese patients. This study aimed to investigate disease severity and antiviral response in non-obese patients with CHB with NAFLD (CHB + NAFLD). METHODS: A total of 809 patients with CHB + NAFLD were prospectively recruited and followed up for 3 years. NAFLD was diagnosed by transient elastography and defined as controlled attenuation parameter ≥248 dB/m, in the absence of excessive alcohol intake. Obesity status was defined by the Asian body mass index (BMI) cutoff of 25 kg/m2. Metabolic abnormality was defined by the presence of dyslipidemia, hypertension or diabetes. Fibrosis staging was defined according to the EASL-ALEH guidelines, with fibrosis progression defined as ≥1-stage increment. RESULTS: In the total cohort (median age 40 years, 59.0% antiviral-treated), 33.3% were non-obese. Non-obese patients were less metabolically abnormal than obese patients (60.2% vs 72.0%, P = 0.003). After 3-year follow up, the rate of fibrosis progression was comparable between non-obese and obese patients (17.5% vs 21.9% in the total cohort, P = 0.145; 15.7% vs 14.6% in antiviral-treated cohort with persistent viral suppression, P = 0.795). No significant differences in virological and biochemical responses were observed between non-obese and obese patients (P >0.05 for all). CONCLUSIONS: Approximately one third of CHB + NAFLD patients were non-obese. Non-obese patients, while less metabolically abnormal, had a similar risk for fibrosis progression as obese patients. Obesity status did not impact the efficiency of antiviral therapy.

Adropin inhibits the progression of atherosclerosis in ApoE-/-/Enho-/- mice by regulating endothelial-to-mesenchymal transition.

Adropin, a secreted protein, coded by energy homeostasis-associated gene (Enho), is recently reported to modulate atherogenesis, with endothelial-to-mesenchymal transition (EndMT) involved in the early process. We explored whether adropin may alleviate atherosclerosis by regulating EndMT. We found that an intraperitoneal injection of adropin [105 µg/(kg·d) for 13 weeks] inhibited the progression of high-fat diet (HFD)-induced aortic atherosclerosis in apolipoprotein E-deficient mice (ApoE-/-) and those with double gene deletion (ApoE-/-/Enho-/-), as detected by Oil Red O and haematoxylin-eosin staining. In the aortas of ApoE-/- mouse, adropin treatment ameliorated the decrease in the mRNA expression of endothelial cell markers (leukocyte differentiation antigen 31, CD31, and vascular endothelial cadherin, VE-cadherin), but increased that of EndMT markers (alpha smooth muscle actin, α-SMA, and fibroblasts specific protein-1). In vitro, an adropin treatment (30 ng/ml) arrested the hydrogen peroxide (H2O2)-induced EndMT in human umbilical vein endothelial cells (HUVECs), attenuated the morphological changes of HUVECs, reduced the number of immunofluorescence-positive α-SMA, increased the mRNA and protein expressions of CD31 and VE-cadherin, and decreased those of α-SMA. Furthermore, the adropin treatment decreased the mRNA and protein expressions of transforming growth factor (TGF)-ß1 and TGF-ß2, and suppressed the phosphorylation of downstream signal protein Smad2/3 in HUVECs. These mitigative effects of adropin on H2O2-induced EndMT were reversed by the transfection of TGF-ß plasmid. The findings signify that adropin treatment may alleviate the atherosclerosis in ApoE-/-/Enho-/- mice by inhibiting EndMT via the TGF-ß/Smad2/3 signaling pathway.

An 18-year evolution of congenital heart disease in China: An echocardiographic database-based study.

OBJECTIVE: Data on the evolution of congenital heart disease (CHD) in China remain scarce. Based on a Chinese echocardiography database, we analyzed the observed rate (OR) and spectrum changes of CHD over the past 18 years with a focus on the congenital aortic valve malformation (CAVM) and Adult CHD (ACHD). METHODS AND RESULTS: The transthoracic echocardiographic data of 682,565 records from 2003 to 2020 were retrospectively reviewed at Fujian Medical University Union Hospital, China. A total of 37,200 CHD cases were recruited in this study. Over the three periods (from 2003 to 2008, 2009-2014, to 2015-2020), the OR of Total CHD decreased (106.72, 90.64, and 67.43 per 1000 cases, respectively); the proportion of Simple CHD to Total CHD increased (80.96%, 83.41%, and 87.97%, respectively), with a decrease in the proportion of Complex CHD (18.11%, 15.51%, and 10.42%, respectively) (p < 0.05 for all). The proportion of ACHD increased in most types of CHD [Total CHD: 25.79%, 27.84%, and 31.43%; CAVM: 69.02%, 73.42%, and 78.16%; CAVM with aortic stenosis (AS): 67.42%, 70.73%, and 79.25%; respectively, p < 0.05 for all], with a much higher proportion in both CAVM and CAVM with AS than in the other CHD types. The proportion of CHD patients receiving intervention increased over the designated periods. CONCLUSIONS: This study depicts the longitudinal changes of CHD in the Chinese population with a single-center echocardiographic data, revealing an increased proportion of Simple CHD, ACHD (including CAVM and CAVM with AS), and a decreased OR of Total CHD and proportion of Complex CHD.

A Multi-modal Heterogeneous Graph Forest to Predict Lymph Node Metastasis of Non-small Cell Lung Cancer.

Lymph node metastasis (LNM) is critical for treatment decision-making for cancer patients, but it is difficult to diagnose accurately before surgery. Machine learning can learn nontrivial knowledge from multi-modal data to support accurate diagnosis. In this paper, we proposed a Multi-modal Heterogeneous Graph Forest (MHGF) approach to extract the deep representations of LNM from multi-modal data. Specifically, we first extracted the deep image features from CT images to represent the pathological anatomic extent of the primary tumor (pathological T stage) using a ResNet-Trans network. And then, a heterogeneous graph with six vertices and seven bi-directional relations was defined by medical experts to describe the possible relations between the clinical and image features. After that, we proposed a graph forest approach to construct the sub-graphs by removing each vertex in the complete graph iteratively. Finally, we used graph neural networks to learn the representations of each sub-graph in the forest to predict LNM and averaged all the prediction results as final results. We conducted experiments on 681 patients' multi-modal data. The proposed MHGF achieves the best performances with a 0.806 AUC value and 0.513 AP value compared with state-of-art machine learning and deep learning methods. The results indicate that the graph method can explore the relations between different types of features to learn effective deep representations for LNM prediction. Moreover, we found that the deep image features about the pathological anatomic extent of the primary tumor are useful for LNM prediction. And the graph forest approach can further improve the generalization ability and stability of the LNM prediction model.

Inter-hospital transfer in patients with acute myocardial infarction in China: Findings from the improving care for cardiovascular disease in China-acute coronary syndrome project.

Background: Little is known about the current scenario of inter-hospital transfer for patients with acute myocardial infarction (AMI) in China. Methods: From November 2014 to December 2019, 94,623 AMI patients were enrolled from 241 hospitals in 30 provinces in China. We analyzed the pattern of inter-hospital transfer, and compared in-hospital treatments and outcomes between transferred patients and directly admitted patients. Results: Of these patients, 40,970 (43.3%) were transferred from hospitals that did not provide percutaneous coronary intervention (PCI). The proportion of patients who were transferred from non-PCI hospital was 46.3% and 11.9% (P < 0.001) in tertiary hospitals and secondary hospitals, respectively; 56.2% and 37.3% (P < 0.001) in hospitals locating in low-economic regions and affluent areas, respectively. Compared with directly admitted patients, transferred patients had lower rates of reperfusion for STEMI (57.8% vs. 65.2%, P < 0.001) and timely PCI for NSTEMI (34.7%vs. 41.1%, P < 0.001). The delay for STEMI patients were long, with 6.5h vs. 4.5h from symptom onset to PCI for transferred and directly admitted patients, respectively. The median time-point was 9 days for in-hospital outcomes. Compared with direct admission, the hazard ratios and 95% confidence intervals associated with inter-hospital transfer were 0.87 (0.75-1.01) and 0.87 (0.73-1.03) for major adverse cardiovascular events and total mortality, respectively, in inverse probability of treatment weighting models in patients with STEMI, and 1.02 (0.71-1.48) and 0.98 (0.70-1.35), respectively, in patients with NSTEMI. Conclusion: More than 40% of the hospitalized AMI patients were transferred from non-PCI-capable hospitals in China. Further strategies are needed to enhance the capability of revascularization and reduce the inequality in management of AMI.

An ensemble learning with active sampling to predict the prognosis of postoperative non-small cell lung cancer patients.

BACKGROUND: Lung cancer is the leading cause of cancer death worldwide. Prognostic prediction plays a vital role in the decision-making process for postoperative non-small cell lung cancer (NSCLC) patients. However, the high imbalance ratio of prognostic data limits the development of effective prognostic prediction models. METHODS: In this study, we present a novel approach, namely ensemble learning with active sampling (ELAS), to tackle the imbalanced data problem in NSCLC prognostic prediction. ELAS first applies an active sampling mechanism to query the most informative samples to update the base classifier to give it a new perspective. This training process is repeated until no enough samples are queried. Next, an internal validation set is employed to evaluate the base classifiers, and the ones with the best performances are integrated as the ensemble model. Besides, we set up multiple initial training data seeds and internal validation sets to ensure the stability and generalization of the model. RESULTS: We verified the effectiveness of the ELAS on a real clinical dataset containing 1848 postoperative NSCLC patients. Experimental results showed that the ELAS achieved the best averaged 0.736 AUROC value and 0.453 AUPRC value for 6 prognostic tasks and obtained significant improvements in comparison with the SVM, AdaBoost, Bagging, SMOTE and TomekLinks. CONCLUSIONS: We conclude that the ELAS can effectively alleviate the imbalanced data problem in NSCLC prognostic prediction and demonstrates good potential for future postoperative NSCLC prognostic prediction.

Prevalence, awareness, treatment, and control rates of hypertension in patients hospitalized with atrial fibrillation in China: Findings from the CCC-AF project.

Background: The status of hypertension in patients with atrial fibrillation (AF) remains unknown in China. Methods: This study used data from patients hospitalized with AF recruited by the Improving Care for Cardiovascular Disease in China-AF (CCC-AF) project from 236 hospitals enrolled by geographic-economic level in China from 2015 to 2019. The prevalence, awareness, treatment, and control rates of hypertension in patients hospitalized with AF were estimated. Multivariable logistic regression was used to analyze the factors associated with uncontrolled hypertension. Results: Among 60,390 patients hospitalized with AF, the prevalence of hypertension according to the 2018 Chinese hypertension guidelines was 66.1%. The awareness, treatment, and control rates of hypertension were 80.3, 55.8, and 39.9%, respectively. Among patients treated for hypertension, the treatment control rate was 46.2%. These rates varied according to patient clinical characteristics and geographic regions. The young (18-44 and 45-54 years old), rural insurance, alcohol drinking, history of heart failure, valvular AF, first diagnosed AF, and permanent AF, were associated with uncontrolled hypertension. Under the 2017 American College of Cardiology/American Heart Association (ACC/AHA) hypertension guidelines, the prevalence of hypertension was 79.3%, and the control and treatment control rates dropped to 16.7 and 21.2%, respectively. Conclusion: Hypertension is common in patients hospitalized with AF in China. Although most patients were aware of their hypertensive status, the treatment and control rates of hypertension were still low. The management of hypertension in patients with AF needs to be further improved.

Pre-hospital delay in patients with acute myocardial infarction in China: findings from the Improving Care for Cardiovascular Disease in China-Acute Coronary Syndrome (CCC-ACS) project.

OBJECTIVE: To describe the duration of the pre-hospital delay time and identify factors associated with prolonged pre-hospital delay in patients with acute myocardial infarction (AMI) in China. METHODS: Data were collected from November 2014 to December 2019 as part of the Improving Care for Cardiovascular Disease in China-Acute Coronary Syndrome (CCC-ACS) project. A total of 33,386 patients with AMI admitted to the index hospitals were included in this study. Two-level logistic regression was conducted to explore the factors associated with the pre-hospital delay and the associations between different pre-hospital delay and in-hospital outcomes. RESULTS: Of the 33,386 patients with AMI, 70.7% of patients arrived at hospital ≥ 2 h after symptom onset. Old age, female, rural medical insurance, symptom onset at early dawn, and non-use of an ambulance predicted a prolonged pre-hospital delay (all P < 0.05). Hypertension and heart failure at admission were only significant in predicting a longer delay in patients with ST-segment elevation myocardial infarction (STEMI) (all P < 0.05). A pre-hospital delay of ≥ 2 h was associated with an increased risk of mortality [odds ratio (OR) = 1.36, 95% CI: 1.09-1.69, P = 0.006] and major adverse cardiovascular events (OR = 1.22, 95% CI: 1.02-1.47, P = 0.033) in patients with STEMI compared with a pre-hospital delay of < 2 h. CONCLUSIONS: Prolonged pre-hospital delay is associated with adverse in-hospital outcomes in patients with STEMI in China. Our study identifies that patient characteristics, symptom onset time, and type of transportation are associated with pre-hospital delay time, and provides focuses for quality improvement.

Using Natural Language Processing and Machine Learning to Preoperatively Predict Lymph Node Metastasis for Non-Small Cell Lung Cancer With Electronic Medical Records: Development and Validation Study.

BACKGROUND: Lymph node metastasis (LNM) is critical for treatment decision making of patients with resectable non-small cell lung cancer, but it is difficult to precisely diagnose preoperatively. Electronic medical records (EMRs) contain a large volume of valuable information about LNM, but some key information is recorded in free text, which hinders its secondary use. OBJECTIVE: This study aims to develop LNM prediction models based on EMRs using natural language processing (NLP) and machine learning algorithms. METHODS: We developed a multiturn question answering NLP model to extract features about the primary tumor and lymph nodes from computed tomography (CT) reports. We then combined these features with other structured clinical characteristics to develop LNM prediction models using machine learning algorithms. We conducted extensive experiments to explore the effectiveness of the predictive models and compared them with size criteria based on CT image findings (the maximum short axis diameter of lymph node >10 mm was regarded as a metastatic node) and clinician's evaluation. Since the NLP model may extract features with mistakes, we also calculated the concordance correlation between the predicted probabilities of models using NLP-extracted features and gold standard features to explore the influence of NLP-driven automatic extraction. RESULTS: Experimental results show that the random forest models achieved the best performances with 0.792 area under the receiver operating characteristic curve (AUC) value and 0.456 average precision (AP) value for pN2 LNM prediction and 0.768 AUC value and 0.524 AP value for pN1&N2 LNM prediction. And all machine learning models outperformed the size criteria and clinician's evaluation. The concordance correlation between the random forest models using NLP-extracted features and gold standard features is 0.950 and improved to 0.984 when the top 5 important NLP-extracted features were replaced with gold standard features. CONCLUSIONS: The LNM models developed can achieve competitive performance using only limited EMR data such as CT reports and tumor markers in comparison with the clinician's evaluation. The multiturn question answering NLP model can extract features effectively to support the development of LNM prediction models, which may facilitate the clinical application of predictive models.

The Role of Metabolic Factors and Steatosis in Treatment-Naïve Patients with Chronic Hepatitis B and Normal Alanine Aminotransferase.

INTRODUCTION: We aimed to elucidate the impact of metabolic syndrome (MS) and nonalcoholic fatty liver disease (NAFLD) on treatment-naïve patients with chronic hepatitis B (CHB) and normal alanine aminotransferase (ALT). METHODS: We analyzed the clinical characteristics of a cross-sectional cohort of treatment-naïve patients with CHB and ALT in the upper limit of normal (ULN) from October 2018 to July 2021. ALT ≤ 0.5 ULN was stratified as low-normal ALT (LNALT) and 0.5 ULN < ALT ≤ ULN as high-normal ALT (HNALT). Transient elastography (TE) was used to evaluate liver steatosis and fibrosis. RESULTS: Among 733 patients with CHB enrolled, 23.1% of them had MS, 37.2% of them had NAFLD, and 5.9% of them had significant fibrosis. The proportions of patients with MS, steatosis, and significant fibrosis in the HNALT group were higher than those in the LNALT group (31.4% vs. 14.1%, p < 0.001; 48.7% vs. 25.2%, p < 0.001; and 8.0% vs. 3.6%, p = 0.013, respectively). Multiple linear regression showed that steatosis (beta = 0.098, p = 0.001) and MS (beta = 0.092, p = 0.002) were independently related to ALT levels in the normal range. Multivariate logistic regression showed that age (OR 1.049, 95% CI 1.012-1.087, p = 0.010), aspartate aminotransferase (AST) (OR 1.059, 95% CI 1.005-1.115, p = 0.030), and severe steatosis (OR 2.559, 95% CI 1.212-5.403, p = 0.014) were independently associated with significant fibrosis. When analyzed in the subgroup of CHB with NAFLD, age (OR 1.060, 95% CI 1.006-1.117, p = 0.029) and severe steatosis (OR 2.962, 95% CI 1.126-7.792, p = 0.028) were still statistically significant. CONCLUSION: The accumulation of MS components exacerbated hepatic steatosis. Severe NAFLD was independently associated with significant fibrosis. This emphasizes the importance of screening for MS and NAFLD in patients with CHB and normal ALT, where a more active intervention may apply.

End-of-treatment HBcrAg and HBsAb levels identify durable functional cure after Peg-IFN-based therapy in patients with CHB.

BACKGROUND & AIMS: Functional cure can be sustained in a proportion of patients with chronic hepatitis B (CHB) who lose hepatitis B surface antigen (HBsAg) after pegylated interferon alpha (Peg-IFN-ɑ)-based treatment. In this study, we aimed to identify biomarkers associated with a durable functional cure and to dissect potential immunological mechanisms. METHODS: Of 257 nucleos(t)ide analogue-suppressed patients with CHB in the ANCHOR study, 80 patients randomly assigned to 96-week Peg-IFN-α-based therapy with 24-week off-treatment follow-up were included in this parallel study. Virologic and immunological biomarkers were examined dynamically. A response was defined as HBsAg loss or hepatitis B surface antibody (HBsAb) appearance at the end of treatment (EOT). Sustained response (SR) or durable functional cure was defined as sustained HBsAg loss with or without the appearance of HBsAb at the end of follow-up (EOF). RESULTS: Thirty-six (45.0%) out of 80 patients achieved a response at EOT; 58.3% (21/36) of responders maintained SR at EOF. Quantitative hepatitis B core-related antigen (qHBcrAg) and HBsAb at EOT were associated with SR, with AUROCs of 0.697 (0.512-0.882, p = 0.047) and 0.744 (0.573-0.915, p = 0.013), respectively. A combination of HBcrAg <4 log10U/ml and HBsAb >2 log10IU/L at EOT had a positive predictive value of 100% for SR with an AUROC of 0.822 (0.684-0.961, p = 0.001). These patients showed maintained proportions of HBV envelope-specific CD8+T and B cells, a markedly increased proportion of T follicular helper cells after Peg-IFN-ɑ discontinuation, and significantly higher proportions of HBV polymerase-specific CD8+T and CD86+CD19+B cells at EOF. CONCLUSIONS: Lower HBcrAg and higher HBsAb levels at EOT were associated with sustained cellular and humoral immune responses. They can be used to identify patients likely to achieve durable functional cure post Peg-IFN-based therapy. GOV IDENTIFIER: NCT02327416 LAY SUMMARY: Functional cure can be sustained in a proportion of patients with chronic hepatitis B after pegylated interferon alpha-based treatment. However, predicting who will achieve durable functional cure remains challenging. Herein, we show that low levels of hepatitis B core-related antigen and higher levels of hepatitis B surface antibodies at the end of treatment are linked to immunological responses and are associated with durable functional cure.

Automatic SNOMED CT coding of Chinese clinical terms via attention-based semantic matching.

BACKGROUND: A considerable amount of meaningful information is routinely recorded in Chinese clinical data in text format, referred to as Chinese clinical terms. The lack of coding is a major difficulty hindering the application of clinical terms. SNOMED CT is a widely used and comprehensive clinical health care terminology collection because of its coverage, granularity, clinical orientation, and logical underpinning. It is useful and efficient for automatically assigning SNOMED CT codes to Chinese clinical terms, but it still faces several problems. Current cross-language clinical term matching studies rely on external resources, such as machine translation and rule-based methods. Semantic matching methods have achieved strong performance on text matching, but few studies have been done on cross-language clinical term matching. We present an effective attention-based semantic matching algorithm to automatically cross-language code Chinese clinical terms with SNOMED CT. METHOD: Firstly, BERT was used to turn the input into word embedding. Then, the word embeddings were encoded through a BiLSTM with self-attention to focus on capturing distant relationships among words with different weights depending on their contribution to semantic matching. Then, decomposable attention was used to make semantic matching trivially parallelizable to speed up calculation. Finally, fully connected layers and a sigmoid were utilized to output matching results. RESULTS: The 29,960 manually coded Chinese clinical terms, 30,040 unmatched Chinese clinical terms and SNOMED CT codes were collected to evaluate the proposed method. Compared with the existing semantic matching method, the proposed approach achieves state-of-the-art results demonstrating the effectiveness of the method with an accuracy of 0.905, a precision of 0.856, a recall of 0.518, and an F-measure of 0.645. The proposed Chinese-English bilingual term mapping, Chinese character-level and word-level encoder, English word-level encoder, BERT model, and attention mechanism performed better than other methods. CONCLUSION: The proposed automatic SNOMED CT coding approach of Chinese clinical terms via attention-based semantic matching can improve the performance of automated SNOMED CT code assignment for Chinese clinical terms and improve the efficiency of the code assignment.

Practice of reperfusion in patients with ST-segment elevation myocardial infarction in China: findings from the Improving Care for Cardiovascular Disease in China-Acute Coronary Syndrome project.

BACKGROUND: Reperfusion therapy is fundamental for ST-segment elevation myocardial infarction (STEMI). However, the details of contemporary practice and factors associated with reperfusion therapy in China are largely unknown. Therefore, this study aimed to explore reperfusion practice and its associated factors among hospitalized patients with STEMI in China. METHODS: Patients with STEMI who were admitted to 159 tertiary hospitals from 30 provinces in China were included in the Improving Care for Cardiovascular Disease in China-Acute Coronary Syndrome project from November 2014 to December 2019. The associations of the characteristics of patients and hospitals with reperfusion were examined using hierarchical logistic regression. The associations between therapies and in-hospital major adverse cardiovascular events were examined with a mixed effects Cox regression model. RESULTS: Among the 59,447 patients, 37,485 (63.1%) underwent reperfusion, including 4556 (7.7%) receiving fibrinolysis and 32,929 (55.4%) receiving primary percutaneous coronary intervention (PCI). The reperfusion rate varied across geographical regions (48.0%-73.5%). The overall rate increased from 60.0% to 69.7% from 2014 to 2019, mainly due to an increase in primary PCI within 12 h of symptom onset. Timely PCI, but not fibrinolysis alone, was associated with a decreased risk of in-hospital major adverse cardiovascular events compared with no reperfusion, with an adjusted hazard ratio (95% confidence interval) of 0.64 (0.54,0.76) for primary PCI at <12 h, 0.53 (0.37,0.74) for primary PCI at 12 to 24 h, 0.46 (0.25,0.82) for the pharmaco-invasive strategy, and 0.79 (0.54,1.15) for fibrinolysis alone. CONCLUSIONS: Nationwide quality improvement initiatives should be strengthened to increase the reperfusion rate and reduce inequality in China. TRIAL REGISTRATION: www.ClinicalTrials.gov , NCT02306616.