A Machine Learning-Based Applied Prediction Model for Identification of Acute Coronary Syndrome (ACS) Outcomes and Mortality in Patients during the Hospital Stay.

Nowadays, machine learning (ML) is a revolutionary and cutting-edge technology widely used in the medical domain and health informatics in the diagnosis and prognosis of cardiovascular diseases especially. Therefore, we propose a ML-based soft-voting ensemble classifier (SVEC) for the predictive modeling of acute coronary syndrome (ACS) outcomes such as STEMI and NSTEMI, discharge reasons for the patients admitted in the hospitals, and death types for the affected patients during the hospital stay. We used the Korea Acute Myocardial Infarction Registry (KAMIR-NIH) dataset, which has 13,104 patients' data containing 551 features. After data extraction and preprocessing, we used the 125 useful features and applied the SMOTETomek hybrid sampling technique to oversample the data imbalance of minority classes. Our proposed SVEC applied three ML algorithms, such as random forest, extra tree, and the gradient-boosting machine for predictive modeling of our target variables, and compared with the performances of all base classifiers. The experiments showed that the SVEC outperformed other ML-based predictive models in accuracy (99.0733%), precision (99.0742%), recall (99.0734%), F1-score (99.9719%), and the area under the ROC curve (AUC) (99.9702%). Overall, the performance of the SVEC was better than other applied models, but the AUC was slightly lower than the extra tree classifier for the predictive modeling of ACS outcomes. The proposed predictive model outperformed other ML-based models; hence it can be used practically in hospitals for the diagnosis and prediction of heart problems so that timely detection of proper treatments can be chosen, and the occurrence of disease predicted more accurately.

Construction of a lncRNA/pseudogene-hsa-miR-30d-5p-GJA1 regulatory network related to metastasis of pancreatic cancer.

Pancreatic cancer, the most lethal malignant tumor, is notorious for its poor prognosis and metastatic potential. Non-coding RNAs (ncRNAs) are reported to play key roles in cancer metastasis. In this study, miRNA and gene expression profiles between metastatic pancreatic cancer cell M8 and its parental cell BxPC.3 were determined. Using differential expression analysis, survival analysis, target gene prediction, pathway enrichment analysis, intersection analysis and correlation analysis, hsa-miR-30d-5p/GJA1 axis was identified as the most potential pathway involved in metastasis of pancreatic cancer. Subsequently, two upstream lncRNAs (HELLPAR and OIP-AS1) and four upstream pseudogenes (AC093616.1, AC009951.1, TMEM183B and PABPC1P4) of hsa-miR-30d-5p/GJA1 axis were predicted and were then identified via assessment of RNA-RNA expression relationship. Furthermore, CTNNA1, CTNNB1 and CTNND1 were regarded as three crucial molecules to be participated in hsa-miR-30d-5p/GJA1-mediated metastatic potential in pancreatic cancer. In conclusion, we established a novel lncRNA/pseudogene-hsa-miR-30d-5p-GJA1 regulatory network linked to metastasis of pancreatic cancer.

Influence of dialysis duration on outcomes of simultaneous pancreas-kidney transplant.

OBJECTIVE: The objective of this study was to assess how pre-transplant dialysis duration affects transplant outcomes after simultaneous pancreas-kidney transplant (SPK) in patients with type 1 diabetes mellitus (T1DM). METHODS: Data of 6887 T1DM patients who underwent SPK transplantation between 2008 and 2018 were obtained from the Scientific Registry of Transplant Recipients database. According to pre-transplant dialysis duration, the patients were divided into the preemptive SPK, 0-2 years, 2-5 years, and >5 years dialysis groups. Kaplan-Meier survival analysis was performed to compare patient and graft survival among the groups. Univariate and multivariate Cox regression analyses were used to identify predictors of transplant outcomes. RESULTS: The mean follow-up period was 56.7 ± 34.7 months. Compared with no dialysis or preemptive SPK, dialysis for 0-2 years was not significantly associated with patient or kidney graft survival, while long-term dialysis of 2-5 years and >5 years was significantly associated with increased risk of death and kidney graft failure. However, the duration of dialysis was not associated with pancreas graft survival. CONCLUSION: Long-term dialysis duration before SPK transplant is an independent predictor of patient death and kidney graft failure in T1DM patients.

Identification of invasion-metastasis-associated microRNAs in hepatocellular carcinoma based on bioinformatic analysis and experimental validation.

BACKGROUND: Hepatocellular carcinoma (HCC) is one of the most lethal cancer, mainly attributing to its high tendency to metastasis. Vascular invasion provides a direct path for solid tumor metastasis. Mounting evidence has demonstrated that microRNAs (miRNAs) are related to human cancer onset and progression including invasion and metastasis. METHODS: In search of invasion-metastasis-associated miRNAs in HCC, microarray dataset GSE67140 was downloaded from the Gene Expression Omnibus database. Differentially expressed miRNAs (DE-miRNAs) were obtained by R software package and the potential target genes were predicted by miRTarBase. The database for annotation, visualization and integrated discovery (DAVID) was introduced to perform functional annotation and pathway enrichment analysis for these potential targets of DE-miRNAs. Protein-protein interaction (PPI) network was established by STRING database and visualized by Cytoscape software. The effects of the miR-494-3p and miR-126-3p on migration and invasion of HCC cell lines were evaluated by conducting wound healing assay and transwell assay. RESULTS: A total of 138 DE-miRNAs were screened out, including 57 upregulated miRNAs and 81 downregulated miRNAs in human HCC tumors with vascular invasion compared with human HCC tumors without vascular invasion. 762 target genes of the top three upregulated and downregulated miRNAs were predicted, and they were involved in HCC-related pathways, such as pathway in cancer, focal adhesion and MAPK signaling pathway. In the PPI network, the top 10 hub nodes with higher degrees were identified as hub genes, such as TP53 and MYC. Through constructing the miRNA-hub gene network, we found that most of hub genes could be potentially modulated by miR-494-3p and miR-126-3p. Of note, miR-494-3p and miR-126-3p was markedly upregulated and downregulated in HCC cell lines and tissues, respectively. In addition, overexpression of miR-494-3p could significantly promote HCC migration and invasion whereas overexpression of miR-126-3p exerted an opposite effect. CONCLUSIONS: Targeting miR-494-3p and miR-126-3p may provide effective and promising approaches to suppress invasion and metastasis of HCC.

Astaxanthin Expedites the Healing of Acute Skin Wounds in Rats by Facilitating M2 Macrophage Polarization and Enhancing Collagen Secretion.

BACKGROUND: Facilitating the healing process of skin post-trauma is crucial for minimizing infection risks and reinstating normal tissue functionality. While past studies have established astaxanthin (ASX) as an effective compound in promoting wound healing, the precise mechanism of its action remains unclear. Consequently, the objective of this study was to explore the impact of ASX on the acute wound healing of rat skin by modulating macrophage polarization. METHODS: Eighteen male SD rats were randomly assigned to control, dimethylsulfoxide (DMSO), and ASX groups. Acute skin wounds were induced in the rats, and the effects of different treatments on wound area and healing were assessed. Hematoxylin-eosin (H&E) staining was employed to detect histopathological changes in the skin, while Masson staining was utilized to observe collagen expression. Immunohistochemistry was conducted to identify clusters of differentiation (CD) 206 macrophages in the tissues. Furthermore, enzyme-linked immunosorbent assay (ELISA) was used to measure the levels of inflammatory cytokines such as tumor necrosis factor-α (TNF-α), interleukin (IL)-6, IL-8, IL-10, IL-4, and IL-13. The expression of inducible nitric oxide synthase (iNOS), arginase (Arg)-1, and mannose receptor C-type 1 (Mrc1) proteins in the injured skin of rats was assessed through Western blot analysis. RESULTS: On postoperative days 7 and 14, the ASX treatment demonstrated notable reductions in inflammatory cell infiltration and inflammatory cytokine expression when compared to the Control and DMSO groups. This was accompanied by evident improvements in the pathological changes in skin tissue, characterized by the regeneration of new epidermis, dermal repair, and increased thickness of granulation, contributing to enhanced scar formation. Furthermore, ASX therapy exhibited an upregulation in the expression levels of collagen I and collagen III, along with markers indicative of M2 macrophages. These findings collectively signify the accelerated progression of wound healing attributed to ASX intervention. CONCLUSIONS: In summary, these findings collectively indicate that ASX facilitates the healing of rat skin wounds by suppressing inflammatory responses and fostering M2 macrophage polarization. Consequently, ASX holds promise as a potentially effective drug for the treatment of skin wounds.

A cost-sensitive deep neural network-based prediction model for the mortality in acute myocardial infarction patients with hypertension on imbalanced data.

Background and objectives: Hypertension is one of the most serious risk factors and the leading cause of mortality in patients with cardiovascular diseases (CVDs). It is necessary to accurately predict the mortality of patients suffering from CVDs with hypertension. Therefore, this paper proposes a novel cost-sensitive deep neural network (CSDNN)-based mortality prediction model for out-of-hospital acute myocardial infarction (AMI) patients with hypertension on imbalanced data. Methods: The synopsis of our research is as follows. First, the experimental data is extracted from the Korea Acute Myocardial Infarction Registry-National Institutes of Health (KAMIR-NIH) and preprocessed with several approaches. Then the imbalanced experimental dataset is divided into training data (80%) and test data (20%). After that, we design the proposed CSDNN-based mortality prediction model, which can solve the skewed class distribution between the majority and minority classes in the training data. The threshold moving technique is also employed to enhance the performance of the proposed model. Finally, we evaluate the performance of the proposed model using the test data and compare it with other commonly used machine learning (ML) and data sampling-based ensemble models. Moreover, the hyperparameters of all models are optimized through random search strategies with a 5-fold cross-validation approach. Results and discussion: In the result, the proposed CSDNN model with the threshold moving technique yielded the best results on imbalanced data. Additionally, our proposed model outperformed the best ML model and the classic data sampling-based ensemble model with an AUC of 2.58% and 2.55% improvement, respectively. It aids in decision-making and offers a precise mortality prediction for AMI patients with hypertension.

Predicting microbe-disease association based on graph autoencoder and inductive matrix completion with multi-similarities fusion.

Background: Clinical studies have demonstrated that microbes play a crucial role in human health and disease. The identification of microbe-disease interactions can provide insights into the pathogenesis and promote the diagnosis, treatment, and prevention of disease. Although a large number of computational methods are designed to screen novel microbe-disease associations, the accurate and efficient methods are still lacking due to data inconsistence, underutilization of prior information, and model performance. Methods: In this study, we proposed an improved deep learning-based framework, named GIMMDA, to identify latent microbe-disease associations, which is based on graph autoencoder and inductive matrix completion. By co-training the information from microbe and disease space, the new representations of microbes and diseases are used to reconstruct microbe-disease association in the end-to-end framework. In particular, a similarity fusion strategy is conducted to improve prediction performance. Results: The experimental results show that the performance of GIMMDA is competitive with that of existing state-of-the-art methods on 3 datasets (i.e., HMDAD, Disbiome, and multiMDA). In particular, it performs best with the area under the receiver operating characteristic curve (AUC) of 0.9735, 0.9156, 0.9396 on abovementioned 3 datasets, respectively. And the result also confirms that different similarity fusions can improve the prediction performance. Furthermore, case studies on two diseases, i.e., asthma and obesity, validate the effectiveness and reliability of our proposed model. Conclusion: The proposed GIMMDA model show a strong capability in predicting microbe-disease associations. We expect that GPUDMDA will help identify potential microbe-related diseases in the future.

Establishment and characterization of a highly metastatic hepatocellular carcinoma cell line.

The prevalence of alcohol-related hepatocellular carcinoma (HCC) has been increasing during the last decade. Cancer research requires cell lines suitable for both in vitro and in vivo assays. However, there is a lack of cell lines with a high in vivo metastatic capacity for this HCC subtype. Herein, a new HCC cell line was established, named HCC-ZJ, using cells from a patient diagnosed with alcohol-related HCC. The karyotype of HCC-ZJ was 46, XY, del (p11.2). Whole-exome sequencing identified several genetic variations in HCC-Z that occur frequently in alcohol-associated HCC, such as mutations in TERT, CTNNB1, ARID1A, CDKN2A, SMARCA2, and HGF. Cell counting kit-8 assays, colony formation assays, and Transwell assays were performed to evaluate the proliferation, migration, and sensitivity to sorafenib and lenvatinib of HCC-Z in vitro. HCC-ZJ showed a robust proliferation rate, a weak foci-forming ability, a strong migration capacity, and a moderate invasion tendency in vitro. Finally, the tumorigenicity and metastatic capacity of HCC-Z were evaluated using a subcutaneous xenograft model, an orthotopic xenograft model, and a tail-veil injection model. HCCZJ exhibited strong tumorigenicity in the subcutaneous xenograft and orthotopic tumor models. Moreover, HCC-ZJ spontaneously formed pulmonary metastases in the orthotopic tumor model. In summary, a new HCC cell line derived from a patient with alcohol-related HCC was established, which showed a high metastatic capacity and could be applied for in vitro and in vivo experiments during pre-clinical research.Highlights• An alcohol-related HCC cell line, HCC-ZJ, was established• HCC-ZJ was applicable for in vitro functional experiment and gene editing• HCC-ZJ was applicable for in vivo tumor growth and spontaneous metastasis models.

scGAA: a general gated axial-attention model for accurate cell-type annotation of single-cell RNA-seq data.

Single-cell RNA sequencing (scRNA-seq) is a key technology for investigating cell development and analysing cell diversity across various diseases. However, the high dimensionality and extreme sparsity of scRNA-seq data pose great challenges for accurate cell type annotation. To address this, we developed a new cell-type annotation model called scGAA (general gated axial-attention model for accurate cell-type annotation of scRNA-seq). Based on the transformer framework, the model decomposes the traditional self-attention mechanism into horizontal and vertical attention, considerably improving computational efficiency. This axial attention mechanism can process high-dimensional data more efficiently while maintaining reasonable model complexity. Additionally, the gated unit was integrated into the model to enhance the capture of relationships between genes, which is crucial for achieving an accurate cell type annotation. The results revealed that our improved transformer model is a promising tool for practical applications. This theoretical innovation increased the model performance and provided new insights into analytical tools for scRNA-seq data.

Impact of Donor Age on Liver Transplant Outcomes in Patients with Acute-on-Chronic Liver Failure: A Cohort Study.

Background/Aims: Liver transplantation is the most effective treatment for the sickest patients with acute-on-chronic liver failure (ACLF). However, the influence of donor age on liver transplantation, especially in ACLF patients, is still unclear. Methods: In this study, we used the data of the Scientific Registry of Transplant Recipients. We included patients with ACLF who received liver transplantation from January 1, 2007, to December 31, 2017, and the total number was 13,857. We allocated the ACLF recipients by age into group I (donor age ≤17 years, n=647); group II (donor age 18-59 years, n=11,423); and group III (donor age ≥60 years, n=1,787). Overall survival (OS), graft survival, and mortality were compared among the three age groups and the four ACLF grades. Cox regression was also analyzed. Results: The 1-, 3-, and 5-year OS rates were 89.6%, 85.5%, and 82.0% in group I; 89.4%, 83.4%, and 78.2% in group II; and 86.8%, 78.4%, and 71.4% in group III, respectively (p<0.001). When we analyzed the different effects of donor age on OS with different ACLF grades, in groups II and III, we observed statistical differences. Finally, the cubic spline curve told us that the relative death rate changed linearly with increasing donor age. Conclusions: Donor age is related to OS and graft survival of ACLF patients after transplantation, and poorer results were associated with elderly donors. In addition, different donor ages have different effects on recipients with different ACLF grades.

Single-Cell Transcriptome Analysis Identified Core Genes and Transcription Factors in Mesenchymal Cell Differentiation during Liver Cirrhosis.

BACKGROUND: Mesenchymal cells, including hepatic stellate cells (HSCs), fibroblasts (FBs), myofibroblasts (MFBs), and vascular smooth muscle cells (VSMCs), are the main cells that affect liver fibrosis and play crucial roles in maintaining tissue homeostasis. The dynamic evolution of mesenchymal cells is very important but remains to be explored for researching the reversible mechanism of hepatic fibrosis and its evolution mechanism of hepatic fibrosis to cirrhosis. METHODS: Here, we analysed the transcriptomes of more than 50,000 human single cells from three cirrhotic and three healthy liver tissue samples and the mouse hepatic mesenchymal cells of two healthy and two fibrotic livers to reconstruct the evolutionary trajectory of hepatic mesenchymal cells from a healthy to a cirrhotic state, and a subsequent integrative analysis of bulk RNA sequencing (RNA-seq) data of HSCs from quiescent to active (using transforming growth factor ß1 (TGF-ß1) to stimulate LX-2) to inactive states. RESULTS: We identified core genes and transcription factors (TFs) involved in mesenchymal cell differentiation. In healthy human and mouse livers, the expression of NR1H4 and members of the ZEB families (ZEB1 and ZEB2) changed significantly with the differentiation of FB into HSC and VSMC. In cirrhotic human livers, VSMCs transformed into HSCs with downregulation of MYH11, ACTA2, and JUNB and upregulation of PDGFRB, RGS5, IGFBP5, CD36, A2M, SOX5, and MEF2C. Following HSCs differentiation into MFBs with the upregulation of COL1A1, TIMP1, and NR1H4, a small number of MFBs reverted to inactivated HSCs (iHSCs). The differentiation trajectory of mouse hepatic mesenchymal cells was similar to that in humans; however, the evolution trajectory and proportion of cell subpopulations that reverted from MFBs to iHSCs suggest that the mouse model may not accurately reflect disease progression and outcome in humans. CONCLUSIONS: Our analysis elucidates primary genes and TFs involved in mesenchymal cell differentiation during liver fibrosis using scRNA-seq data, and demonstrated the core genes and TFs in process of HSC activation to MFB and MFB reversal to iHSC using bulk RNA-seq data of human fibrosis induced by TGF-ß1. Furthermore, our findings suggest promising targets for the treatment of liver fibrosis and provide valuable insights into the molecular mechanisms underlying its onset and progression.

A Novel Four-Gene Signature Based on Nonsense-Mediated RNA Decay for Predicting Prognosis in Hepatocellular Carcinoma: Bioinformatics Analysis and Functional Validation.

Purpose: Nonsense-mediated RNA decay (NMD), a surveillance pathway for selective degradation of aberrant mRNAs, is associated with cancer progression. Its potential as a predictor for aggressive hepatocellular carcinoma (HCC) is unclear. Here, we present an innovative NMD risk model for predicting HCC prognosis. Methods: The Cancer Genome Atlas (TCGA) data of 374 liver HCC (LIHC) and 50 normal liver samples were extracted. A risk model based on NMD-related genes was developed through least absolute shrinkage and selection operator Cox (LASSO-Cox) regression of the LIHC-TCGA data. Prognostic validation was done using GSE54236, GSE116174, and GSE76427 data. Univariate and multivariate Cox regression analyses were conducted to assess the prognostic value of the model. We also constructed nomograms for survival prediction. Tumor immune infiltration was evaluated using the CIBERSORT algorithm, and the tumor cell phenotype was assessed. Finally, mouse experiments verified UPF3B knockdown effects on HCC tumor characteristics. Results: We developed a risk model based on four NMD-related genes (PABPC1, RPL8, SMG5, and UPF3B) and validated it using GSE54236, GSE116174, and GSE76427 data. The model effectively distinguished high- and low-risk groups corresponding to unfavorable and favorable HCC outcomes. Its prognostic prediction accuracy was confirmed through time-dependent ROC analysis, and clinical-use nomograms with calibration curves were developed. Single-cell RNA sequencing results indicated significantly higher expression of SMG5 and UPF3B in tumor cells. Knockdown of SMG5 and UPF3B inhibited HCC cell proliferation, invasion, and migration, while affecting cell-cycle progression and apoptosis. In vivo, UPF3B knockdown delayed tumor growth and increased immune cell infiltration. Conclusion: Our NMD-related gene-based risk model can help identify therapeutic targets and biomarkers for HCC. Additionally, it assists clinicians in predicting the prognosis of HCC patients.

MicroRNAs as Biomarkers and Therapeutic Targets for Nonalcoholic Fatty Liver Disease: A Narrative Review.

PURPOSE: Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in the world. However, biomarkers for NAFLD diagnosis and liver-specific drugs for treatment are lacking. This article reviews the possibility of circulating miRNAs in the diagnosis and treatment of NAFLD diseases and focuses on several well-studied miRNAs to provide preclinical data for subsequent related studies. METHODS: Related articles were identified through searches of the PubMed database for literature published from 2010 to December 2022. Search terms included NAFLD, microRNA, biomarker, diagnosis, and therapy. FINDINGS: Current research data indicate that some key circulating miRNAs may be used as diagnostic biomarkers of NAFLD and the combination of several miRNAs improves diagnostic performance. In addition, some preclinical trials using cell and mouse models provide a basis for some miRNAs as potential therapeutic targets. IMPLICATIONS: Current evidence suggests that circulating miRNAs are potential noninvasive biomarkers for clinical diagnosis of NAFLD, which needs to be validated in more heterogeneous and larger cohorts. In addition, several miRNAs regulate multiple downstream pathways related to the pathophysiology of NAFLD in a cell- and tissue-specific manner, making them attractive drug therapeutic targets for NAFLD. However, more preclinical and clinical trials are needed for these miRNAs to become therapeutic targets of NAFLD.

Hsa_circ_0007401 regulates gemcitabine resistance of pancreatic cancer through the hsa-miR-6509-3p/fli1 axis.

Pancreatic cancer is a highly malignant cancer with a poor prognosis. Owing to the strong drug resistance of pancreatic cancer, adjuvant chemotherapy has failed to achieve good results in clinical practice. The expression profile data of circular RNA (circRNA) (GSE110580), microRNA (miRNA) (GSE79234), and messenger RNA (mRNA) (GSE140077, GES35141) were obtained from the gene expression omnibus database. The Cancer-Specific circRNA Database identified the structural pattern of circRNA, and the starBase and circBank databases together predicted the miRNA of circRNA. The mirDIP database predicts the target mRNAs of miRNAs and identifies the ceRNA network of circRNA-miRNA-mRNA via negative regulatory mechanisms. The final validation was performed using clinical data from the cancer treatment response gene signature database of patients treated with gemcitabine for pancreatic cancer of the cancer genome atlas. By differential expression analysis, 22 differential circRNAs (8 upregulated and 14 downregulated), 70 differential microRNAs (37 upregulated and 33 downregulated), and 256 differential messenger RNA (DEmRNA) (161 upregulated and 95 downregulated) were obtained. Gene ontology and Kyoto encyclopedia of genes and genomes enrichment analyses showed that DEmRNAs were associated with drug response, exogenous cellular stimulation, and the tumor necrosis factor signaling pathway. The screened downregulated differential circular RNA (hsa_circ_0007401), upregulated differential microRNA (hsa-miR-6509-3p), and downregulated DEmRNA (FLI1) were consistent with the negative regulation mechanism of the ceRNA network, and FLI1 was significantly downregulated in the data of gemcitabine-resistant pancreatic cancer patients in the cancer genome atlas (n = 26).

Integrated analysis of circRNA-associated ceRNA network in ischemic stroke.

Introduction: Stroke, of which ischemic stroke (IS) is the major type, is the second leading cause of disability and death worldwide. Circular RNAs (circRNAs) are reported to play important role in the physiology and pathology of IS. CircRNAs often act as competing endogenous RNA (ceRNA) to regulate gene expression by acting as miRNA sponges. However, whole transcriptome-wide screenings of circRNA-mediated ceRNA networks associated with IS are still lacking. In the present study, we constructed a circRNA-miRNA-mRNA ceRNA network by whole transcriptome-wide analysis. Methods: CircRNAs, miRNAs and mRNAs expression profiles were downloaded from the Gene Expression Omnibus (GEO) datasets. We identified differentially expressed (DE) circRNAs, miRNAs, and mRNAs in IS patients. StarBase and CircBank databases were used to predict the miRNA targets of DEcircRNAs, and mirDIP database was used to predict the mRNA targets of DEmiRNAs. CircRNA-miRNA pairs and miRNA-mRNA pairs were established. Then, we identified hub genes via protein-protein interaction analysis and constructed a core ceRNA sub-network. Results: In total, 276 DEcircRNAs, 43 DEmiRNAs, and 1926 DEmRNAs were explored. The ceRNA network included 69 circRNAs, 24 miRNAs, and 92 mRNAs. The core ceRNA subnetwork included hsa_circ_0011474, hsa_circ_0023110, CDKN1A, FHL2, RPS2, CDK19, KAT6A, CBX1, BRD4, and ZFHX3. Discussion: In conclusion, we established a novel hsa_circ_0011474 - hsa-miR-20a-5p/hsa-miR-17-5p - CDKN1A ceRNA regulatory axis associated with IS. Our findings provide new insights into the pathogenesis of IS and offer promising diagnostic and predictive biomarkers.

The interaction of T2DM and BMI with NASH in recipients of liver transplants: an SRTR database analysis.

BACKGROUND: NASH-related liver transplants are increasing because of the obesity epidemic, but the influence of T2DM on various levels of BMI among NASH recipients is unclear. RESEARCH DESIGN AND METHODS: We analyzed data retrieved from SRTR on 4,515 patients. We divided patients by BMI into five groups: normal weight; overweight; class 1 obesity; class 2 obesity; and class 3 obesity. Statistical analysis was done. RESULTS: Patients in the NASH group with T2DM had a lower patient and graft survival than patients without T2DM (5-year patient and graft survival: 77.5% vs. 79.8%; P = 0.001 and 76.4% vs. 78.2%; P = 0.002, respectively). Multivariate Cox proportional regression showed an independent association between T2DM and decreased patient and graft survival (HR, 1.170; P = 0.015 and HR, 1.133; P = 0.048, respectively). In the lean and the class 3 obesity NASH groups, patients with T2DM had lower patient and graft survival than the patients without T2DM. In the class 3 obesity NASH group, T2DM was independently associated with decreased patient survival (HR, 1.581; P = 0.027). CONCLUSION: Our research reveals that the focus of the post-transplantation treatment should be different for different BMI patients.

Impact of donor diabetes mellitus status on liver transplant outcomes in patients with acute-on-chronic liver failure.

BACKGROUND: Liver transplantation (LT) is the most effective way to save patients with acute-on-chronic liver failure (ACLF). However, the impact of donor diabetes mellitus (DM) on LT outcomes in patients with ACLF has not been fully investigated. RESEARCH DESIGN AND METHODS: We retrospectively analyzed data from the Scientific Registry of Transplant Recipients (SRTR) between January 1st, 2008 to December 31st, 2017 in this study. All the patients were divided into donors with DM and without DM group (DM: 1,394; non-DM: 11138). We compared the overall survival (OS) and graft survival (GS) across different estimated ACLF (estACLF) grades between two groups. RESULTS: There were 25.10% estACLF-3 patients in the entire cohort. And in estACLF-3 patients, 318 patients had DM donors. The estACLF-3 associated 5-year OS rate in the non-DM group was 74.6%, significantly better than that in the DM group, with corresponding survival rate at 64.9% (P < 0.001). Donor DM was an independent predictor for OS in the entire cohort as well as in estACLF-3 patients. CONCLUSIONS: Donor DM was associated with inferior outcomes of LT in patients with estACLF-3. However, the differences were not obvious in recipients with other estACLF grades.

Deep learning-based prediction of heart failure rehospitalization during 6, 12, 24-month follow-ups in patients with acute myocardial infarction.

Heart failure is a clinical syndrome that occurs when the heart is too weak or stiff and cannot pump enough blood that our body needs. It is one of the most expensive diseases due to frequent hospitalizations and emergency room visits. Reducing unnecessary rehospitalizations is also an important and challenging task that has the potential of saving healthcare costs, enabling discharge planning, and identifying patients at high risk. Therefore, this paper proposes a deep learning-based prediction model of heart failure rehospitalization during 6, 12, 24-month follow-ups after hospital discharge in patients with acute myocardial infarction (AMI). We used the Korea Acute Myocardial Infarction-National Institutes of Health (KAMIR-NIH) registry which included 13,104 patient records and 551 features. The proposed deep learning-based rehospitalization prediction model outperformed traditional machine learning algorithms such as logistic regression, support vector machine, AdaBoost, gradient boosting machine, and random forest. The performance of the proposed model was accuracy, the area under the curve, precision, recall, specificity, and F1 score of 99.37%, 99.90%, 96.86%, 98.61%, 99.49%, and 97.73%, respectively. This study showed the potential of a deep learning-based model for cardiology, which can be used for decision-making and medical diagnosis tool of heart failure rehospitalization in patients with AMI.

Recurrence Risk of Liver Cancer Post-hepatectomy Using Machine Learning and Study of Correlation With Immune Infiltration.

Postoperative recurrence of liver cancer is the main obstacle to improving the survival rate of patients with liver cancer. We established an mRNA-based model to predict the risk of recurrence after hepatectomy for liver cancer and explored the relationship between immune infiltration and the risk of recurrence after hepatectomy for liver cancer. We performed a series of bioinformatics analyses on the gene expression profiles of patients with liver cancer, and selected 18 mRNAs as biomarkers for predicting the risk of recurrence of liver cancer using a machine learning method. At the same time, we evaluated the immune infiltration of the samples and conducted a joint analysis of the recurrence risk of liver cancer and found that B cell, B cell naive, T cell CD4+ memory resting, and T cell CD4+ were significantly correlated with the risk of postoperative recurrence of liver cancer. These results are helpful for early detection, intervention, and the individualized treatment of patients with liver cancer after surgical resection, and help to reveal the potential mechanism of liver cancer recurrence.

Identification of Key mRNAs as Prediction Models for Early Metastasis of Pancreatic Cancer Based on LASSO.

Pancreatic cancer is a highly malignant and metastatic tumor of the digestive system. Even after surgical removal of the tumor, most patients are still at risk of metastasis. Therefore, screening for metastatic biomarkers can identify precise therapeutic intervention targets. In this study, we analyzed 96 pancreatic cancer samples from The Cancer Genome Atlas (TCGA) without metastasis or with metastasis after R0 resection. We also retrieved data from metastatic pancreatic cancer cell lines from Gene Expression Omnibus (GEO), as well as collected sequencing data from our own cell lines, BxPC-3 and BxPC-3-M8. Finally, we analyzed the expression of metastasis-related genes in different datasets by the Limma and edgeR packages in R software, and enrichment analysis of differential gene expression was used to gain insight into the mechanism of pancreatic cancer metastasis. Our analysis identified six genes as risk factors for predicting metastatic status by LASSO regression, including zinc finger BED-Type Containing 2 (ZBED2), S100 calcium-binding protein A2 (S100A2), Jagged canonical Notch ligand 1 (JAG1), laminin subunit gamma 2 (LAMC2), transglutaminase 2 (TGM2), and the transcription factor hepatic leukemia factor (HLF). We used these six EMT-related genes to construct a risk-scoring model. The receiver operating characteristic (ROC) curve showed that the risk score could better predict the risk of metastasis. Univariate and multivariate Cox regression analyses revealed that the risk score was also an important predictor of pancreatic cancer. In conclusion, 6-mRNA expression is a potentially valuable method for predicting pancreatic cancer metastasis, assessing clinical outcomes, and facilitating future personalized treatment for patients with ductal adenocarcinoma of the pancreas (PDAC).