Multi-omics regulatory network inference in the presence of missing data.

A key problem in systems biology is the discovery of regulatory mechanisms that drive phenotypic behaviour of complex biological systems in the form of multi-level networks. Modern multi-omics profiling techniques probe these fundamental regulatory networks but are often hampered by experimental restrictions leading to missing data or partially measured omics types for subsets of individuals due to cost restrictions. In such scenarios, in which missing data is present, classical computational approaches to infer regulatory networks are limited. In recent years, approaches have been proposed to infer sparse regression models in the presence of missing information. Nevertheless, these methods have not been adopted for regulatory network inference yet. In this study, we integrated regression-based methods that can handle missingness into KiMONo, a Knowledge guided Multi-Omics Network inference approach, and benchmarked their performance on commonly encountered missing data scenarios in single- and multi-omics studies. Overall, two-step approaches that explicitly handle missingness performed best for a wide range of random- and block-missingness scenarios on imbalanced omics-layers dimensions, while methods implicitly handling missingness performed best on balanced omics-layers dimensions. Our results show that robust multi-omics network inference in the presence of missing data with KiMONo is feasible and thus allows users to leverage available multi-omics data to its full extent.

Identification of methylation markers for diagnosis of autism spectrum disorder.

Autism spectrum disorder (ASD) is a hereditary heterogeneous neurodevelopmental disorder characterized by social and speech dysplasia. We collected the expression profiles of ASD in GSE26415, GSE42133 and GSE123302 from the gene expression omnibus (GEO) database, as well as methylation data of GSE109905. Differentially expressed genes (DEGs) between ASD and controls were obtained by differential expression analysis. Enrichment analysis identified the biological functions and signaling pathways involved by common genes in three groups of DEGs. Protein-protein interaction (PPI) networks were used to identify genes with the highest connectivity as key genes. In addition, we identified methylation markers by associating differentially methylated positions. Key methylation markers were identified using the least absolute shrink and selection operator (LASSO) model. Receiver operating characteristic curves and nomograms were used to identify the diagnostic role of key methylation markers for ASD. A total of 57 common genes were identified in the three groups of DEGs. These genes were mainly enriched in Sphingolipid metabolism and PPAR signaling pathway. In the PPI network, we identified seven key genes with higher connectivity, and used qRT-PCR experiments to verify the expressions. In addition, we identified 31 methylation markers and screened 3 key methylation markers (RUNX2, IMMP2L and MDM2) by LASSO model. Their methylation levels were closely related to the diagnostic effects of ASD. Our analysis identified RUNX2, IMMP2L and MDM2 as possible diagnostic markers for ASD. Identifying different biomarkers and risk genes will contribute to the diagnosis of ASD and the development of new clinical and drug treatments.

Changes in efficiency of tertiary public general hospitals during the reform of public hospitals in Beijing, China.

BACKGROUND: In March 2009, the Chinese government formally launched a new round of healthcare reform. As the city with the highest concentration of high-quality medical resources in China, in the past 10 years, Beijing has also been exploring medical reforms. This article studied the performance and development of the 10 tertiary general public hospitals managed by Beijing Municipal Hospital Management Centre to provide policy basis for further deepening Beijing public hospital reform and improving hospital efficiency. METHODS: The 2011, 2015 and 2018 'Beijing Health Work Statistics' were used to evaluate the performance of Beijing's tertiary public general hospitals, based on the Pabon Lasso model and the data envelopment analysis (DEA) model. RESULTS: Based on the Pabon Lasso model, 60%, 70% and 70% of the hospitals were entirely efficient (zone 3) in 2011, 2015 and 2018. It shows that among the 10 general public hospitals in Beijing, efficient hospitals accounted for the majority and further increased during the reform period. The DEA model further illustrates this point and shows more effective hospitals (80%) than the Pabon Lasso model, showing the efficiency of these hospitals to be improved during the reform period. CONCLUSIONS: The efficiency of the 10 hospitals has gradually improved during the reform period, and the difficulty of seeing a doctor in Beijing at a national medical centre has been relieved to a certain extent. Combining the Pabon Lasso model and the DEA model can analyse hospital efficiency more comprehensively, and can prompt initial information for improving hospital efficiency, but the results also reflect some problems.

Assessment of public hospitals performance in Gaza governorates using the Pabón Lasso Model.

BACKGROUND: Measuring hospital performance can be undertaken in different areas including efficiency, productivity, and quality. Many studies have been conducted using several indicators in this regard. AIM: This study aimed at measuring efficiency of obstetrics and gynaecology, and paediatrics departments at the public hospitals affiliated with the Palestinian Ministry of Health in 2016, 2017 and 2018. METHOD: This descriptive study includes 12 hospitals; four (A-D) and eight (E-L) providing obstetrics and gynaecology, and paediatrics services, respectively. Data were collected from health annual reports for the 3-year study period. Pabón Lasso charts were drawn using Microsoft Excel 2013. RESULTS AND CONCLUSION: During the 3 years, hospitals B, D and C lied in zones 3, 1 and 2, respectively. Hospital A was in Zone 4 in 2016 and 2018, but in Zone 1 in 2017. In 2016, hospitals E, F and H were in Zone 3, while, hospitals I, J, K, and L were in Zone 1. In 2017, hospital G shifted from Zone 2 to Zone 3, and backed to Zone 2 in 2018. Hospital L has moved from Zone 1 to Zone 2 in 2018, whereas other hospitals have remained in the same zones. Inefficiency in hospitals provides an important opportunity for addressing the gaps in a quality and less costly manner. Emphasis antenatal care in primary healthcare is important. Further researches are required to cover other types of hospitals through employing frontier techniques of efficiency measurement.

Evaluation of performance at Palestinian public hospitals using Pabón Lasso model.

BACKGROUND: Hospitals have a paramount role in provision of health care services, which in turn significantly impacts the performance of any health system, especially in developing countries. AIM: This study aimed to evaluate the performance of the Ministry of Health general hospitals in Gaza according to their surgical and internal medicine departments during a 3-year period (2016, 2017 and 2018) using Pabón Lasso model. METHOD: This descriptive study includes the overall public general hospitals in Gaza Strip (7). Data have been collected from the officially disseminated reports, mainly about average length of stay, bed occupancy rate, and bed turnover ratio in the surgical and internal medicine departments for the study period. Pabón Lasso charts have been drawn using MS Excel 2013. RESULTS AND CONCLUSION: For surgical departments, about 42.8% were efficient (zone 3) during the 3 years, while, 42.8% were inefficient (zone 1). Only one hospital was relatively efficient (zone 4) in 2017 and 2018. Regarding internal medicine departments, 28.6%-42.8% of hospitals were efficient during the study period, 14.3%-42.8% were inefficient, and 28.6%-42.8% were relatively efficient (zone 2 and 4). It is recommended to count on such type of analysis in decision-making and identify obstacles for best utilization of the available resources. Inefficiency in hospitals serves as a good opportunity for resources mobilization or innovation in demand-creating interventions, thereby, regular evaluation of resources' distribution. Further research is required by applying frontier techniques of efficiency measurement.

Is bed turnover rate a good metric for hospital scale efficiency? A measure of resource utilization rate for hospitals in Southeast Nigeria.

BACKGROUND: Nigeria health sector, like that of other sub-Saharan African countries, increasingly faces critical resource constraints. Thus, there is need to seek for ways of improving efficient use of scarce health resources. The aim of this study was to determine resource utilization rate of teaching hospitals in Southeast Nigeria as a means of estimating their efficiency. METHODS: The study is a longitudinal cross sectional study. It applied ratio indicators and Pabon Lasso model using data on the number of hospital bed, number of inpatients and total inpatient-days from purposefully selected teaching hospitals in Southeast Nigeria to measure efficiency over a period of 6 years (2011-2011). RESULTS: The hospitals' mean bed occupancy rate was as low as 42.14%, far below standard benchmark of 80-85%. The mean average length of stay was as high as 8.15 days and observed mean bed turnover was 21.27 patients/bed/year. These findings portrayed high level of inefficiency in Nigeria teaching hospitals, which was further illustrated by Pabon Lasso graph, with only 10-20% of the hospital-years located within or near the efficient zone or quadrant. CONCLUSION: The study was able to show that health ratio indicators such as hospital bed turnover rate (BTR) and bed occupancy rate (BOR), as well as patients' average length of stay (ALS) can be used as tools for assessing hospital performance or its efficiency in resource utilization. Thus, in low and middle income countries where medical record keeping may be inadequate or poor, ratio indicators used alone or with Pabon Lasso graph/chart could be an optional metrics for hospital efficiency.

Assessing impacts and determinants of China's environmental protection tax on improving air quality at provincial level based on Bayesian statistics.

Facing severe PM2.5 pollution, China has adopted a series of clean air policies since 2013, but PM2.5 concentrations in China remain serious. Weighing both sustainable development and environmental protection, the Environmental Protection Tax Law was implemented in 2018 in China. This research employed a Bayesian space-time model to identify the impacts of the environmental protection tax on PM2.5 pollution (IEPTPM2.5P) at the provincial level in 2018 in China, combining remotely sensed and in-situ monitoring data. Then the influence factors of the IEPTPM2.5P was investigated using a Bayesian LASSO regression model. Results indicate that the IEPTPM2.5P resulted in a decreasing trend of annual PM2.5 concentrations in 31 provinces. The spatial pattern of the IEPTPM2.5P presented a distinct geographical feature. The highest five IEPTPM2.5P occurred in Beijing, Tianjin, Shanghai, Shandong, and Hebei, and the corresponding values were -1.81, -1.79, -1.52, -1.51, and -1.47 µg/m3 per year, respectively. Tourism output value associated negatively with the IEPTPM2.5P, and the other five variables associated positively with the IEPTPM2.5P. The urbanisation rate and relief amplitude were the top two influencing factors, with contributions of 36.3% and 19.3%, respectively. The IEPTPM2.5P increased 0.0141 µg/m3 per year (95% credibility interval (CI): 0.0013, 0.0259) if the urbanisation rate increased one percentage point. The influencing contributions and magnitudes of the tax rate for air pollutants and the environmental tax revenue are 9.6% and 12.1%, 0.0016 (95% CI:-0.0038, 0.0076) and 0.0108 (95% CI:-0.0188, 0.0412) µg/m3 per year, respectively.

Iranian hospital efficiency: a systematic review and meta-analysis.

PURPOSE: The purpose of this paper (systematic review and meta-analysis) is to synthesize and analyze studies that assessed Iranian hospital efficiency. DESIGN/METHODOLOGY/APPROACH: A systematic literature search was conducted using both international (the Institute for Scientific Information, Scopus and PubMed) and Iranian scientific (Magiran, IranMedex and Scientific Information Database) databases. The review included original studies that used the Pabon Lasso Model to examine Iranian hospital performance, published in Persian or English. A self-administered checklist was used to collect data. In total, 12 questions were used for quality assessment. FINDINGS: In total, 34 studies met our inclusion criteria. The fixed-effects meta-analysis indicated that 19.2 percent (95% confidence interval (CI): 15.6-23.2 percent) of hospitals were in Zone 1 (poor performance: low bed turnover rate (BTR) and bed occupancy rate (BOR) and high average hospital stay (ALoS)), 23.7 percent (95% CI: 20.1-27.8 percent) were in Zone 2, 31.7 percent (95% CI: 27.7-36 percent) in Zone 3 (good performance: high BTR and BOR and low ALoS) and 25.4 percent (95% CI: 21.7-29.5 percent) in Zone 4. PRACTICAL IMPLICATIONS: Results help Iranian health policymakers to understand hospital performance, which, in turn, may lead to promoting greater awareness and policy attention to improve Iranian hospital efficiency. ORIGINALITY/VALUE: This study indicated that most Iranian hospitals had sub-optimal performance. Further studies are required to understand factors that explain the country's hospital inefficiency.

Assessing health facility performance in Indonesia using the Pabón-Lasso model and unit cost analysis of health services.

Total health care costs have dramatically increased in Indonesia, and health facilities consume the largest share of health resources. This study aims to provide a better understanding of the characteristics of the best-performing health facilities. We use 4 national Indonesian datasets for 2011 and analysed 200 hospitals and 95 health centres. We first apply the Pabón-Lasso model to assess the relative performance of health facilities in terms of bed occupancy rate and the number of admissions per bed; the model gathers together health facilities into 4 sectors representing different levels of productivity. We then use a step-down costing method to estimate the cost per outpatient visit, inpatient, and bed days in hospitals and health centres. We combined both ratio analysis and applied bivariate and multivariate analyses to identify the predictors of the best-performing health facility; 37% of hospitals and 33% of health centres were located in the high-performing sector of the Pabón-Lasso model. The wide variation in unit costs across health facilities presented a basis for benchmarking and identifying relatively efficient units. Combining the unit cost analysis and Pabón-Lasso model, we find that health facility performance is affected by both internal (size and capacity, financing, type of patients, ownership, accreditation status, and staff availability) and external factors (economic status, population education level, location, and population density). Our study demonstrates that it is feasible to identify the best-performing health facilities and provides information about how to improve efficiency using simplistic methods.

A Prognostic Index Derived From LASSO-Selected Preoperative Inflammation and Nutritional Markers for Non-Muscle-Invasive Bladder Cancer.

BACKGROUND: There is an urgent need to identify a robust predictor for BCG response in patients with non-muscle-invasive bladder cancer (NMIBC). We aimed to employ the Lasso regression model for the selection and construction of an index (BCGI) utilizing inflammation and nutrition indicators to predict the response to BCG therapy. METHODS: After acquiring the ethics approval, we searched the electric medical records in our institution and performed data screening. Then, we developed the BCGI using a Lasso regression model and subsequently evaluated its performance in both the train and internal test datasets through Kaplan-Meier survival curves and Cox regression analysis. Then, we also evaluated the prognostic value of BCGI alongside the EAU2021 model. RESULTS: The training dataset and internal test dataset contained 295 and 196 patients, respectively. Referring to the Lasso results, BCGI consisted of hemoglobin, albumin, and platelet count, which could significantly predict the recurrence of NMIBC patients who accepted BCG in train (P = .012) and test (P = .004) datasets. The BCGI also exhibited statistically prognostic value in no smoking history, World Health Organization high grade, and T1 subgroups, both in train and test datasets. In multivariable analysis, BCGI exhibited independent prognostic value in train (P = .012) and test (P = .012) datasets. Finally, we constructed a nomogram that consisted of smoking history, T stage, World Health Organization grade, tumor size, and BCGI. Then, BCGI demonstrated significant independent prognostic value in NMIBC patients treated with BCG, a result not observed with the EAU2021 score or classification. CONCLUSION: Based on the results, we reasonably suggest that BCGI may be a useful predictor for NMIBC patients who accepted BCG. Furthermore, we have demonstrated the efficacy of constructing a prognostic index using clinical factors and a Lasso regression model, a versatile approach applicable to various medical conditions.

Prospective Prediction of Treatment Response in High-Grade Glioma Patients using Pre-Treatment Tumor ADC Value and miR-222 and miR-205 Expression Levels in Plasma.

Background: Treatment response in High-grade Glioma (HGG) patients changes based on their genetic and biological characteristics. MiRNAs, as important regulators of drug and radiation resistance, and the Apparent Diffusion Coefficients (ADC) value of tumor can be used as a prognostic predictor for glioma. Objective: This study aimed to identify some of the pre-treatment individual patient features for predicting the treatment response in HGG patients. Material and Methods: In this prospective study, 18 HGG patients, who were candidated for chemo-radiation treatment, participated after informed consent of the patients. The investigated features were the expression level of miR-222 and miR-205 in plasma, the ADC value of tumor, Body Mass Index (BMI), and age. Treatment response was assessed, and Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to obtain a model to predict the treatment response. Mann-Whitney U test was also applied to select the variables with a significant relationship with patients' treatment response. Results: The LASSO coefficients for miR-205, miR-222, tumor's mean ADC value, BMI, and age were 3.611, -1.683, 2.468, -0.184, and -0.024, respectively. Mann-Whitney U test results showed miR-205 and tumor's mean ADC significantly related to treatment response (P-value<0.05). Conclusion: The miR-205 expression level of the patient in plasma and tumor's mean ADC value has the potential for prognostic predictors in HGG.

A hub and spoke model to supply the Sicilian neurorehabilitation demand: effects on hospitalization rates and patient mobility.

Introduction: Cerebrovascular diseases in Sicily have led to high mortality and healthcare challenges, with a notable gap between healthcare demand and supply. The mobility of patients seeking care, both within and outside Sicily, has economic and organizational impacts on the healthcare system. The Hub and Spoke model implemented by the IRCCS Centro Neurolesi "Bonino-Pulejo" of Messina aims to distribute advanced neurorehabilitation services throughout Sicily, potentially reducing health mobility and improving service accessibility. Methods: The evaluation was based on calculating hospitalization rates, examining patient mobility across Sicilian provinces, and assessing the financial implications of neurorehabilitation admissions. Data from 2016 to 2018, covering the period before and after the implementation of the Hub and Spoke network, were analyzed to understand the changes brought about by this model. Results: The analysis revealed a significant increase in hospitalization rates for neurorehabilitation in the Sicilian provinces where spokes were established. This increase coincided with a marked decrease in interregional health mobility, indicating that patients were able to receive high-quality care closer to their residences. Furthermore, there was a decrease in both intra-regional and inter-regional escape rates in provinces within the Hub and Spoke network, demonstrating the network's efficacy in improving accessibility and quality of healthcare services. Discussion: The implementation of the Hub and Spoke network substantially improved neurorehabilitation healthcare in Sicily, enhancing both accessibility and quality of care for patients. The network's establishment led to a more efficient utilization of healthcare resources and balanced distribution of services. These advancements are vital steps toward equitable and effective healthcare delivery in Sicily.

Deciphering KDM8 dysregulation and CpG methylation in hepatocellular carcinoma using multi-omics and machine learning.

Aim: This study investigates the altered expression and CpG methylation patterns of histone demethylase KDM8 in hepatocellular carcinoma (HCC), aiming to uncover insights and promising diagnostics biomarkers.Materials & methods: Leveraging TCGA-LIHC multi-omics data, we employed R/Bioconductor libraries and Cytoscape to analyze and construct a gene correlation network, and LASSO regression to develop an HCC-predictive model.Results: In HCC, KDM8 downregulation is correlated with CpGs hypermethylation. Differential gene correlation analysis unveiled a liver carcinoma-associated network marked by increased cell division and compromised liver-specific functions. The LASSO regression identified a highly accurate HCC prediction signature, prominently featuring CpG methylation at cg02871891.Conclusion: Our study uncovers CpG hypermethylation at cg02871891, possibly influencing KDM8 downregulation in HCC, suggesting these as promising biomarkers and targets.

Changes in gene function can play a role in causing cancer. In this study, we looked at how a specific gene called KDM8 behaves in liver cancer. By analyzing a large set of liver cancer samples, we investigated how gene interactions are different in this disease and if they can help predict liver cancer risk. Our results show that the KDM8 gene is less active, and its DNA gets chemically modified more often in liver cancer. We also found a group of genes and DNA changes, which are linked to the disease. Using this information, we identified 16 important markers and built a computer model that can accurately predict liver cancer. We found that DNA methylation at a specific spot called cg02871891 is especially important for predicting liver cancer. Overall, our study suggests that high levels of DNA methylation may lead to reduced KDM8 activity in liver cancer, which could be important for future research and better diagnostic tools.

Prognostic Value of Platelet-to-Lymphocyte Ratio Combined with CHA2DS2-VASc Score for Nonvalvular Atrial Fibrillation Induced Cardiogenic Cerebral Embolism.

Aim: To determine the predictive significance of the platelet-to-lymphocyte ratio (PLR) combined with the CHA2DS2-VASc score for cardiogenic cerebral embolism (CCE) in patients with nonvalvular atrial fibrillation (NVAF). Methods: A total of 553 patients with NVAF were included in this retrospective study. The general data, PLR, CHA2DS2-VASc score and echocardiography indicators were compared. The risk factors for CCE and the predictive value of PLR and CHA2DS2-VASc were analyzed. Stratified analysis was performed based on the cut-off value. Least absolute shrinkage and selection operator (LASSO) regression analysis was utilized to build a model. The relationship between risk score and different anticoagulants was evaluated. Results: Multiple regression analysis showed hypertension (OR=3.95, 95% CI=2.12-7.35, p=1.40×10-5), diabetes mellitus (OR=2.95, 95% CI=1.57-5.58, p=7.65×10-4), PLR (OR=1.01, 95% CI=1.00-1.01, p<10-6), creatinine level (OR=1.01, 95% CI=1.00-1.02, p=7.44×10-3), left atrial diameter (LAD) (OR=1.90, 95% CI=1.13-3.19, p=1.51×10-2), ejection fraction (EF) (OR=0.93, 95% CI=0.87-0.98, p=8.06×10-3) and CHA2DS2-VASc score (OR=3.79, 95% CI=2.95-4.85, p<10-6) were independent risk factors for CCE. A one-way linear analysis also showed the above seven indexes were significantly correlated with CCE (F=56.4, p<10-6). The area under the receiver operating characteristic (ROC) curve of PLR and CHA2DS2-VASc score was 0.760 (95% CI:0.721-0.800), and 0.855 (95% CI: 0.824-0.886), respectively. Pearson correlation analysis showed that PLR was correlated with CHA2DS2-VASc score (r=0.331, p<10-6). Stratified analysis indicated there was a positive correlation between different risk group (p<10-6). Using the LASSO model, a composite indicator displayed differential power for distinguishing CCE with an AUC value of 0.884 (95% CI: 0.857-0.911). Patients with dabigatran and rivaroxaban exhibited higher risk score than those with warfarin (warfarin vs dabigatran, p=1.40×10-2; warfarin vs rivaroxaban p=3.00×10-3). Conclusion: PLR and CHA2DS2-VASc score are independent risk factors for CCE with NVAF, and the combination of the two indices can improve the prediction of CCE.

Erratum: A metabolism- associated gene signature for prognosis prediction of hepatocellular carcinoma.

[This corrects the article DOI: 10.3389/fmolb.2022.988323.].

Conceptualised psycho-medical footprint for health status outcomes and the potential impacts for early detection and prevention of chronic diseases in the context of 3P medicine.

Background: The Suboptimal Health Status Questionnaire-25 (SHSQ-25) is a distinctive medical psychometric diagnostic tool designed for the early detection of chronic diseases. However, the synaptic connections between the 25 symptomatic items and their relevance in supporting the monitoring of suboptimal health outcomes, which are precursors for chronic diseases, have not been thoroughly evaluated within the framework of predictive, preventive, and personalised medicine (PPPM/3PM). This baseline study explores the internal structure of the SHSQ-25 and demonstrates its discriminatory power to predict optimal and suboptimal health status (SHS) and develop photogenic representations of their distinct relationship patterns. Methods: The cross-sectional study involved healthy Ghanaian participants (n = 217; aged 30-80 years; ~ 61% female), who responded to the SHSQ-25. The median SHS score was used to categorise the population into optimal and SHS. Graphical LASSO model and multi-dimensional scaling configuration methods were employed to describe the network structures for the two populations. Results: We observed differences in the structural, node placement and node distance of the synaptic networks for the optimal and suboptimal populations. A statistically significant variance in connectivity levels was noted between the optimal (58 non-zero edges) and suboptimal (43 non-zero edges) networks (p = 0.024). Fatigue emerged as a prominently central subclinical condition within the suboptimal population, whilst the cardiovascular system domain had the greatest relevance for the optimal population. The contrast in connectivity levels and the divergent prominence of specific subclinical conditions across domain networks shed light on potential health distinctions. Conclusions: We have demonstrated the feasibility of creating dynamic visualizers of the evolutionary trends in the relationships between the domains of SHSQ-25 relative to health status outcomes. This will provide in-depth comprehension of the conceptual model to inform personalised strategies to circumvent SHS. Additionally, the findings have implications for both health care and disease prevention because at-risk individuals can be predicted and prioritised for monitoring, and targeted intervention can begin before their symptoms reach an irreversible stage. Supplementary information: The online version contains supplementary material available at 10.1007/s13167-023-00344-2.

A Machine Learning Approach to the Diagnosis of Autism Spectrum Disorder and Multi-Systemic Developmental Disorder Based on Retrospective Data and ADOS-2 Score.

Early and accurate diagnosis of autism spectrum disorders (ASD) and tailored therapeutic interventions can improve prognosis. ADOS-2 is a standardized test for ASD diagnosis. However, owing to ASD heterogeneity, the presence of false positives remains a challenge for clinicians. In this study, retrospective data from patients with ASD and multi-systemic developmental disorder (MSDD), a term used to describe children under the age of 3 with impaired communication but with strong emotional attachments, were tested by machine learning (ML) models to assess the best predictors of disease development as well as the items that best describe these two autism spectrum disorder presentations. Maternal and infant data as well as ADOS-2 score were included in different ML testing models. Depending on the outcome to be estimated, a best-performing model was selected. RIDGE regression model showed that the best predictors for ADOS social affect score were gut disturbances, EEG retrievals, and sleep problems. Linear Regression Model showed that term pregnancy, psychomotor development status, and gut disturbances were predicting at best for the ADOS Repetitive and Restricted Behavior score. The LASSO regression model showed that EEG retrievals, sleep disturbances, age at diagnosis, term pregnancy, weight at birth, gut disturbances, and neurological findings were the best predictors for the overall ADOS score. The CART classification and regression model showed that age at diagnosis and weight at birth best discriminate between ASD and MSDD.

A risk prediction model for type 2 diabetes mellitus complicated with retinopathy based on machine learning and its application in health management.

Objective: This study aimed to establish a risk prediction model for diabetic retinopathy (DR) in the Chinese type 2 diabetes mellitus (T2DM) population using few inspection indicators and to propose suggestions for chronic disease management. Methods: This multi-centered retrospective cross-sectional study was conducted among 2,385 patients with T2DM. The predictors of the training set were, respectively, screened by extreme gradient boosting (XGBoost), a random forest recursive feature elimination (RF-RFE) algorithm, a backpropagation neural network (BPNN), and a least absolute shrinkage selection operator (LASSO) model. Model I, a prediction model, was established through multivariable logistic regression analysis based on the predictors repeated ≥3 times in the four screening methods. Logistic regression Model II built on the predictive factors in the previously released DR risk study was introduced into our current study to evaluate the model's effectiveness. Nine evaluation indicators were used to compare the performance of the two prediction models, including the area under the receiver operating characteristic curve (AUROC), accuracy, precision, recall, F1 score, balanced accuracy, calibration curve, Hosmer-Lemeshow test, and Net Reclassification Index (NRI). Results: When including predictors, such as glycosylated hemoglobin A1c, disease course, postprandial blood glucose, age, systolic blood pressure, and albumin/urine creatinine ratio, multivariable logistic regression Model I demonstrated a better prediction ability than Model II. Model I revealed the highest AUROC (0.703), accuracy (0.796), precision (0.571), recall (0.035), F1 score (0.066), Hosmer-Lemeshow test (0.887), NRI (0.004), and balanced accuracy (0.514). Conclusion: We have built an accurate DR risk prediction model with fewer indicators for patients with T2DM. It can be used to predict the individualized risk of DR in China effectively. In addition, the model can provide powerful auxiliary technical support for the clinical and health management of patients with diabetes comorbidities.

Comprehensive analysis to identify the neurotransmitter receptor-related genes as prognostic and therapeutic biomarkers in hepatocellular carcinoma.

Background: Hepatocellular carcinoma (HCC) is a highly heterogeneous disease with high morbidity and mortality, which accounts for the fourth most common cause of cancer-related deaths. Reports suggest that the neurotransmitter receptor-related genes (NRGs) may influence the tumor microenvironment and the prognosis of patients with HCC. Methods: The clinical information and RNA-seq data of patients with HCC were acquired from the ICGC-LIRI-JP dataset and the TCGA-LIHC dataset. Effects of 115 NRGs on the prognosis of HCC patients were analyzed in the ICGC-LIRI-JP dataset. The least absolute shrinkage and selection operator (LASSO) regression model was utilized to generate a risk score formula based on the critical NRGs. Next, the risk score effectiveness was validated both in the TCGA-LIHC dataset and in our clinical HCC samples. Based on the risk scores, patients with HCC were divided into two groups. Moreover, differentially expressed genes (DEGs) were screened. The gene ontology (GO) was used to analyze the functional enrichments of DEGs and to identify potential signaling pathways. To test the diagnostic effectiveness of our model, the receiver operator characteristic curve (ROC) analysis and nomogram were used. Finally, potential targeted drug prediction was performed based on DEGs of nine clinical HCC samples. Results: Nine NRGs were correlated significantly with the prognosis of patients with HCC, and eight NRGs were successfully included in the LASSO regression model. The Kaplan-Meier analysis of overall survival (OS) suggested that patients in the high-risk score group had worse prognosis; on the other hand, ROC analysis revealed a high prognostic value of the risk score in HCC. Several critical signaling pathways, such as lipid metabolism, organic acid metabolism, cell migration, cell adhesion, and immune response, were enriched both in public datasets and clinical samples. Nomogram results also suggested that the risk scores correlated well with the patients' prognosis. Potential targeted drugs prediction revealed that tubulin inhibitors might be the promising drugs for patients with HCC who have high risk scores based on the NRGs. Conclusion: We established a prognostic model based on critical NRGs. NRGs show a promising prognostic prediction value in HCC and are potential therapeutic targets for the disease treatment.

Spatiotemporal Evolution and Drivers of Total Health Expenditure across Mainland China in Recent Years.

A substantially growing health expenditure has become an important global issue. Thus, how and why health expenditure is rising should be urgently investigated in systematic research. The Bayesian space-time model and the Bayesian least absolute shrinkage and selection operator (LASSO) model were employed in this study to investigate the spatiotemporal trends and influence patterns of total health expenditure per capita (THEPC) and total health expenditure (THEE) as a share of the gross domestic product (GDP) on the Chinese mainland from 2009 to 2018. The spatial distribution of THEE as a share of GDP in mainland China has shaped a distinct geographical structure with the characteristic of 'west high/east low'. Its local increasing trends formed a geographical structure that exhibited a 'north high/south low' feature. The heterogeneity of the influence patterns of health expenditure was observed from east to west across China. Natural environmental factors, such as air pollution and green coverage, along with changes in dietary structures, have increasingly influenced the growth of health expenditures.