Search

Liver transplantation (LT) is the "cure" therapy for patients with hepatocellular carcinoma (HCC). However, some patients encounter HCC recurrence after LT. Unfortunately, there is no effective methods to identify the LT patients who have high risk of HCC recurrence and would benefit from adjuvant targeted therapy. The present study aimed to establish a scoring system to predict HCC recurrence of HCC patients after LT among the Chinese population, and to evaluate whether these patients are suitable for adjuvant targeted therapy.Clinical data of HCC patients who underwent LT from March 2015 to June 2019 were retrospectively collected and analyzed.A total of 201 patients were included in the study. The multivariate Cox analysis suggested that preoperative alpha fetoprotein (AFP)200 µg/L (HR = 2.666, 95% CI: 1.515-4.690; P = 0.001), glutamyl transferase (GGT)96 U/L (HR = 1.807, 95% CI: 1.012-3.224; P = 0.045), and exceeding the Hangzhou criteria (HR = 2.129, 95% CI: 1.158-3.914; P = 0.015) were independent risk factors for poor disease-free survival (DFS) in patients with HCC who underwent LT. We established an AFP-GGT-Hangzhou (AGH) scoring system based on these factors, and divided cases into high-, moderate-, and low-risk groups. The differences in overall survival (OS) and disease-free survival (DFS) rates among the three groups were significant (P 0.05). The efficacy of the AGH scoring system to predict DFS was better than that of the Hangzhou criteria, UCSF criteria, Milan criteria, and TNM stage. Only in the high-risk group, we found that lenvatinib significantly improved prognosis compared with that of the control group (P 0.05).The AGH scoring system provides a convenient and effective way to predict HCC recurrence after LT in HCC patients in China. Patients with a high-risk AGH score may benefit from lenvatinib adjuvant therapy after LT.

This study aimed to develop and evaluate a population pharmacokinetic (PPK) combined machine learning approach to predict tacrolimus trough concentrations for Chinese adult liver transplant recipients in the early post-transplant period. Tacrolimus trough concentrations were retrospectively collected from routine monitoring records of liver transplant recipients and divided into the training dataset (1287 concentrations in 145 recipients) and the test dataset (296 concentrations in 36 recipients). A PPK model was first established using NONMEM. Then a machine learning model of Xgboost was adapted to fit the estimated individual pharmacokinetic parameters obtained from the PPK model with Bayesian forecasting. The performance of the final PPK model and Xgboost model was compared in the test dataset. In the final PPK model, tacrolimus daily dose, postoperative days, hematocrit, aspartate aminotransferase, and concomitant with voriconazole were identified to significantly influence the clearance. The postoperative days along with hematocrit significantly influence the volume of distribution. In the Xgboost model, the first five predictors for predicting the clearance were concomitant with voriconazole, sex, SNPs of CYP3A4*1G and CYP3A5*3 in recipients, and tacrolimus daily dose, for the volume of distribution were postoperative days, age, weight, total bilirubin and Graft: recipient weight ratio. In the test dataset, the Xgboost model showed the minimum median prediction error of tacrolimus concentrations than the PPK model with or without Bayesian forecasting. In conclusion, a PPK combined machine learning approach could improve the prediction of tacrolimus concentrations for Chinese adult liver transplant recipients in the early post-transplant period. This article is protected by copyright. All rights reserved.

Background Diabetes mellitus has become an increasing global health burden with rapid growing prevalence. Patients with diabetes have higher susceptibility to acute kidney injury (AKI). Liver transplantation (LT) predisposes the kidney to injury. However, the association between diabetes and AKI in LT patients remains unclear. Methods We conducted a retrospective cohort study examining risk factors for AKI in patients undergone orthotopic LT. Potential risk factors including baseline estimated glomerular filtration rate (eGFR), the model for end-stage liver disease (MELD) score, diabetes, hypertension and intraoperative blood loss were screened. The primary endpoint was AKI occurrence. Multivariate logistic regression was used to analyze the association between potential risk factors and AKI. Results A total of 291 patients undergone orthotopic LT were included in the present study. Among them, 102 patients (35.05%) developed AKI within 5 days after LT. Diabetes was identified as an independent risk factor for AKI. Patients who developed AKI had worse graft function recovery and higher mortality within 14 days after LT compared to those who did not develop AKI. AKI patients with diabetes had a significant decline of eGFR within the first postoperative year, compared with patients who did not develop AKI and who developed AKI but without diabetes. Conclusions Diabetes is an independent risk factor for AKI after orthotopic LT. AKI is associated with delayed graft function recovery and higher mortality in short-term postoperative period. Diabetic patients who developed AKI after LT experience a faster decline of eGFR within the first year after surgery.

We assess the safety and feasibility of the left hepatic vein preferential approach (LHVPA) based on left hepatic vein (LHV) anatomy for living donor laparoscopic left lateral sectionectomy (LLLS). Data from 50 donors who underwent LLLS in Huashan Hospital from October 2016 to November 2019 were analyzed retrospectively. On the basis of the classification of the LHV anatomy, the vein was defined as the direct import type, upper branch type, or indirect import type. A subgroup analysis was performed to compare the outcomes between the LHVPA and non-LHVPA groups. All 50 patients underwent pure LLLS. The mean operative duration was 157.5 ± 29.7 minutes. The intraoperative blood loss was 160.4 ± 97.5 mL. No complications more severe than grade 3 occurred. LHVPA was applied in 13 patients, whereas non-LHVPA was applied in 10 patients with the direct import type and upper branch type anatomy. The operative duration was shorter in the LHVPA group than the non-LHVPA group (142.7 ± 22.0 versus 173.0 ± 22.8 minutes; P = 0.01). Intraoperative blood loss was reduced in the LHVPA group compared with the non-LHVPA group (116.2 ± 45.6 versus 170.0 ± 63.3 mL; P = 0.02). The length of the LHV reserved extrahepatically in the LHVPA group was longer than in the non-LHVPA group (4.3 ± 0.2 versus 3.3 ± 0.3 mm; P = 0.01). Fewer reconstructions of the LHV in the direct import type anatomy were required for the LHVPA group than for the non-LHVPA group (0/8 versus 4/6). LHVPA based on the LHV anatomy is recommended in LLLS because it can further increase the safety and the efficiency of surgery for suitable donors.

Infections still represent the main factors influencing morbidity and mortality following liver transplantation. This study aimed to evaluate the incidence and risk factors for infection and survival after liver transplantation.We retrospectively examined medical records in 210 liver recipients who underwent liver transplantation between April 2015 and October 2017 in our hospital. Clinical manifestations and results of pathogen detection test were used to define infection. We analyzed the prevalence, risk factors and prognosis of patients with infection.The median follow-up was 214 days; the incidence of infection after liver transplantation was 46.7% (n = 98) which included pneumonia (43.4%), biliary tract infection (21.9%), peritonitis (21.4%) and bloodstream infection (7.6%). Among the pathogens in pneumonia, the most frequently isolated was Acinetobacter baumanii (23.5%) and Klebsiella pneumoniae (21.2%). Model for end-stage liver disease (MELD) score (OR = 1.083, 95% CI: 1.045-1.123; P0.001), biliary complication (OR = 4.725, 95% CI: 1.119-19.947; P = 0.035) and duration of drainage tube (OR = 1.040, 95% CI: 1.007-1.074; P = 0.017) were independent risk factors for posttransplant infection. All-cause mortality was 11.0% (n = 23). The prognostic factors for postoperative infection in liver recipients were prior-transplant infection, especially pneumonia within 2 weeks before transplantation. Kaplan-Meier curves of survival showed that recipients within 2 weeks prior infection had a significantly lower cumulative survival rate compared with those without infection (65.2% vs. 90.0%; hazard ratio: 4.480; P0.001).Infection, especially pneumonia within 2 weeks before transplantation, complication with impaired renal function and MELD score after 7 days of transplantation was an independent prognostic factor for postoperative infection in liver transplant recipients.

PurposeSince the issuance in 2009, the digital currency has enjoyed an increasing popularity and has become one of the most important options for global investors. The purpose of this paper is to propose a hybrid model ( KDJ–Markov chain) which integrates the advantages of the stochastic index (KDJ) and grey Markov chain methods and provide a useful decision support tool for investors participating in the digital currency market.Design/methodology/approachTaking Litecoin's closing price prediction as an example, the closing prices from May 2 to June 20, 2017, are used as the training set, while those from June 21 to August 9, 2017, are used as the test set. In addition, an adaptive KDJ–Markov chain is proposed to enhance the adaptability for dynamic transaction information. And the paper verifies the effectiveness of the KDJ–Markov chain method and adaptive KDJ–Markov chain method.FindingsThe results show that the proposed methods can provide a reliable foundation for market analysis and investment decisions. Under the circumstances the accuracy of the training set and the accuracy of the test set are 76% and 78%, respectively.Practical implicationsThis study not only solves the problems that KDJ method cannot accurately predict the next day's state and the grey Markov chain method cannot divide the states very well, but it also provides two useful decision support tools for investors to make more scientific and reasonable decisions for digital currency where there are no existing methods to analyze the fluctuation.Originality/valueA new approach to analyze the fluctuation of digital currency, in which there are no existing methods, is proposed based on the stochastic index (KDJ) and grey Markov chain methods. And both of these two models have high accuracy.

The environmental Kuznets curve (EKC) hypothesis is used to describe the relationship between economic development and environmental pollution. In this paper, an EKC-estimating method based on an improved nonlinear gray Bernoulli model (NGBM) is proposed from the perspective of gray system modeling. First, a non-equigap NGBM is established taking the GDP per capita and pollutant emission as the input and output of the gray system, respectively. Then, a particle swarm optimization algorithm is used to find the parameters in the nonlinear model. Finally, the EKC is validated by applying it to the per capita emission of wastewater, SO2, CO2, and soot in China. The results show that the new method proposed in this paper optimizes the exponent of the NGBM which allows it to describe the trends in the different morphological data very well, resulting in a higher fitting accuracy. China’s per capita emission of wastewater, SO2, CO2, and soot show trends corresponding to monotonically increasing, inverted U-shaped, S-shaped, and N-shaped changes, respectively.

Background. Congenital hepatic fibrosis is a hereditary fibropolycystic disease caused by ductal plate malformation. It is characterized by portal hypertension, but the manifestations, management, and outcome vary in children and adults. To raise awareness of medical staff, we have comprehensively compared the clinical features of congenital hepatic fibrosis between children and adults. Methods. We retrospectively enrolled all patients diagnosed with congenital hepatic fibrosis at the Huashan Hospital from August 2015 to August 2017 and analyzed their familial, clinical, laboratory, imaging, treatment, and follow-up data in detail. In addition, we reviewed cases with congenital hepatic fibrosis reported in the past 20 years in China and analyzed them according to the patients’ age. Results. A total of eight patients were diagnosed with congenital hepatic fibrosis in the study, including four children and four adults. The onset age of the children, who suffered from severe complications of portal hypertension and needed liver transplantation, ranged from 1 to 15 years old. The disorder developed in adults aged 26 to 60 years old. Three adults complained of recurrent abnormal liver function at the onset of illness, and they mainly received conservative treatments. The literature review included 30 children and 33 adults. In comparison, hepatomegaly was more common in children than in adults (57% vs. 21%, p=0.004). Malformation of kidneys and bile duct abnormalities were common, and multisystem involvement included eyes, other digestive organs, and genital and central nervous systems. Conclusions. Serious complications of portal hypertension developed in children requiring liver transplantation, while adults often had mild-to-moderate liver injuries upon onset. Adults with CHF varied a lot in clinical manifestations. Multiorgan involvement and unusual course are helpful to make a diagnosis. Timely histological assessment by liver biopsy and multidisciplinary cooperation are crucial for definitive diagnosis and early intervention.

PurposeSeasonal fluctuation interference often affects the relational analysis of economic time series. The main purpose of this paper is to propose a new grey relational model for relational analysis of seasonal time series and apply it to identify and eliminate the influence of seasonal fluctuation of retail sales of consumer goods in China.Design/methodology/approachFirst, the whole quarterly time series is divided into four groups by data grouping method. Each group only contains the time series data in the same quarter. Then, the new series of four-quarters are used to establish the grey correlation model and calculate its correlation coefficient. Finally, the correlation degree of factors in each group of data was calculated and sorted to determine its importance.FindingsThe data grouping method can effectively reflect the correlation between time series in different quarters and eliminate the influence of seasonal fluctuation.Practical implicationsIn this paper, the main factors influencing the quarterly fluctuations of retail sales of consumer goods in China are explored by using the grouped grey correlation model. The results show that the main factors are different from quarter to quarter: in the first quarter, the main factors are money supply, tax and per capita disposable income of rural residents. In the second quarter are money supply, fiscal expenditure and tax. In the third quarter are money supply, fiscal expenditure and per capita disposable income of rural residents. In the fourth quarter are money supply, fiscal expenditure and tax.Originality/valueThis paper successfully realizes the application of grey relational model in quarterly time series and extends the applicable scope of grey relational model.

An assessment model is built to investigate the degree of order in the organisational structure of the electricity regulatory institution (ERI) in China. The model is based on Shannon entropy and is constructed from the perspective of timeliness and accuracy of the flow of information. The model is then used to evaluate the degree of order in the organisational structures of the ERI during three stages of reform that occurred during 2002–13. The results indicate that the reforms and improvements made in the organisational structure of China's ERI have resulted in a stepwise increase in their degree of order (corresponding to 0.3156, 0.3277, and 0.3324 in the three stages, respectively). On this basis, a scheme is put forward to optimise the degree of order in the structure of the energy regulatory institution in the current stage. The results show that downsizing the internal and subordinate departments appropriately and creating more governmental agencies to regulate energy are conducive to further improving the degree of order of the energy regulatory institution. Finally, we use principal component analysis to propose a priority scheme for adding more regulatory governmental agencies based on sorted energy production and consumption data.

Background An individual prognostic model that includes inflammation caused by the delayed recovery of liver function after surgery for the early recurrence of hepatocellular carcinoma (HCC) following liver transplantation (LT) has not been well determined. Our aim was to develop a nomogram model for predicting individual survival and early recurrence following LT for patients. Methods Retrospective data, including clinical pathology and follow-up data, on HCC patients were collected between October 2016 and October 2019 at Huashan Hospital Affiliated to Fudan University. A nomogram estimating recurrence post-transplantation was constructed using multivariate Cox regression analysis. Results A total of 210 patients were included in the present study. The multivariate estimators of recurrence consisted of age, maximum tumor diameter, tumor thrombus, microvascular invasion (MVI), alanine aminotransferase and alpha-fetoprotein on postoperative day 7. Nomogram of recurrence-free survival was developed. The calibration and discrimination of the novel model were assessed with the calibration curves and concordance index (C-index). Its reliability and advantages were evaluated by comparing it with the conventional American Joint Committee on Cancer (AJCC) 8th edition staging system using integrated discrimination improvement (IDI) and net reclassification improvement (NRI). In comparison to the AJCC 8th edition staging system, the C-index (development set: 0.796 vs. 0.643, validation set: 0.741 vs. 0.563), the area under the receiver operating characteristic curve (AUC) of the validation set (1-year AUC: 0.732 vs. 0.586, 2-year AUC: 0.705 vs. 0.504), the development set (1-year AUC: 0.799 vs. 0.551, 2-year AUC: 0.801 vs. 0.512), and this model's calibration plots all showed improved performance. In addition, NRI and IDI verified that the nomogram is an accurate prognostic tool. Subsequently, a web calculator was generated to assess the risk of tumor recurrence post-LT. Conclusions The nomogram, based on clinical and pathological factors, showed good accuracy in estimating prognostic recurrence and can be used to guide individual patient follow-up and treatment.

Mammalian target of rapamycin (mTOR) inhibitor as an attractive drug target with promising antitumor effects has been widely investigated. High quality clinical trial has been conducted in liver transplant (LT) recipients in Western countries. However, the pertinent studies in Eastern world are paucity. Therefore, we designed a clinical trial to test whether sirolimus can improve recurrence-free survival (RFS) in hepatocellular carcinoma (HCC) patients beyond the Milan criteria after LT. This is an open-labeled, single-arm, prospective, multicenter, and real-world study aiming to evaluate the clinical outcomes of early switch to sirolimus-based regimens in HCC patients after LT. Patients with a histologically proven HCC and beyond the Milan criteria will be enrolled. The initial immunosuppressant regimens are center-specific for the first 4-6 weeks. The following regimens integrated sirolimus into the regimens as a combination therapy with reduced calcineurin inhibitors based on the condition of patients and centers. The study is planned for 4 years in total with a 2-year enrollment period and a 2-year follow-up. We predict that sirolimus conversion regimen will provide survival benefits for patients particular in the key indicator RFS as well as better quality of life. If the trial is conducted successfully, we will have a continued monitoring over a longer follow-up time to estimate indicator of overall survival. We hope that the outcome will provide better evidence for clinical decision-making and revising treatment guidelines based on Chinese population data. Trial register: Trial registered at http://www.chictr.org.cn: ChiCTR2100042869.

In recent years, the constant advancement of market reform in China improves the market allocation efficiency of production factors and further promotes the rapid growth of economic aggregate output in the country. Additionally, the CO2 emitted in China also constantly increases. Based on panel data pertaining to 30 provinces from 1998 to 2015 in China, the CO2 emission efficiency and marketisation level of China were separately measured using a super Slacks-based measure approach and generalised principal component analysis. On this basis, by utilising a dynamic panel regression model and a panel smooth transition regression model, the influence of market reform on CO2 emission efficiency was explored. The results show that the marketisation level exerts a significantly positive effect on CO2 emission efficiency in both models and the results are robust. On the condition of taking energy price as a threshold variable, the influence of process of marketisation is significantly greater than that when taking technological progress as a threshold variable. It implied that although energy price and technological progress both significantly promote the marginal effect of market reform on CO2 emission efficiency, the energy price shows a greater influence than technological progress. Infrastructure construction and urbanisation can inhibit the improvement of CO2 emission efficiency. Additionally, the proportion of secondary industries can delay the improvement of CO2 emission efficiency.

In recent years, residential solar energy consumption of United States under the effect of a series of encouragement policies has exhibited a growth trend characterized by seasonal leap. To predict it, a new grey model based on data grouping and buffer operator is proposed. The model groups on a monthly or quarterly basis are grouped, which are then buffered separately to cope with prediction error caused by seasonal fluctuations and sudden changes in trend. In addition, a genetic algorithm is used to obtain the most appropriate degree of buffering. And then, the predictive effects of classical grey model, grey model based on data grouping, non-linear autoregressive neural network, echo state network, and the proposed model are compared. The results show that the mean absolute percentage errors of predicted results obtained by using these five models are 32.73%, 30.23%, 46.94%, 39.15%, and 6.17%, respectively, implying that the proposed model confers a significant advantage. Compared with the other four models, the new model can more effectively recognize the seasonal fluctuation and structural mutation of time series data. After conducting out-of-sample forecasting, the results demonstrate that the residential solar energy consumption of United States will maintain its rapid growth with an average annual growth rate of 24%.

Based on the traditional export sophistication indices used at a national level, three indices for measuring the export sophistication of the new energy industry are proposed. Meanwhile, data from 30 countries from 2000 to 2015 in the UN Comtrade database are used to calculate and compare export sophistication of global sub-divided new energy products and new energy industries at a transnational level. Moreover, the development levels of new energy industries in typical counties and regions are analysed through use of a regional anomaly test. The research results demonstrate that new energy products are ranked, in descending order, as: wind, biomass, nuclear, and solar energy products according to their technical contents, while nuclear power products show higher technical content than biomass energy products when considering product quality differences. Besides, the technical contents of new energy industries in different countries show regional heterogeneity to a certain extent: the new energy industries in Europe present increasingly evident superiority with time in terms of the technical content; the technical content of new energy industries in Asia presents a growing gap with other countries; while the technical content of new energy industries in America gradually catches up with other countries.

We propose a new estimation method of the Lorenz curve based on dummy variable regression for granting clear economic connotations to parameters in the estimation of the Lorenz curve. The dummy variable regression model of the Lorenz curve is established by introducing dummy variables to group data about income level. On this basis, it is proved that the regression function is endowed with the convexity that the Lorenz curve should have. To verify the effectiveness of the new method, empirical study is carried out taking the income data of urban and rural residents in the Chinese Household Income Project Survey (CHIP) database as examples. In the empirical study, the proposed model is compared with several existing classical parameter models. The results indicate that the slope parameter of the model represents the cumulative income–population resilience of corresponding residents, and reflects the income structure of residents to some extent. Meanwhile, compared with classical parameter models, the newly proposed method has advantages of higher fitting precision and stronger adaptability.

A new method for discussing the relationship between CO2 emissions and economic growth is proposed using grey systems theory. GDP and CO2 emissions per capita are separately regarded as the input to, and output of, a grey system to establish a non-equigap grey Verhulst model (NE grey Verhulst model). To avoid the errors resulting from substituting a difference equation for a differential equation in grey modelling theory, the derived non-equigap grey Verhulst model (DNE grey Verhulst model) is deduced. Moreover, the structural parameters of the model are optimised using a particle swarm optimisation (PSO) algorithm in an attempt to further improve modelling accuracy. Based on data relating to CO2 emissions and GDP per capita in China from 1990 to 2014, empirical research is conducted which shows that the relationship between CO2 emissions and economic growth exhibits an inverted U-shape and the emissions are in a rapid growth stage on the left of the curve. It is predicted that CO2 emissions per capita will continue to rise from 2016 to 2020 and will not reach their peak before 2030, so the Chinese government should take effective measures to reduce carbon emissions.

Capital intensity is an important indicator for reflecting the relative changes of production factors of an industry. The facilitating effect of capital deepening, i.e., the positive impact of the augment of capital intensity, towards the structural transformation of the industry is definite. Therefore, the accurate prediction of capital intensity of the new energy industry is of great significance in facilitating the structural transformation and upgrading of the industry. Based on the Cobb-Douglas production function, an industry-level capital-labour ratio (KLR) model is established to describe the dynamic characteristics of the capital intensity. Then, by combining the estimation and prediction method of the nonlinear grey Bernoulli model (NGBM(1, 1)) with the KLR model, a new hybrid grey model, i.e., NGBM(1, 1)-KLR model is proposed. In this way, the economic meaning of the KLR and the advantage of the NGBM(1, 1) model in solving small-sample and nonlinear problems are complemented with each other, which enables one to more favourably predict the capital intensity of the industry. To verify the effectiveness and superiority of the proposed model, the NGBM (1, 1)-KLR model is used to predict the capital intensity of the new energy industry in China, and the model is compared with the GM(1, 1) and the NGBM(1, 1) models in the prediction performance. The empirical results show that the NGBM(1, 1)-KLR model can more accurately predict the capital intensity of the industry in China than the GM(1, 1) and the NGBM(1, 1) models. Moreover, the new hybrid grey model is used to carry out the out-of-sample prediction for the capital intensity of the new energy industry in China in the period of 2017–2020. The predicted results demonstrate that the structure of the industry in China will further transform and upgrade towards the capital deepening.

To investigate whether the microRNA-144 (miR-144) had immune regulation effect on matured immune cells, we firstly used quantitative RT-PCR (qRT-PCR) to detect the expression changes of miR-144 between the matured and immature dendritic cells (DCs), macrophages, and the peripheral blood mononuclear cells (PBMCs). Then we went on inspecting the expression changes of TNF-α, IL-1β, IL-6 and IL-23 in the matured DCs treated with miR-144 mimics or inhibitors using qRT-PCR, and also performed western blot to test phosphorylation state of ERK, JNK, p38 and p65 in these cells. Next, TargetScan was conducted to forecast the target gene of miR-144, receptor activator for nuclear factor-κB ligand (RANKL), and double luciferase reporter system was applied to research their banding sites. We also determined the expression changes of RANKL in the DCs treated with miR-144 mimics or inhibitors using qRT-PCR and ELISA, respectively. The siRNA of RANKL was synthesized and transfected into DCs to inspect how the immune regulation effect of miR-144 was conducted to inhibit the expression of TNF-α using qRT-PCR, and lastly we used flow cytometry to investigate whether this effect applied to Th17 cells. As results, we found that miR-144 was down-regulated in the matured DCs, macrophages and PBMCs of liver transplantation patients, and the miR-144 mimics could inhibit the expression levels of TNF-α, IL-1β, IL-6 and IL-23 in the matured DCs. Furthermore, miR-144 interacted with RANKL at position 679-685 of RANKL 3'UTR, and suppressed the translation of RANKL mRNA to realize the negative-regulation. Besides, the silence of RANKL enhanced the suppression effect of miR-144 on TNF-α and this immune regulation effect was applied to Th17 cells, too. In conclusion, this study clearly illustrated that miR-144 could inhibit the expression of cytokine in matured immune cells through suppressing the translation of RANKL mRNA.

The outbreak of a novel coronavirus (SARS-CoV-2) has caused a large number of residents in China to be infected with a highly contagious pneumonia recently. Despite active control measures taken by the Chinese government, the number of infected patients is still increasing day by day. At present, the changing trend of the epidemic is attracting the attention of everyone. Based on data from 21 January to 20 February 2020, six rolling grey Verhulst models were built using 7-, 8- and 9-day data sequences to predict the daily growth trend of the number of patients confirmed with COVID-19 infection in China. The results show that these six models consistently predict the S-shaped change characteristics of the cumulative number of confirmed patients, and the daily growth decreased day by day after 4 February. The predicted results obtained by different models are very approximate, with very high prediction accuracy. In the training stage, the maximum and minimum mean absolute percentage errors (MAPEs) are 4.74% and 1.80%, respectively, in the testing stage, the maximum and minimum MAPEs are 4.72% and 1.65%, respectively. This indicates that the predicted results show high robustness. If the number of clinically diagnosed cases in Wuhan City, Hubei Province, China, where COVID-19 was first detected, is not counted from 12 February, the cumulative number of confirmed COVID-19 cases in China will reach a maximum of 60,364&ndash, 61,327 during 17&ndash, 22 March, otherwise, the cumulative number of confirmed cases in China will be 78,817&ndash, 79,780.

In recent years, the Chinese government has proposed a policy to replace coal use with natural gas and electricity in the northern region to reduce the air pollution caused by the large consumption of coal. In order to assess the air pollution reduction effect of the coal substitution policy in Liaoning Province, this paper proposes a data grouping grey model with a fractional order accumulation (FDGGM (1,1)). The empirical analysis results show that the new grey model can predict the monthly coal consumption more accurately than the traditional DGGM (1,1) model. The MAPEs of the training set in the FDGGM(1,1) and DGGM(1,1) models are 4.58% and 5.48%, and the MAPEs of the test set are 23.89% and 33.78%, respectively. And the policy achieves a great success based on the FDGGM(1,1) model. During the policy implementation period (from January 2015 to December 2018), the coal consumption in Liaoning Province decreased by 27.2501 million tons, while the emissions of CO2, SO2 and NOx fell by 0.714, 0.2316 and 0.2017 million tons, respectively. The results also provide a necessary support to the further implementation of the coal substitution policy.