ObjectivesTo estimate the latest burden of disability adjusted life years (DALYs) for liver cancer in China and the long-term trend, and to make future prediction.MethodsBased on the visualization platform of Global Burden of Disease 2016, data on the DALYs for liver cancer in China was extracted. The very recent status in 2016 and the previous trend from 1990 to 2016 were described, using annualized rate of change (ARC). The burden from 2017 to 2050 was further predicted by combining the ARC and the Chinese population data projected by the United Nation.ResultsIn 2016, the total DALYs for liver cancer in China was estimated as 11 539 000 person years (accounting for 54.6% of the global burden), and years of life lost (YLLs) and years lived with disability (YLDs) contributed 98.9% and 1.1%, respectively. The age-standardized DALY rate was 844.1 per 100 000 (3.0 times of the global average) and the male-to-female ratio was 3.4. The DALY rate continuously increased from 1990–2016 (ARC=0.57%), particularly in recent 5 years (ARC=1.75%). Among the DALYs for all cancers, liver cancer contributed approximately 20% and constantly remained as the top 2 (ranking as the number one before year 2005). There were inverse trends in gender, with increasing in males and decreasing in females (ARC was 0.77% and –0.11%, respectively). Hepatitis B infection continually kept the leading cause of DALYs for liver cancer (accounting for nearly 57%), and the DALY rate was gradually increasing (ARC=0.43%). Although the peak age of DALY rate was stable at 65to 69 years, the peak age of the DALYs changed from 55 to 59 years in 1990 to 60 ~ 64 years in 2016. In 2050, the estimated DALYs for liver cancer in China will reach 14.37 million person years, 20.0% more than that in 2017.ConclusionsThe DALYs caused by liver cancer in China exceeds the overall burden of all other countries in the world, and accounts for 1/5 of DALYs for all cancers in local population. The burden in males has been continuously rising, and the leading cause remained unchanged as hepatitis B infection. With population aging, the DALYs for liver cancer in China will be incessant to increase, suggesting the necessity to implement continuous effort in risk factors prevention (e.g. hepatitis B infection), and efficient management in high risk population of liver cancer.
Clinical prediction models typically utilize a combination of multiple variables to predict individual health outcomes. However, multiple prediction models for the same outcome often exist, making it challenging to determine the suitable model for guiding clinical practice. In recent years, an increasing number of studies have evaluated and summarized prediction models using the systematic review/meta-analysis method. However, they often report poorly on critical information. To enhance the reporting quality of systematic reviews/meta-analyses of prediction models, foreign scholars published the TRIPOD-SRMA reporting guideline in BMJ in March 2023. As the number of such systematic reviews/meta-analyses is increasing rapidly domestically, this paper interprets the reporting guideline with a published example. This study aims to assist domestic scholars in better understanding and applying this reporting guideline, ultimately improving the overall quality of relevant research.
ObjectiveTo systematically review mortality risk prediction models for acute type A aortic dissection (AAAD). MethodsPubMed, EMbase, Web of Science, CNKI, WanFang Data, VIP and CBM databases were electronically searched to collect studies of mortality risk prediction models for AAAD from inception to July 31th, 2021. Two reviewers independently screened literature, extracted data and assessed the risk of bias of included studies. Systematic review was then performed. ResultsA total of 19 studies were included, of which 15 developed prediction models. The performance of prediction models varied substantially (AUC were 0.56 to 0.92). Only 6 studies reported calibration statistics, and all models had high risk of bias. ConclusionsCurrent prediction models for mortality and prognosis of AAAD patients are suboptimal, and the performance of the models varies significantly. It is still essential to establish novel prediction models based on more comprehensive and accurate statistical methods, and to conduct internal and a large number of external validations.
Objective To observe the correlation between the level of advanced glycosylation end products (AGE) in skin and diabetic retinopathy (DR), and establish and preliminatively verify the nomogramolumbaric model for predicting the risk of DR. MethodsA clinical case-control study. A total of 346 patients with type 2 diabetes mellitus (T2DM) who were admitted to the Department of Endocrinology and Ophthalmology of the First Affiliated Hospital of Zhengzhou University from January 2023 to June 2024 were included in the study. Among them, 198 were males and 148 were females. The mean age was (54.77±10.92). According to whether the patients were accompanied by DR, the patients were divided into the non-DR group (NDR group) and the DR group (DR group), 174 and 172 cases, respectively. All patients underwent skin AGE detection using a noninvasive diabetes detector. Diabetes duration, hemoglobin A1c (HbA1c), fasting plasma glucose, Urea, creatinine (Crea), uric acid, total cholesterol, triglyceride, estimated glomerular filtration rate (eGFR), urinary albumin concentration (UALB), and body mass index (BMI) were collected in detail. Univariate analysis and multivariate logistic regression analysis were used to determine the independent risk factors for T2DM concurrent DR, and to construct a nomogram prediction model for DR risk. Receiver operating characteristic curve (ROC curve), calibration curve and decision curve (DCA) were used to evaluate the model. ResultsHypertension prevalence rate (χ2=3.892), Diabetes duration (Z=?7.708), BMI (Z=?2.627), HbA1c (Z=?4.484), Urea (Z=?4.620), Crea (Z=?3.526), UALB (Z=?6.999), AGE (Z=?8.097) in DR group were significantly higher than those in NDR group, with statistical significance (P＜0.05); eGFR was lower than that in NDR group, the difference was statistically significant (Z=?6.061, P＜0.05). Logistic regression analysis showed that AGE, diabetes duration, HbA1c, UALB and eGFR were independent risk factors for DR (P＜0.05). Based on the results of multi-factor regression analysis, a nomogram prediction model was constructed. The area under ROC curve of the model was 0.843, 95% confidence interval was 0.802-0.884, sensitivity and specificity were 79.1% and 75.9%, respectively. The calibration curve was basically consistent with the ideal curve. The results of DCA analysis showed that when the model predicted the risk threshold of patients with DR between 0.17 and 0.99, the clinical net benefit provided by the nomogram model was＞0. ConclusionsSkin AGE level is an independent risk factor for DR. The nomogram prediction model based on AGE, diabetes duration, HbA1c, eGFR and UALB can accurately predict the risk of DR, and has good clinical practicability.
ObjectiveTo investigate the predictive factors of portal vein thrombosis (PVT) before and after splenectomy and gastroesophageal devascularization for liver cirrhosis with portal hypertension.
MethodsSixty-one cases of liver cirrhosis with portal hypertension who underwent splenectomy and gastroesophageal devascularization were enrolled retrospectively. The patients were divided into PVT group and non-PVT group based on the presence or absence of postoperative PVT on day 7. The clinical factors related with PVT were analyzed.
ResultsThere were 25 cases in the DVT group and 36 cases in the non-DVT group. The results of univariate analysis showed that the preoperative platelet (P=0.006), activated partial thromboplastin time (P=0.048), prothrombin time (P=0.028), and international normalized ratio (P=0.029), postoperative fibrin degradation product (P=0.002) and D-dimer (P=0.014) on day 1, portal venous diameter (P=0.050) had significant differences between the DVT group and non-DVT group. The results of logistic multivariate regression analysis showed that the preoperative platelet (OR=0.966, 95% CI 0.934-1.000, P=0.048) and postoperative fibrin degradation product on day 1(OR=1.055, 95% CI 1.011-1.103, P=0.017) were correlated with the PVT. The PVT might happen when preoperative platelet was less than 34.5×109/L (sensitibity 80.6%, specificity 60.0%) or postoperative fibrin degradation product on day 1 was more than 64.75 mg/L (sensitibity 48.0%, specificity 91.7%).
ConclusionPreoperative platelet and postoperative fibrin degradation product on day 1 might predict PVT after splenectomy and gastroesophageal devascularization for liver cirrhosis with portal hypertension.
ObjectiveTo systematically review the early clinical prediction value of machine learning (ML) for cardiac arrest (CA).MethodsPubMed, EMbase, WanFang Data and CNKI databases were electronically searched to retrieve all ML studies on predicting CA from January 2015 to February 2021. Two reviewers independently screened literature, extracted data and assessed the risk of bias of included studies. The value of each model was evaluated based on the area under receiver operating characteristic curve (AUC) and accuracy.ResultsA total of 38 studies were included. In terms of data sources, 13 studies were based on public database, and other studies retrospectively collected clinical data, in which 21 directly predicted CA, 3 predicted CA-related arrhythmias, and 9 predicted sudden cardiac death. A total of 51 models had been adopted, among which the most popular ML methods included artificial neural network (n=11), followed by random forest (n=9) and support vector machine (n=5). The most frequently used input feature was electrocardiogram parameters (n=20), followed by age (n=12) and heart rate variability (n=10). Six studies compared the ML models with other traditional statistical models and the results showed that the AUC value of ML was generally higher than that in traditional statistical models.ConclusionsThe available evidence suggests that ML can accurately predict the occurrence of CA, and the performance is significantly superior to traditional statistical model in certain cases.
ObjectiveTo systematically evaluate the risk prediction model of knee osteoarthritis (KOA). MethodsThe CNKI, WanFang Data, VIP, PubMed, Embase, Web of Science and Cochrane Library databases were electronically searched to collect relevant studies on KOA’s risk prediction model from inception to April, 2024. After study screening and data extraction by two independent researchers, the PROBAST bias risk assessment tool was used to evaluate the bias risk and applicability of the risk prediction model. ResultsA total of 12 studies involving 21 risk prediction models for KOA were included. The number of predictors ranged from 3 to 12, and the most common predictors were age, sex, and BMI. The range of modeling AUC included in the model was 0.554-0.948, and the range of testing AUC was 0.6-0.94. The overall predictive performance of the models was mediocre and the risk of overall bias was high, and more than half of the models were not externally verified. ConclusionAt present, the overall quality and applicability of the KOA morbidity risk prediction model still have great room for improvement. Future modeling should follow the CHARMS and PROBAST to reduce the risk of bias, explore the combination of multiple modeling methods, and strengthen the external verification of the model.
ObjectiveConstructing a prediction model for seizures after stroke, and exploring the risk factors that lead to seizures after stroke. MethodsA retrospective analysis was conducted on 1 741 patients with stroke admitted to People's Hospital of Zhongjiang from July 2020 to September 2022 who met the inclusion and exclusion criteria. These patients were followed up for one year after the occurrence of stroke to observe whether they experienced seizures. Patient data such as gender, age, diagnosis, National Institute of Health Stroke Scale (NIHSS) score, Activity of daily living (ADL) score, laboratory tests, and imaging examination data were recorded. Taking the occurrence of seizures as the outcome, an analysis was conducted on the above data. The Least absolute shrinkage and selection operator (LASSO) regression analysis was used to screen predictive variables, and multivariate Logistic regression analysis was performed. Subsequently, the data were randomly divided into a training set and a validation set in a 7:3 ratio. Construct prediction model, calculate the C-index, draw nomogram, calibration plot, receiver operating characteristic (ROC) curve, and decision curve analysis (DCA) to evaluate the model's performance and clinical application value. ResultsThrough LASSO regression, nine non-zero coefficient predictive variables were identified: NIHSS score, homocysteine (Hcy), aspartate aminotransferase (AST), platelet count, hyperuricemia, hyponatremia, frontal lobe lesions, temporal lobe lesions, and pons lesions. Multivariate logistic regression analysis revealed that NIHSS score, Hcy, hyperuricemia, hyponatremia, and pons lesions were positively correlated with seizures after stroke, while AST and platelet count were negatively correlated with seizures after stroke. A nomogram for predicting seizures after stroke was established. The C-index of the training set and validation set were 0.854 [95%CI (0.841, 0.947)] and 0.838 [95%CI (0.800, 0.988)], respectively. The areas under the ROC curves were 0.842 [95%CI (0.777, 0.899)] and 0.829 [95%CI (0.694, 0.936)] respectively. Conclusion These nine variables can be used to predict seizures after stroke, and they provide new insights into its risk factors.
Objective To systematically review prediction models of small for gestational age (SGA) based on machine learning and provide references for the construction and optimization of such a prediction model. Methods The PubMed, EMbase, Web of Science, CBM, WanFang Data, VIP and CNKI databases were electronically searched to collect studies on SGA prediction models from database inception to August 10, 2022. Two researchers independently screened the literature, extracted data, evaluated the risk of bias of the included studies, and conducted a systematic review. Results A total of 14 studies, comprising 40 prediction models constructed using 19 methods, such as logical regression and random forest, were included. The results of the risk of bias assessment from 13 studies were high; the area under the curve of the prediction models ranged from 0.561 to 0.953. Conclusion The overall risk of bias in the prediction models for SGA was high, and the predictive performance was average. Models built using extreme gradient boosting (XGBoost) demonstrated the best predictive performance across different studies. The stacking method can improve predictive performance by integrating different models. Finally, maternal blood pressure, fetal abdominal circumference, head circumference, and estimated fetal weight were important predictors of SGA.
ObjectiveTo establish a predictive model of surgical site infection (SSI) following colorectal surgery using machine learning.MethodsMachine learning algorithm was used to analyze and model with the colorectal data set from Duke Infection Control Outreach Network Surveillance Network. The whole data set was divided into two parts, with 80% as the training data set and 20% as the testing data set. In order to improve the training effect, the whole data set was divided into two parts again, with 90% as the training data set and 10% as the testing data set. The predictive result of the model was compared with the actual infected cases, and the sensitivity, specificity, positive predictive value, and negative predictive value of the model were calculated, the area under receiver operating characteristic (ROC) curve was used to evaluate the predictive capacity of the model, odds ratio (OR) was calculated to tested the validity of evaluation with a significance level of 0.05.ResultsThere were 7 285 patients in the whole data set registered from January 15th, 2015 to June 16th, 2016, among whom 234 were SSI cases, with an incidence of SSI of 3.21%. The predictive model was established by random forest algorithm, which was trained by 90% of the whole data set and tested by 10% of that. The sensitivity, specificity, positive predictive value, and negative predictive value of the model were 76.9%, 59.2%, 3.3%, and 99.3%, respectively, and the area under ROC curve was 0.767 [OR=4.84, 95% confidence interval (1.32, 17.74), P=0.02].ConclusionThe predictive model of SSI following colorectal surgery established by random forest algorithm has the potential to realize semi-automatic monitoring of SSIs, but more data training should be needed to improve the predictive capacity of the model before clinical application.
