Application of artificial intelligence in a real-world research for predicting the risk of liver metastasis in T1 colorectal cancer

Background Liver is the most common metastatic site of colorectal cancer (CRC) and liver metastasis (LM) determines subsequent treatment as well as prognosis of patients, especially in T1 patients. T1 CRC patients with LM are recommended to adopt surgery and systematic treatments rather than endoscopic therapy alone. Nevertheless, there is still no effective model to predict the risk of LM in T1 CRC patients. Hence, we aim to construct an accurate predictive model and an easy-to-use tool clinically. Methods We integrated two independent CRC cohorts from Surveillance Epidemiology and End Results database (SEER, training dataset) and Xijing hospital (testing dataset). Artificial intelligence (AI) and machine learning (ML) methods were adopted to establish the predictive model. Results A total of 16,785 and 326 T1 CRC patients from SEER database and Xijing hospital were incorporated respectively into the study. Every single ML model demonstrated great predictive capability, with an area under the curve (AUC) close to 0.95 and a stacking bagging model displaying the best performance (AUC = 0.9631). Expectedly, the stacking model exhibited a favorable discriminative ability and precisely screened out all eight LM cases from 326 T1 patients in the outer validation cohort. In the subgroup analysis, the stacking model also demonstrated a splendid predictive ability for patients with tumor size ranging from one to50mm (AUC = 0.956). Conclusion We successfully established an innovative and convenient AI model for predicting LM in T1 CRC patients, which was further verified in the external dataset. Ultimately, we designed a novel and easy-to-use decision tree, which only incorporated four fundamental parameters and could be successfully applied in clinical practice.

A review of literature on risk prediction tools for hospital readmissions in older adults

Purpose The purpose of this paper is to identify and analyse the readmission risk prediction tools reported in the literature and their benefits when it comes to healthcare organisations and management.Design/methodology/approach Readmission risk prediction is a growing topic of interest with the aim of identifying patients in particular those suffering from chronic diseases such as congestive heart failure, chronic obstructive pulmonary disease and diabetes, who are at risk of readmission. Several models have been developed with different levels of predictive ability. A structured and extensive literature search of several databases was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-analysis strategy, and this yielded a total of 48,984 records.Findings Forty-three articles were selected for full-text and extensive review after following the screening process and according to the eligibility criteria. About 34 unique readmission risk prediction models were identified, in which their predictive ability ranged from poor to good (c statistic 0.5–0.86). Readmission rates ranged between 3.1 and 74.1% depending on the risk category. This review shows that readmission risk prediction is a complex process and is still relatively new as a concept and poorly understood. It confirms that readmission prediction models hold significant accuracy at identifying patients at higher risk for such an event within specific context.Research limitations/implications Since most prediction models were developed for specific populations, conditions or hospital settings, the generalisability and transferability of the predictions across wider or other contexts may be difficult to achieve. Therefore, the value of prediction models remains limited to hospital management. Future research is indicated in this regard.Originality/value This review is the first to cover readmission risk prediction tools that have been published in the literature since 2011, thereby providing an assessment of the relevance of this crucial KPI to health organisations and managers.

Development of a clinical model to predict vagus nerve stimulation response in pediatric patients with drug-resistant epilepsy

Introduction Epilepsy impacts 470,000 children in the United States, and children with epilepsy are estimated to expend 6 times more on healthcare than those without epilepsy. For patients with antiseizure medication (ASM)-resistant epilepsy and unresectable seizure foci, vagus nerve stimulation (VNS) is a treatment option. Predicting response to VNS has been historically challenging. We aimed to create a clinical prediction score which could be utilized in a routine outpatient clinical setting. Methods We performed an 11-year, single-center retrospective analysis of patients <21 years old with ASM-resistant epilepsy who underwent VNS. The primary outcome was >50% seizure frequency reduction after one year. Univariate and multivariate logistic regressions were performed to assess clinical factors associated with VNS response; 70% and 30% of the sample were used to train and validate the multivariate model, respectively. A prediction score was developed based on the multivariate regression. Sensitivity, specificity, and area under the receiver operating curve (AUC) were calculated. Results This analysis included 365 patients. Multivariate logistic regression revealed that variables associated with VNS response were: <4 years of epilepsy duration before VNS (p=0.008) and focal motor seizures (p=0.037). The variables included in the clinical prediction score were: epilepsy duration before VNS, age at seizure onset, number of pre-VNS ASMs, if VNS was the patient's first therapeutic epilepsy surgery, and predominant seizure semiology. The final AUC was 0.7013 for the "fitted" sample and 0.6159 for the "validation" sample. Conclusions We developed a clinical model to predict VNS response in one of the largest samples of pediatric VNS patients to date. While the presented clinical prediction model demonstrated an acceptable AUC in the training cohort, clinical variables alone likely do not accurately predict VNS response. This score may be useful upon further validation, though its predictive ability underscores the need for more robust biomarkers of treatment response.

Spatiotemporal variations in exposure: Chagas disease in Colombia as a case study

Age-stratified serosurvey data are often used to understand spatiotemporal trends in disease incidence and exposure through estimating the Force-of-Infection (FoI). Typically, median or mean FoI estimates are used as the response variable in predictive models, often overlooking the uncertainty in estimated FoI values when fitting models and evaluating their predictive ability. To assess how this uncertainty impact predictions, we compared three approaches with three levels of uncertainty integration. We propose a performance indicator to assess how predictions reflect initial uncertainty.In Colombia, 76 serosurveys (1980–2014) conducted at municipality level provided age-stratified Chagas disease prevalence data. The yearly FoI was estimated at the serosurvey level using a time-varying catalytic model. Environmental, demographic and entomological predictors were used to fit and predict the FoI at municipality level from 1980 to 2010 across Colombia.A stratified bootstrap method was used to fit the models without temporal autocorrelation at the serosurvey level. The predictive ability of each model was evaluated to select the best-fit models within urban, rural and (Amerindian) indigenous settings. Model averaging, with the 10 best-fit models identified, was used to generate predictions.Our analysis shows a risk of overconfidence in model predictions when median estimates of FoI alone are used to fit and evaluate models, failing to account for uncertainty in FoI estimates. Our proposed methodology fully propagates uncertainty in the estimated FoI onto the generated predictions, providing realistic assessments of both central tendency and current uncertainty surrounding exposure to Chagas disease.

Development and validation of a prediction index for recent mortality in advanced COPD patients

The primary barrier to initiating palliative care for advanced COPD patients is the unpredictable course of the disease. We enroll 752 COPD patients into the study and validate the prediction tools for 1-year mortality using the current guidelines for palliative care. We also develop a composite prediction index for 1-year mortality and validate it in another cohort of 342 patients. Using the current prognostic models for recent mortality in palliative care, the best area under the curve (AUC) for predicting mortality is 0.68. Using the Modified Medical Research Council dyspnea score and oxygen saturation to define the combined dyspnea and oxygenation (DO) index, we find that the AUC of the DO index is 0.84 for predicting mortality in the validated cohort. Predictions of 1-year mortality based on the current palliative care guideline for COPD patients are poor. The DO index exhibits better predictive ability than other models in the study.

Validation of CSR model to predict stroke risk after transient ischemic attack

It is essential to identify high risk transient ischemic attack (TIA) patients. The previous study reported that the CSR (comprehensive stroke recurrence) model, a neuroimaging model, had a high predictive ability of recurrent stroke. The aims of this study were to validate the predictive value of CSR model in TIA patients and compare the predictive ability with ABCD3-I score. Data were analyzed from the prospective hospital-based database of patients with TIA which defined by the World Health Organization time-based criteria. The predictive outcome was stroke occurrence at 90 days. The receiver-operating characteristic (ROC) curves were plotted and the C statistics were calculated as a measure of predictive ability. Among 1186 eligible patients, the mean age was 57.28 ± 12.17 years, and 474 (40.0%) patients had positive diffusion-weighted imaging (DWI). There were 118 (9.9%) patients who had stroke within 90 days. In 1186 TIA patients, The C statistic of CSR model (0.754; 95% confidence interval [CI] 0.729–0.778) was similar with that of ABCD3-I score (0.717; 95% CI 0.691–0.743; Z = 1.400; P = 0.1616). In 474 TIA patients with positive DWI, C statistic of CSR model (0.725; 95% CI 0.683–0.765) was statistically higher than that of ABCD3-I score (0.626; 95% CI 0.581–0.670; Z = 2.294; P = 0.0245). The CSR model had good predictive value for assessing stroke risk after TIA, and it had a higher predictive value than ABCD3-I score for assessing stroke risk for TIA patients with positive DWI.

The prognostic value of m6A-related LncRNAs in patients with HNSCC: bioinformatics analysis of TCGA database

N6-methyladenosine (m6A) modifications play an essential role in tumorigenesis. These modifications modulate RNAs, including mRNAs and lncRNAs. However, the prognostic role of m6A-related lncRNAs in head and neck squamous cell carcinoma (HNSCC) is poorly understood. Based on LASSO Cox regression, enrichment analysis, univariate and multivariate Cox regression analysis, a prognostic risk model, and consensus clustering analysis, we analyzed 12 m6A-related lncRNAs in HNSCC sample data from The Cancer Genome Atlas (TCGA) database. We found 12 m6A-related lncRNAs in the training cohort and validated them in all cohorts by Kaplan–Meier and Cox regression analyses, revealing their independent prognostic value in HNSCC. Moreover, ROC analysis was conducted, confirming the strong predictive ability of this signature for HNSCC survival. GSEA and detailed immune infiltration analyses revealed specific pathways associated with m6A-related lncRNAs. In this study, a novel risk model including twelve genes (SAP30L-AS1, AC022098.1, LINC01475, AC090587.2, AC008115.3, AC015911.3, AL122035.2, AC010226.1, AL513190.1, ZNF32-AS1, AL035587.1 and AL031716.1) was built. It could accurately predict HNSCC outcomes and could provide new therapeutic targets for HNSCC patients.

Research on prediction of China’s financial systematic risk based on the hybrid model

An important feature of the outbreak of systemic financial risk is that the linkage and contagion of risk amongst the various sub-markets of the financial system have increased significantly. In addition, research on the prediction of systemic financial risk plays a significant role in the sustainable development of the financial market. Therefore, this paper takes China’s financial market as its research object, considers the risks co-activity among major financial sub-markets, and constructs a financial composite indicator of systemic stress (CISS) for China, describing its financial systemic stress based on 12 basic indicators selected from the money market, bond market, stock market, and foreign exchange market. Furthermore, drawing on the decomposition and integration technology in the TEI@I complex system research methodology, this paper introduces advanced variational mode decomposition (VMD) technology and extreme learning machine (ELM) algorithms, constructing the VMD-DE-ELM hybrid model to predict the systemic risk of China’s financial market. According to e RMSE , e MAE , and e MAPE , the prediction model’s multistep-ahead forecasting effect is evaluated. The empirical results show that the China’s financial CISS constructed in this paper can effectively identify all kinds of risk events in the sample range. The results of a robustness test show that the overall trend of China’s financial CISS and its ability to identify risk events are not affected by parameter selection and have good robustness. In addition, compared with the benchmark model, the VMD-DE-ELM hybrid model constructed in this paper shows superior predictive ability for systemic financial risk.

High glycemic albumin representing prestroke glycemic variability is associated with hemorrhagic transformation in patients receiving intravenous thrombolysis

We evaluated the impact of prestroke glycemic variability estimated by glycated albumin (GA) on symptomatic hemorrhagic transformation (SHT) in patients with intravenous thrombolysis (IVT). Using a multicenter database, we consecutively enrolled acute ischemic stroke patients receiving IVT. A total of 378 patients were included in this study. Higher GA was defined as GA ≥ 16.0%. The primary outcome measure was SHT. Multivariate regression analysis and a receiver operating characteristic curve were used to assess risks and predictive ability for SHT. Among the 378 patients who were enrolled in this study, 27 patients (7.1%) had SHT as defined by the Safe Implementation of Thrombolysis in Stroke-Monitoring Study (SHTSITS). The rate of SHTSITS was higher in the higher GA group than in the lower GA group (18.0% vs. 1.6%, p < 0.001). A higher GA level (GA ≥ 16.0%) significantly increased the risk of SHTSITS (adjusted odds ratio [OR], [95% confidence interval, CI], 12.57 [3.08–41.54]) in the logistic regression analysis. The predictive ability of the GA level for SHTSITS was good (AUC [95% CI]: 0.83 [0.77–0.90], p < 0.001), and the cutoff value of GA in SHT was 16.3%. GA was a reliable predictor of SHT after IVT in acute ischemic stroke in this study.