Conventional Cox Proportional Hazards Research Articles

Mixture and Non-Mixture Cure Models for the survival analysis of SARS-CoV-2 patients in Khyber Pakhtunkhwa, Pakistan.

To examine the potential difference in survival and risk of death between asymptomatic and symptomatic SARS-CoV-2 patients, controlled by age and gender for all the attendance in hospitals of Khyber Pakhtunkhwa (KP), Pakistan. In this retrospective study, the medical records of 6273 SARS-CoV-2 patients admitted to almost all hospitals in Khyber Pakhtunkhwa during the first wave of the coronavirus outbreak from March to June 2020 were analysed. The effects of gender, age, and being symptomatic on the survival of SARS-CoV-2 patients were assessed using cure-survival models as opposed to the conventional Cox proportional hazards model. The prevalence of initially symptomatic patients was 55.8%, and the overall mortality rate was 11.8%. The fitted cure-survival models suggest that age affects the probability of death (incidence) but not the short-term survival time of patients (latency); symptomatic patients have a higher risk of death than their asymptomatic counterparts, but the survival time of symptomatic patients is longer on average; gender has no significant effect on the probability of death and survival time. The available data and statistical results suggest that asymptomatic and young patients are generally less susceptible to initial infection with SARS-CoV-2 and therefore have a lower risk of death. Our regression models show that uncured asymptomatic patients generally have poorer short-term survival than their uncured symptomatic counterparts. The association between gender and survival outcome was not significant.

Fracture risk assessment in the presence of competing risk of death.

To identify the optimal statistical approach for predicting the risk of fragility fractures in the presence of competing event of death. We used real-world data from the Dubbo Osteoporosis Epidemiology Study that has monitored 3035 elderly participants for bone health and mortality. Fragility fractures were ascertained radiologically. Mortality was confirmed by the State Registry. We considered four statistical models for predicting fracture risk: (i) conventional Cox's proportional hazard model, (ii) cause-specific model, (iii) Fine-Gray sub-distribution model, and (iv) multistate model. These models were fitted and validated in the development (60% of the original sample) and validation (40%) subsets, respectively. The model performance was assessed by discrimination and calibration analyses. During a median follow-up of 11.3 years (IQR: 7.2, 16.2), 628 individuals (34.5%) in the development cohort fractured, and 630 (34.6%) died without a fracture. Neither the discrimination nor the 5-year prediction performance was significantly different among the models, though the conventional model tended to overestimate fracture risk (calibration-in-the-large index = - 0.24; 95% CI: - 0.43, - 0.06). For 10-year risk prediction, the multistate model (calibration-in-the-large index = - 0.05; 95% CI: - 0.20, 0.10) outperformed the cause-specific (- 0.23; - 0.30, - 0.08), Fine-Gray (- 0.31; - 0.46, - 0.16), and conventional model (- 0.54; - 0.70, - 0.39) which significantly overestimated fracture risk. Adjustment for competing risk of death has minimum impact on the short-term prediction of fracture. However, the multistate model yields the most accurate prediction of long-term fracture risk and should be considered for predictive research in the elderly, who are also at high mortality risk. Fracture risk assessment might be compromised by the competing event of death. This study, using real-world data found a multistate model was superior to the current competing risk methods in fracture risk assessment. A multistate model is considered an optimal statistical method for predictive research in the elderly.

On semiparametric accelerated failure time models with time-varying covariates: A maximum penalised likelihood estimation.

The accelerated failure time (AFT) model offers an important and useful alternative to the conventional Cox proportional hazards model, particularly when the proportional hazards assumption for a Cox model is violated. Since an AFT model is basically a log-linear model, meaningful interpretations of covariate effects on failure times can be made directly. However, estimation of a semiparametric AFT model imposes computational challenges even when it only has time-fixed covariates, and the situation becomes much more complicated when time-varying covariates are included. In this paper, we propose a penalised likelihood approach to estimate the semiparametric AFT model with right-censored failure time, where both time-fixed and time-varying covariates are permitted. We adopt the Gaussian basis functions to construct a smooth approximation to the nonparametric baseline hazard. This model fitting method requires a constrained optimisation approach. A comprehensive simulation study is conducted to demonstrate the performance of the proposed method. An application of our method to a motor neuron disease data set is provided.

Machine learning for individualized prediction of hepatocellular carcinoma development after the eradication of hepatitis C virus with antivirals

Accurate risk stratification for hepatocellular carcinoma (HCC) after achieving a sustained viral response (SVR) is necessary for optimal surveillance. We aimed to develop and validate a machine learning (ML) model to predict the risk of HCC after achieving an SVR in individual patients. In this multicenter cohort study, 1742 patients with chronic hepatitis C who achieved an SVR were enrolled. Five ML models were developed including DeepSurv, gradient boosting survival analysis, random survival forest (RSF), survival support vector machine, and a conventional Cox proportional hazard model. Model performance was evaluated using Harrel' c-index and was externally validated in an independent cohort (977 patients). During the mean observation period of 5.4 years, 122 patients developed HCC (83 in the derivation cohort and 39 in the external validation cohort). The RSF model showed the best discrimination ability using seven parameters at the achievement of an SVR with a c-index of 0.839 in the external validation cohort and a high discriminative ability when the patients were categorized into three risk groups (P <0.001). Furthermore, this RSF model enabled the generation of an individualized predictive curve for HCC occurrence for each patient with an app available online. We developed and externally validated an RSF model with good predictive performance for the risk of HCC after an SVR. The application of this novel model is available on the website. This model could provide the data to consider an effective surveillance method. Further studies are needed to make recommendations for surveillance policies tailored to the medical situation in each country. A novel prediction model for HCC occurrence in patients after hepatitis C virus eradication was developed using machine learning algorithms. This model, using seven commonly measured parameters, has been shown to have a good predictive ability for HCC development and could provide a personalized surveillance system.

Analysis of Antibiotic Exposure and Development of Acute Graft-vs-Host Disease Following Allogeneic Hematopoietic Cell Transplantation

Certain antibiotic exposures have been associated with increased rates of acute graft-vs-host disease (aGVHD) after allogeneic hematopoietic cell transplantation (allo-HCT). Since antibiotic exposure can both affect and be affected by infections, analyzing time-dependent exposure in the presence of multiple potential confounders, including prior antibiotic exposures, poses specific analytical challenges, necessitating both a large sample size and unique approaches. To identify antibiotics and antibiotic exposure timeframes associated with subsequent aGVHD. This cohort study assessed allo-HCT at a single center from 2010 to 2021. Participants included all patients aged at least 18 years who underwent their first T-replete allo-HCT, with at least 6 months of follow-up. Data were analyzed from August 1 to December 15, 2022. Antibiotics between 7 days before and 30 days after transplant. The primary outcome was grade II to IV aGVHD. The secondary outcome was grade III to IV aGVHD. Data were analyzed using 3 orthogonal methods: conventional Cox proportional hazard regression, marginal structural models, and machine learning. A total of 2023 patients (median [range] age, 55 [18-78] years; 1153 [57%] male) were eligible. Weeks 1 and 2 after HCT were the highest-risk intervals, with multiple antibiotic exposures associated with higher rates of subsequent aGVHD. In particular, exposure to carbapenems during weeks 1 and 2 after allo-HCT was consistently associated with increased risk of aGVHD (minimum hazard ratio [HR] among models, 2.75; 95% CI, 1.77-4.28), as was week 1 after allo-HCT exposure to combinations of penicillins with a β-lactamase inhibitor (minimum HR among models, 6.55; 95% CI, 2.35-18.20). In this cohort study of allo-HCT recipients, antibiotic choices and schedules in the early course of transplantation were associated with aGVHD rates. These findings should be considered in antibiotic stewardship programs.

Machine-learning predicts time-series prognosis factors in metastatic prostate cancer patients treated with androgen deprivation therapy

Machine learning technology is expected to support diagnosis and prognosis prediction in medicine. We used machine learning to construct a new prognostic prediction model for prostate cancer patients based on longitudinal data obtained from age at diagnosis, peripheral blood and urine tests of 340 prostate cancer patients. Random survival forest (RSF) and survival tree were used for machine learning. In the time-series prognostic prediction model for metastatic prostate cancer patients, the RSF model showed better prediction accuracy than the conventional Cox proportional hazards model for almost all time periods of progression-free survival (PFS), overall survival (OS) and cancer-specific survival (CSS). Based on the RSF model, we created a clinically applicable prognostic prediction model using survival trees for OS and CSS by combining the values of lactate dehydrogenase (LDH) before starting treatment and alkaline phosphatase (ALP) at 120 days after treatment. Machine learning provides useful information for predicting the prognosis of metastatic prostate cancer prior to treatment intervention by considering the nonlinear and combined impacts of multiple features. The addition of data after the start of treatment would allow for more precise prognostic risk assessment of patients and would be beneficial for subsequent treatment selection.

Comparison of prediction ability of two extended Cox models in nonlinear survival data analysis

To compare the predictive ability of two extended Cox models in nonlinear survival data analysis. Through Monte Carlo simulation and empirical study and with the conventional Cox Proportional Hazards model and Random Survival Forests as the reference models, we compared restricted cubic spline Cox model (Cox_RCS) and DeepSurv neural network Cox model (Cox_DNN) for their prediction ability in nonlinear survival data analysis. Concordance index was used to evaluate the differentiation of the prediction results (a larger concordance index indicates a better prediction ability of the model). Integrated Brier Score was used to evaluate the calibration degree of the prediction (a smaller index indicates a better prediction ability). For data that met requirement of the proportion risk, the Cox_RCS model had the best prediction ability regardless of the sample size or deletion rate. For data that failed to meet the proportion risk, the prediction ability of Cox_DNN was optimal for a large sample size (≥500) with a low deletion (< 40%); the prediction ability of Cox_RCS was superior to those of other models in all other scenarios. For example data, the Cox_RCS model showed the best performance. In analysis of nonlinear low maintenance data, Cox_RCS and Cox_DNN have their respective advantages and disadvantages in prediction. The conventional survival analysis methods are not inferior to machine learning or deep learning methods under certain conditions.

A comparison of machine learning algorithms and traditional regression-based statistical modeling for predicting hypertension incidence in a Canadian population

Risk prediction models are frequently used to identify individuals at risk of developing hypertension. This study evaluates different machine learning algorithms and compares their predictive performance with the conventional Cox proportional hazards (PH) model to predict hypertension incidence using survival data. This study analyzed 18,322 participants on 24 candidate features from the large Alberta’s Tomorrow Project (ATP) to develop different prediction models. To select the top features, we applied five feature selection methods, including two filter-based: a univariate Cox p-value and C-index; two embedded-based: random survival forest and least absolute shrinkage and selection operator (Lasso); and one constraint-based: the statistically equivalent signature (SES). Five machine learning algorithms were developed to predict hypertension incidence: penalized regression Ridge, Lasso, Elastic Net (EN), random survival forest (RSF), and gradient boosting (GB), along with the conventional Cox PH model. The predictive performance of the models was assessed using C-index. The performance of machine learning algorithms was observed, similar to the conventional Cox PH model. Average C-indexes were 0.78, 0.78, 0.78, 0.76, 0.76, and 0.77 for Ridge, Lasso, EN, RSF, GB and Cox PH, respectively. Important features associated with each model were also presented. Our study findings demonstrate little predictive performance difference between machine learning algorithms and the conventional Cox PH regression model in predicting hypertension incidence. In a moderate dataset with a reasonable number of features, conventional regression-based models perform similar to machine learning algorithms with good predictive accuracy.

Hyperparathyroidism Is an Independent Risk Factor for Allograft Dysfunction in Pediatric Kidney Transplantation

Little is known about the consequences of deranged chronic kidney disease-mineral and bone disorder (CKD-MBD) parameters on kidney allograft function in children. We examined a relationship between these parameters over time and allograft outcome. This registry study from the Cooperative European Paediatric Renal Transplant Initiative (CERTAIN) collected data at baseline, months 1, 3, 6, 9, and 12 after transplant; and every 6 months thereafter up to 5 years. Survival analysis for a composite end point of graft loss or estimated glomerular filtration rate (eGFR)≤30 ml/min per 1.73 m2 or a≥50% decline from eGFR at month 1 posttransplant was performed. Associations of parathyroid hormone (PTH), calcium, phosphate, and 25-hydroxyvitamin D (25(OH)D) with allograft outcome were investigated using conventional stratified Cox proportional hazards models and further verified with marginal structural models with time-varying covariates. We report on 1210 patients (61% boys) from 16 European countries. The composite end point was reached in 250 grafts (21%), of which 11 (4%) were allograft losses. In the conventional Cox proportional hazards models adjusted for potential confounders, only hyperparathyroidism (hazard ratio [HR], 2.94; 95% confidence interval [CI], 1.82-4.74) and hyperphosphatemia (HR, 1.94; 95% CI, 1.28-2.92) were associated with the composite end point. Marginal structural models showed similar results for hyperparathyroidism (HR, 2.74; 95% CI, 1.71-4.38), whereas hyperphosphatemia was no longer significant (HR, 1.35; 95% CI, 0.87-2.09), suggesting that its association with graft dysfunction can be ascribed to a decline in eGFR. Hyperparathyroidism is a potential independent risk factor for allograft dysfunction in children.

Deep learning-based relapse prediction of neuromyelitis optica spectrum disorder with anti-aquaporin-4 antibody.

ObjectiveWe previously identified the independent predictors of recurrent relapse in neuromyelitis optica spectrum disorder (NMOSD) with anti-aquaporin-4 antibody (AQP4-ab) and designed a nomogram to estimate the 1- and 2-year relapse-free probability, using the Cox proportional hazard (Cox-PH) model, assuming that the risk of relapse had a linear correlation with clinical variables. However, whether the linear assumption fits real disease tragedy is unknown. We aimed to employ deep learning and machine learning to develop a novel prediction model of relapse in patients with NMOSD and compare the performance with the conventional Cox-PH model.MethodsThis retrospective cohort study included patients with NMOSD with AQP4-ab in 10 study centers. In this study, 1,135 treatment episodes from 358 patients in Huashan Hospital were employed as the training set while 213 treatment episodes from 92 patients in nine other research centers as the validation set. We compared five models with added variables of gender, AQP4-ab titer, previous attack under the same therapy, EDSS score at treatment initiation, maintenance therapy, age at treatment initiation, disease duration, the phenotype of the most recent attack, and annualized relapse rate (ARR) of the most recent year by concordance index (C-index): conventional Cox-PH, random survival forest (RSF), LogisticHazard, DeepHit, and DeepSurv.ResultsWhen including all variables, RSF outperformed the C-index in the training set (0.739), followed by DeepHit (0.737), LogisticHazard (0.722), DeepSurv (0.698), and Cox-PH (0.679) models. As for the validation set, the C-index of LogisticHazard outperformed the other models (0.718), followed by DeepHit (0.704), DeepSurv (0.698), RSF (0.685), and Cox-PH (0.651) models. Maintenance therapy was calculated to be the most important variable for relapse prediction.ConclusionThis study confirmed the superiority of deep learning to design a prediction model of relapse in patients with AQP4-ab-positive NMOSD, with the LogisticHazard model showing the best predictive power in validation.

Generalized Structural Equation Model with Survival Outcomes and Time-Varying Coefficients

The conventional Cox proportional hazards (PH) model typically assumes fully observed predictors and constant regression coefficients. However, some predictors are latent variables, each of which must be characterized by multiple observed indicators from various perspectives. Moreover, the predictor effects may vary with time in practice. Accommodating such latent variables and identifying temporal covariate effects are frequently of primary interest. This study proposes a generalized structural equation model to investigate the temporal effects of observed and latent risk factors on the hazards of interest. The proposed model comprises a confirmatory factor analysis model as the measurement equation and a varying-coefficient PH model with observed and latent predictors as the structural equation. A hybrid procedure that combines the expectation-maximization (EM) algorithm and the corrected estimating equation approach is developed to estimate unknown parameters and coefficient functions. Simulation studies demonstrate the satisfactory performance of the proposed method. An application to a health survey study reveals insights into risk factors for elders’ life expectancy.

Prediction and Risk Stratification of Cardiovascular Disease in Diabetic Kidney Disease Patients.

BackgroundDiabetic kidney disease (DKD) patients are facing an extremely high risk of cardiovascular disease (CVD), which is a major cause of death for DKD patients. We aimed to build a deep learning model to predict CVD risk among DKD patients and perform risk stratifying, which could help them perform early intervention and improve personal health management.MethodsA retrospective cohort study was conducted to assess the risk of the occurrence of composite cardiovascular disease, which includes coronary heart disease, cerebrovascular diseases, congestive heart failure, and peripheral artery disease, in DKD patients. A least absolute shrinkage and selection operator (LASSO) regression was used to perform the variable selection. A deep learning-based survival model called DeepSurv, based on a feed-forward neural network was developed to predict CVD risk among DKD patients. We compared the model performance with the conventional Cox proportional hazards (CPH) model and the Random survival forest (RSF) model using the concordance index (C-index), the area under the curve (AUC), and integrated Brier scores (IBS).ResultsWe recruited 890 patients diagnosed with DKD in this retrospective study. During a median follow-up of 10.4 months, there are 289 patients who sustained a subsequent CVD. Seven variables, including age, high density lipoprotein (HDL), hemoglobin (Hb), systolic blood pressure (SBP), smoking status, 24 h urinary protein excretion, and total cholesterol (TC), chosen by LASSO regression were used to develop the predictive model. The DeepSurv model showed the best performance, achieved a C-index of 0.767(95% confidence intervals [CI]: 0.717–0.817), AUC of 0.780(95%CI: 0.721–0.839), and IBS of 0.067 in the validation set. Then we used the cut-off value determined by ROC (receiver operating characteristic) curve to divide the patients into different risk groups. Moreover, the DeepSurv model was also applied to develop an online calculation tool for patients to conduct risk monitoring.ConclusionA deep-learning-based predictive model using seven clinical variables can effectively predict CVD risk among DKD patients and perform risk stratification. An online calculator allows its easy implementation.

Machine learning–based personalized prediction of gastric cancer incidence using the endoscopic and histologic findings at the initial endoscopy

Accurate risk stratification for gastric cancer is required for optimal endoscopic surveillance in patients with chronic gastritis. We aimed to develop a machine learning (ML) model that incorporates endoscopic and histologic findings for an individualized prediction of gastric cancer incidence. We retrospectively evaluated 1099 patients with chronic gastritis who underwent EGD and biopsy samplingof the gastric mucosa. Patients were randomly divided into training and test sets (4:1). We constructed a conventional Cox proportional hazard model and 3 ML models. Baseline characteristics, endoscopic atrophy, and Operative Link on Gastritis-Intestinal Metaplasia Assessment (OLGIM)/Operative Link on Gastritis Assessment (OLGA) stage at initial EGD were comprehensively assessed. Model performance was evaluated using Harrel's c-index. During a mean follow-up of 5.63 years, 94 patients (8.55%) developed gastric cancer. The gradient-boosting decision tree (GBDT) model achieved the best performance (c-index from the test set, .84) and showed high discriminative ability in stratifying the test set into 3 risk categories (P< .001). Age, OLGIM/OLGA stage, endoscopic atrophy, and history of malignant tumors other than gastric cancer were important predictors of gastric cancer incidence in the GBDT model. Furthermore, the proposed GBDT model enabled the generation of a personalized cumulative incidence prediction curve for each patient. We developed a novel ML model that incorporates endoscopic and histologic findings at initial EGD for personalized risk prediction of gastric cancer. This model may lead to the development of effective and personalized follow-up strategies after initial EGD.

Machine Learning-Based Personalized Prediction of Hepatocellular Carcinoma Recurrence After Radiofrequency Ablation.

Radiofrequency ablation (RFA) is a widely accepted, minimally invasive treatment for hepatocellular carcinoma (HCC). This study aimed to develop a machine learning (ML) model to predict the risk of HCC recurrence after RFA treatment for individual patients. We included a total of 1778 patients with treatment-naïve HCC who underwent RFA. The cumulative probability of overall recurrence after the initial RFA treatment was 78.9% and 88.0% at 5 and 10 years, respectively. We developed a conventional Cox proportional hazard model and 6 ML models-including the deep learning-based DeepSurv model. Model performance was evaluated using Harrel's c-index and was validated externally using the split-sample method. The gradient boosting decision tree (GBDT) model achieved the best performance with a c-index of 0.67 from external validation, and it showed a high discriminative ability in stratifying the external validation sample into 2, 3, and 4 different risk groups (P < .001 among all risk groups). The c-index of DeepSurv was 0.64. In order of significance, the tumor number, serum albumin level, and des-gamma-carboxyprothrombin level were the most important variables for the prediction of HCC recurrence in the GBDT model. Also, the current GBDT model enabled the output of a personalized cumulative recurrence prediction curve for each patient. We developed a novel ML model for the personalized risk prediction of HCC recurrence after RFA treatment. The current model may lead to the personalization of effective follow-up strategies after RFA treatment according to the risk stratification of HCC recurrence.

Bi-Centric Independent Validation of Outcome Prediction after Radioembolization of Primary and Secondary Liver Cancer.

Background: Yttrium-90 radioembolization (RE) plays an important role in the treatment of liver malignancies. Optimal patient selection is crucial for an effective and safe treatment. In this study, we aim to validate the prognostic performance of a previously established random survival forest (RSF) with an external validation cohort from a different national center. Furthermore, we compare outcome prediction models with different established metrics. Methods: A previously established RSF model, trained on a consecutive cohort of 366 patients who had received RE due to primary or secondary liver tumor at a national center (center 1), was used to predict the outcome of an independent consecutive cohort of 202 patients from a different national center (center 2) and vice versa. Prognostic performance was evaluated using the concordance index (C-index) and the integrated Brier score (IBS). The prognostic importance of designated baseline parameters was measured with the minimal depth concept, and the influence on the predicted outcome was analyzed with accumulated local effects plots. RSF values were compared to conventional cox proportional hazards models in terms of C-index and IBS. Results: The established RSF model achieved a C-index of 0.67 for center 2, comparable to the results obtained for center 1, which it was trained on (0.66). The RSF model trained on center 2 achieved a C-index of 0.68 on center 2 data and 0.66 on center 1 data. CPH models showed comparable results on both cohorts, with C-index ranging from 0.68 to 0.72. IBS validation showed more differentiated results depending on which cohort was trained on and which cohort was predicted (range: 0.08 to 0.20). Baseline cholinesterase was the most important variable for survival prediction. Conclusion: The previously developed predictive RSF model was successfully validated with an independent external cohort. C-index and IBS are suitable metrics to compare outcome prediction models, with IBS showing more differentiated results. The findings corroborate that survival after RE is critically determined by functional hepatic reserve and thus baseline liver function should play a key role in patient selection.

Prognostic impact of six versus eight cycles of standard regimen in patients with diffuse large B-cell lymphoma: propensity score-matching analysis

BackgroundR-CHOP-21 has been the standard treatment for diffuse large B-cell lymphoma (DLBCL), but there is a paucity of evidence focusing on the number of cycles of regimens.Patients and methodsWe conducted a retrospective study to compare the effectiveness of six cycles of standard regimens versus eight cycles for overall survival (OS) in DLBCL patients using propensity score matching, in consideration of relative dose intensity (RDI).ResultsA total of 685 patients with newly diagnosed DLBCL were identified in three institutions from 2007 to 2017. Patients treated using six cycles of standard regimens were matched by propensity scores with those treated using eight cycles. A 1 : 1 propensity score matching yielded 138 patient pairs. Eight cycles did not significantly improve OS in the conventional Cox proportional hazards model (hazard ratio 0.849, 95% confidence interval 0.453-1.588, P = 0.608). Restricted cubic spline Cox models for OS confirmed that the effect of the number of cycles was not modified by total average RDI, the International Prognostic Index, and age. Occurrence of adverse events did not differ between six and eight cycles.ConclusionEven considering the impact of RDI, six cycles of the initial standard regimen for DLBCL is not inferior to eight cycles.

Brachial-ankle pulse wave velocity and cardio-ankle vascular index are associated with future cardiovascular events in a general population: The Nagahama Study.

Faster pulse wave velocity (PWV) is known to be associated with the incidence of cardiovascular diseases (CVD). The aim of this study was to clarify the hypothesis that PWV may be associated with future CVD events even when its time‐dependent changes were adjusted. We also investigated a prognostic significance of cardio‐ankle vascular index, another index of arterial stiffness. Study participants included 8850 community residents. The repeated measures of the clinical parameters at 5.0 years after the baseline were available for 7249 of the participants. PWV was calculated using the arterial waveforms measured at the brachia and ankles (baPWV). The cardio‐ankle vascular index was calculated by estimated pulse transit time from aortic valve to tibial artery. During the 8.53 years follow‐up period, we observed 215 cases of CVD. The incidence rate increased linearly with baPWV quartiles (per 10 000 person‐years: Q1, 2.7; Q2, 12.6; Q3, 22.5; Q4, 76.2), and the highest quartile was identified as an independent determinant of incident CVD by conventional Cox proportional hazard analysis adjusted for known risk factors [hazard ratio (HR), 4.00; p = .007]. Per unit HR of baPWV (HR, 1.15; p < .001) remained significant in the time‐dependent Cox regression analysis including baPWV and other clinical values measured at 5‐year after the baseline as time‐varying variables (HR, 1.14; p < .001). The cardio‐ankle vascular index was also associated with CVD with similar manner though the associations were less clear than that of baPWV. baPWV is a good risk marker for the incidence of CVD.

Two-stage Cox-nnet: biologically interpretable neural-network model for prognosis prediction and its application in liver cancer survival using histopathology and transcriptomic data.

Pathological images are easily accessible data with the potential of prognostic biomarkers. Moreover, integration of heterogeneous data types from multi-modality, such as pathological image and gene expression data, is invaluable to help predicting cancer patient survival. However, the analytical challenges are significant. Here, we take the hepatocellular carcinoma (HCC) pathological image features extracted by CellProfiler, and apply them as the input for Cox-nnet, a neural network-based prognosis prediction model. We compare this model with the conventional Cox proportional hazards (Cox-PH) model, CoxBoost, Random Survival Forests and DeepSurv, using C-index and log-rank P-values. The results show that Cox-nnet is significantly more accurate than Cox-PH and Random Survival Forests models and comparable with CoxBoost and DeepSurv models, on pathological image features. Further, to integrate pathological image and gene expression data of the same patients, we innovatively construct a two-stage Cox-nnet model, and compare it with another complex neural-network model called PAGE-Net. The two-stage Cox-nnet complex model combining histopathology image and transcriptomic RNA-seq data achieves much better prognosis prediction, with a median C-index of 0.75 and log-rank P-value of 6e−7 in the testing datasets, compared to PAGE-Net (median C-index of 0.68 and log-rank P-value of 0.03). Imaging features present additional predictive information to gene expression features, as the combined model is more accurate than the model with gene expression alone (median C-index 0.70). Pathological image features are correlated with gene expression, as genes correlated to top imaging features present known associations with HCC patient survival and morphogenesis of liver tissue. This work proposes two-stage Cox-nnet, a new class of biologically relevant and interpretable models, to integrate multiple types of heterogenous data for survival prediction.

Factors affecting the survival of patients with colorectal cancer using random survival forest.

Colorectal cancer is one of the most common cancers and the leading cause of cancer death in Iran. This study aimed to develop and validate a random survival forest (RSF) to identify important risk factors on mortality in colorectal patients based on their demographic and clinical-related variables. In this retrospective cohort study, the information of 317 patients with colorectal cancer who were referred to Imam Khomeini Clinic of Hamadan during the years of 2002 to 2017 were examined. Patient survival was calculated from the time of diagnosis to death. In the present study, the RSF model was used to identify factors affecting patient survival. Also, the results of the RSF model were compared with the Cox model. The data were analyzed using R software (version 3.6.1) and survival packages. One-, 2-, 3-, 4-, 5-, and 10-year survival rates of included patients were 81.4%, 63%, 57%, 52%, 45%, and 34%, respectively, and the median survival was obtained to be 53 months. The number of 150 patients was died at this time period. The four most important predictors of survival included metastasis to other organs, WBC count, disease stage, and number of lymphomas involved. RSF method predicted survival better than the conventional Cox proportional hazard model. We found that metastasis to other organs, WBC count, disease stage, and number of lymphomas involved were the most four most important predictors of low survival for colorectal cancer patients.

Predictors of mortality among hemodialysis patients in Hamadan province using random survival forests.

SummaryBackgroundHemodialysis patients are at a high risk for morbidity and mortality. This study aimed to find the predictors of mortality and survival in hemodialysis patients in Hamadan province of Iran.MethodsA number of 785 patients during the entire 10 years were enrolled into this historical cohort study. Data were gathered by a checklist of hospital records. The survival time was the time between the start of hemodialysis treatment to patient’s death as the end point. Random survival forests (RSF) method was used to identify the main predictors of survival among the patients.ResultsThe median survival time was 613 days. The number of 376 deaths was occurred. The three most important predictors of survival were hemoglobin, CRP and albumin. RSF method predicted survival better than the conventional Cox-proportional hazards model (out-of-bag C-index of 0.808 for RSF vs. 0.727 for Cox model).ConclusionsWe found that positivity of CRP, low serum albumin and low serum hemoglobin were the top three most important predictors of low survival for HD patients