Random Survival Forest Model Research Articles

AI-Based Risk Score from Tumour-Infiltrating Lymphocyte Predicts Locoregional-Free Survival in Nasopharyngeal Carcinoma.

Locoregional recurrence of nasopharyngeal carcinoma (NPC) occurs in 10% to 50% of cases following primary treatment. However, the current main prognostic markers for NPC, both stage and plasma Epstein-Barr virus DNA, are not sensitive to locoregional recurrence. We gathered 385 whole-slide images (WSIs) from haematoxylin and eosin (H&E)-stained NPC sections (n = 367 cases), which were collected from Sun Yat-sen University Cancer Centre. We developed a deep learning algorithm to detect tumour nuclei and lymphocyte nuclei in WSIs, followed by density-based clustering to quantify the tumour-infiltrating lymphocytes (TILs) into 12 scores. The Random Survival Forest model was then trained on the TILs to generate risk score. Based on Kaplan-Meier analysis, the proposed methods were able to stratify low- and high-risk NPC cases in a validation set of locoregional recurrence with a statically significant result (p < 0.001). This finding was also found in distant metastasis-free survival (p < 0.001), progression-free survival (p < 0.001), and regional recurrence-free survival (p < 0.05). Furthermore, in both univariate analysis (HR: 1.58, CI: 1.13-2.19, p < 0.05) and multivariate analysis (HR:1.59, CI: 1.11-2.28, p < 0.05), we also found that our methods demonstrated a strong prognostic value for locoregional recurrence. The proposed novel digital markers could potentially be utilised to assist treatment decisions in cases of NPC.

Machine learning-based overall and cancer-specific survival prediction of M0 penile squamous cell carcinoma：A population-based retrospective study

BackgroundPenile cancer is a rare tumor and few studies have focused on the prognosis of M0 penile squamous cell carcinoma (PSCC). This retrospective study aimed to identify independent prognostic factors and construct predictive models for the overall survival (OS) and cancer-specific survival (CSS) of patients with M0 PSCC. MethodsData was extracted from the Surveillance, Epidemiology, and End Results database for patients diagnosed with malignant penile cancer. Eligible patients with M0 PSCC were selected according to predetermined inclusion and exclusion criteria. These patients were then divided into a training set, a validation set, and a test set. Univariate and multivariate COX regression analyses were initially performed to identify independent prognostic factors for OS and CSS in M0 PSCC patients. Subsequently, traditional and machine learning prognostic models, including random survival forest (RSF), COX, gradient boosting, and component-wise gradient boosting modelling, were constructed using the scikit-survival framework. The performance of each model was assessed by calculating time-dependent area under curve (AUC), C-index, and integrated Brier score (IBS), ultimately identifying the model with the highest performance. Finally, the Shapley additive explanation (SHAP) value, feature importance, and cumulative rates analyses were used to further estimate the selected model. ResultsA total of 2, 446 patients were included in our study. Cox regression analyses demonstrated that age, N stage, and tumor size were predictors of OS, while the N stage, tumor size, surgery, and residential area were predictors of CSS. The RSF and COX models had a higher time-independent AUC and C-index, and lower IBS value than other models in OS and CSS prediction. Feature importance analysis revealed the N stage as a common significant feature for predicting M0 PSCC patients’ survival. The SHAP and cumulative rate analyses demonstrated that the selected models can effectively evaluate the prognosis of M0 PSCC patients. ConclusionIn M0 PSCC patients, age, N stage, and tumor size were predictors of OS. In addition, the N stage, tumor size, surgery, and residential area were predictors of CSS. The machine learning-based RSF and COX models effectively predicted the prognosis of M0 PSCC patients.

Comparison of risk prediction models for the progression of pelvic inflammatory disease patients to sepsis: Cox regression model and machine learning model

IntroductionThe present study presents the development and validation of a clinical prediction model using random survival forest (RSF) and stepwise Cox regression, aiming to predict the probability of pelvic inflammatory disease (PID) progressing to sepsis. MethodsA retrospective cohort study was conducted, gathering clinical data of patients diagnosed with PID between 2008 and 2019 from the Medical Information Mart for Intensive Care (MIMIC)-IV database. Patients who met the Sepsis 3.0 diagnostic criteria were selected, with sepsis as the outcome. Univariate Cox regression and stepwise Cox regression were used to screen variables for constructing a nomogram. Moreover, an RSF model was created using machine learning algorithms. To verify the model's performance, a calibration curve, decision curve analysis (DCA), and receiver operating characteristic (ROC) curve were utilized. Furthermore, the capabilities of the two models for estimating the incidence of sepsis in PID patients within 3 and 7 days were compared. ResultsA total of 1064 PID patients were included, of whom 54 had progressed to sepsis. The established nomogram highlighted dialysis, reduced platelet (PLT) counts, history of pneumonia, medication of glucocorticoids, and increased leukocyte counts as significant predictive factors. The areas under the curve (AUCs) of the nomogram for prediction of PID progression to sepsis at 3-day and 7-day (3-/7-day) in the training set and the validation set were 0.886/0.863 and 0.824/0.726, respectively, and the C-index of the model was 0.8905. The RSF displayed excellent performance, with AUCs of 0.939/0.919 and 0.712/0.571 for 3-/7-day risk prediction in the training set and validation set, respectively. ConclusionThe nomogram accurately predicted the incidence of sepsis in PID patients, and relevant risk factors were identified. While the RSF model outperformed the Cox regression models in predicting sepsis incidence, its performance exhibited some instability. On the other hand, the Cox regression-based nomogram displayed stable performance and improved interpretability, thereby supporting clinical decision-making in PID treatment.

Use of survival support vector machine combined with random survival forest to predict the survival of nasopharyngeal carcinoma patients.

The Cox regression model is not sufficiently accurate to predict the survival prognosis of nasopharyngeal carcinoma (NPC) patients. It is impossible to calculate and rank the importance of impact factors due to the low predictive accuracy of the Cox regression model. So, we developed a system. Using the SEER (The Surveillance, Epidemiology, and End Results) database data on NPC patients, we proposed the use of random survival forest (RSF) and survival-support vector machine (SVM) from the machine learning methods to develop a survival prediction system specifically for NPC patients. This approach aimed to make up for the insufficiency of the Cox regression model. We also used the Cox regression model to validate the development of the nomogram and compared it with machine learning methods. A total of 1,683 NPC patients were extracted from the SEER database from January 2010 to December 2015. We used R language for modeling work, established the nomogram of survival prognosis of NPC patients by Cox regression model, ranked the correlation of influencing factors by RSF model VIMP (variable important) method, developed a survival prognosis system for NPC patients based on survival-SVM, and used C-index for model evaluation and performance comparison. Although the Cox regression models can be developed to predict the prognosis of NPC patients, their accuracy was lower than that of machine learning methods. When we substituted the data for the Cox model, the C-index for the training set was only 0.740, and the C-index for the test set was 0.721. In contrast, the C index of the survival-SVM model was 0.785. The C-index of the RSF model was 0.729. The importance ranking of each variable could be obtained according to the VIMP method. The prediction results from the Cox model are not as good as those of the RSF method and survival-SVM based on the machine learning method. For the survival prognosis of NPC patients, the machine learning method can be considered for clinical application.

Identification of a Novel Oxidative Stress- and Anoikis-Related Prognostic Signature and Its Immune Landscape Analysis in Non-Small Cell Lung Cancer.

The objective of this study was to identify a kind of prognostic signature based on oxidative stress- and anoikis-related genes (OARGs) for predicting the prognosis and immune landscape of NSCLC. Initially, We identified 47 differentially expressed OARGs that primarily regulate oxidative stress and epithelial cell infiltration through the PI3K-Akt pathway. Subsequently, 10 OARGs related to prognosis determined two potential clusters. A cluster was associated with a shorter survival level, lower immune infiltration, higher stemness index and tumor mutation burden. Next, The best risk score model constructed by prognostic OARGs was the Random Survival Forest model, and it included SLC2A1, LDHA and PLAU. The high-risk group was associated with cluster A and poor prognosis, with a higher tumor mutation burden, stemness index and proportion of M0-type macrophages, and a lower immune checkpoint expression level, immune function score and IPS score. The calibration curve and decision-making curve showed that the risk score combined with clinical pathological characteristics could be used to construct a nomogram for guiding the clinical treatment strategies. Finally, We found that all three hub genes were highly expressed in tumor tissues, and LDHA expression was mainly regulated by has-miR-338-3p, has-miR-330-5p and has-miR-34c-5p. Altogether, We constructed an OARG-related prognostic signature to reveal potential relationships between the signature and clinical characteristics, TME, stemness, tumor mutational burden, drug sensitivity and immune landscape in NSCLC patients.

Use of machine learning to predict mortality in patients with type 2 diabetes mellitus, according to socioeconomic status

Abstract Introduction Patients with Type 2 diabetes mellitus (T2DM) are at increased risk of developing cardiovascular disease (CVD) that adversely affects prognosis. Socioeconomic deprivation may be associated with lifestyle choices that contribute to adverse outcomes. The influence of social deprivation on the prognosis of T2DM is rarely investigated in large cohorts. Therefore, we assessed the associations between socioeconomic status and survival in patients with T2DM by applying conventional statistical methods and state-of-the-art, machine learning (ML) models. Methods We obtained routinely collected, linked administrative data for individuals with T2DM aged &amp;gt;50 years from the National Health Service (NHS) Scotland, including demographic data, laboratory measurements, prescriptions, death records and primary and secondary care diagnostic codes. We developed a random survival forest model for predicting all-cause mortality, overall and for each quintile of socioeconomic status using the Scottish Index of Multiple Deprivation (SIMD), which is based on postcodes that reflect a combination of income, employment, local crime rates, education, housing and health. We used Cox proportional hazards models to investigate the risk between the most and least deprived quintiles. Subsequently, we applied Shapely Additive Explanations (SHAP), a state-of-the-art ML interpretability method, to identify key prognostic factors that influence the survival probability prediction score for each sub-group. Results 30,495 people had a newly recorded diagnosis of T2DM between 2009-2019, of whom 4,300 died within 10 years. Table 1 shows the c-index (global assessment of model discrimination) and time-brier score (measure of calibration for time-dependent probability prediction) for the whole cohort and for each SIMD quintile: use of loop diuretics, older age, lower serum concentrations of albumin and alanine transaminase (ALT) and estimated glomerular function rate (eGFR) were strong predictors of death for all quintiles. Prevalence of chronic obstructive pulmonary disease (COPD) was strongly associated with mortality for the most deprived quintile (Q1, whilst strokes were strongly associated with mortality for the most affluent (Q5). A history of heart failure, in addition to use of loop diuretics, predicted a higher risk, in 3 of the 5 quintiles. Overall Q1 had a 36% higher mortality than Q5 (HR: adjusted for age and sex 1.36 [95% CI 1.24 – 1.50 (&amp;lt;0.005)]). Conclusion Greater socioeconomic deprivation is associated with a worse prognosis in patients with T2DM. Results of commonly measured blood tests, allied to readily available patient characteristics (including heart failure and use of loop diuretics) predict mortality risks across all deprivation groups for people with T2DM.Table:Predictors of Mortality in T2DM

Abstract 16870: Short-Term Risk Prognostication for All-Cause Mortality After Endovascular Revascularization in the Medicare Claims-Linked VQI Registry

Background: Patients with chronic limb-threatening ischemia (CLTI) have high mortality rates after revascularization. Balancing short-term mortality risk against limb salvage gains is important when triaging patients with CLTI to revascularization. Data for risk stratification is limited. We aimed to develop machine learning models to rank predictive variables for 30-day and 90-day all-cause mortality for CLTI after peripheral vascular intervention (PVI). Methods: Patients undergoing PVI for CLTI between 2017 and 2018 in the Medicare-linked Vascular Quality Initiative (VQI) were included. Sixty-six preprocedural variables were included in random survival forest (RSF) models. RSF models were constructed for 30-day and 90-day all-cause mortality in the training sample (80% of data) and evaluated in the validation sample (20% of data). Predictive variables were ranked based on frequency that they caused branch splitting nearest the root node. Goodness of fit was assessed by Harrell’s C-index. Results: A total of 10,114 patients (mean age 72 ± 11, 59% male, 74% white) were included. The mortality rate was 4.4% at 30 days and 10.6% at 90 days. The RSF model for 30-day mortality identified stage 5 chronic kidney disease (CKD), dementia, and urgent procedures as the most predictive variables. For the 90-day mortality prediction model, the most important variables were stage 5 CKD, dementia, and poor functional status (Figure) . For both models, 8 of the top 10 variables were either medical history or functional status variables. The C-index was 0.72 for the 30-day model and 0.73 for the 90-day model. Conclusion Using data from the claims-linked VQI registry, medical comorbidities and functional status variables were identified as the most predictive variables for 30-day and 90-day all-cause mortality using RSF models. Results may help guide individualized risk conversations regarding PVI versus conservative therapy for the highest risk patients.

Abstract 14312: Comparison of Machine Learning and Conventional Statistical Modeling for Predicting Readmissions Following Acute Heart Failure Hospitalization

Introduction: Developing accurate models for predicting risk of 30-day readmission has been a major healthcare interest. Evidence suggests that models developed using machine learning (ML) may have better discrimination than conventional statistical models (CSM), but the calibration of such models is unclear. Objectives: To compare models developed using CSM or ML to predict 30-day readmission for cardiovascular and non-cardiovascular causes in HF patients. Methods: We studied 10,919 patients with HF (&gt; 18 years) discharged alive from a hospital or emergency department (2004-2007) in Ontario, Canada, linked to administrative databases for hospitalization and vital status resulting in complete follow-up. The study sample was randomly divided into training and validation sets in a 2:1 ratio. CSMs to predict 30-day readmission were developed using Fine-Gray subdistribution hazards regression (treating death as a competing risk), and the ML algorithm employed random survival forests. Models were evaluated in the validation set using both discrimination and calibration metrics. Results: In the validation sample of 3602 patients (median age 76 [IQR, 67-82] years, 46.6% females), Random Survival Forests (c-statistic = 0.620) showed similar discrimination to the Fine-Gray competing risk model (c-statistic= 0.621) for 30-day cardiovascular readmission. In contrast, for 30-day non-cardiovascular readmission, the Fine-Gray model (c-statistic= 0.641) slightly outperformed the random survival forests model (c-statistic = 0.632). For both outcomes, The Fine-Gray model displayed better calibration than random survival forests when deciles of observed vs. predicted risks were compared (Panels A-D). Conclusions: In HF patients, time-to-event analysis of outcomes using Fine-Gray models had similar discrimination but superior calibration to ML model, highlighting the importance of reporting calibration metrics for ML-based prediction models.

Epigenetic Profiling and Machine Learning for Enhanced Risk Stratification in Pediatric Acute Lymphoblastic Leukemia

Introduction: Acute lymphoblastic leukemia (ALL) is the most common childhood malignancy with high clinical and biological heterogeneity. Despite advancements in treatment, relapsed patients still face significant mortality rates and long-term complications. Risk-adapted strategies have been implemented to optimize treatment intensity based on prognostic factors. Genomic techniques offer potential for improved risk stratification, but accurate grouping remains challenging. Epigenetic alterations, particularly DNA methylation, have shown promise in tumor classification and prognostication. In this study, we trained machine learning survival models based on DNA methylation signatures to refine risk grouping in pediatric ALL patients. Methods: Clinical annotation and DNA methylation data from pediatric ALL samples were retrieved from the Norlund et al. cohort (n = 763). Age, sex, risk group, and cytogenetic subtype were selected as clinical covariates. The cohort was randomly divided into training (80%) and test (20%) sets. Univariate Cox regression and variable importance analysis were performed to select CpG sites associated with relapse-free survival (RFS) and overall survival (OS). Random survival forest models were constructed using the training set, and their performance was evaluated on the test set using the concordance index (c-index), the time-dependent area under the ROC curve (AUC), and the continuous rank probability scores (CRPS). To assess the generalizability of the risk predictors, we performed external validation using two independent pediatric ALL datasets ( Busche et al., n = 42; and Krali et al., n = 384). Results: The relapse risk predictor (RRP) was constructed using random survival forests based on a signature of 16 CpG sites. The model achieved good predictive accuracy with c-indexes of 0.667 and 0.677 in the training and test sets, respectively. The addition of cytogenetic subtype or age at diagnosis did not significantly change the model's performance. Longitudinal assessment of the RRP revealed its superior performance compared to clinical risk grouping. Combining the RRP with clinical risk grouping improved prognostic accuracy, with a 20-month AUC over 80%. The mortality risk predictor (MRP) was constructed using a signature of 53 CpG sites. The model achieved strong predictive performance with c-indexes of 0.751 and 0.755 in the training and test sets, respectively. Similar to the RRP, the addition of cytogenetic subtype or age at diagnosis did not significantly impact the model's performance. Longitudinal assessment of the MRP demonstrated a higher performance rate than clinical risk grouping at all evaluated time points. Combining the MRP with clinical risk grouping yielded the highest prognostic accuracy. In the external validation of the MRP in the Krali dataset, the MRP score was strongly associated with OS: c-index 0.621, p-value 1.06 x 10 -4. The hazard ratio was 1.073 (95% confidence interval: 1.035-1.112) for each incremental increase in the risk score. On the contrary, a lower reproducibility of the RRP was observed (c-index 0.529), presumptively related to the addition of MRD-driven risk stratification in more recent protocols. Conclusions: Machine learning models built on DNA methylation signatures surpassed the traditional clinical risk grouping both for predicting relapse risk and mortality, while the combination of molecular and clinical factors provided the best prognostic accuracy. Further validation and implementation of these predictors could contribute to personalized risk-adapted treatment strategies for pediatric ALL patients.

Predicting survival of advanced laryngeal squamous cell carcinoma: comparison of machine learning models and Cox regression models

Laryngeal squamous cell carcinoma (LSCC) is a common tumor type. High recurrence rates remain an important factor affecting the survival and quality of life of advanced LSCC patients. We aimed to build a new nomogram and a random survival forest model using machine learning to predict the risk of LSCC progress. The study included 671 patients with AJCC stages III–IV LSCC. To develop a prognostic model, Cox regression analyses were used to assess the relationship between clinic-pathologic factors and disease-free survival (DFS). RSF analysis was also used to predict the DFS of LSCC patients. The ROC curve revealed that the Cox model exhibited good sensitivity and specificity in predicting DFS in the training and validation cohorts (1 year, validation AUC = 0.679, training AUC = 0.693; 3 years, validation AUC = 0.716, training AUC = 0.655; 5 years, validation AUC = 0.717, training AUC = 0.659). Random survival forest analysis showed that N stage, clinical stage, and postoperative chemoradiotherapy were prognostically significant variables associated with survival. The random forest model exhibited better prediction ability than the Cox regression model in the training cohort; however, the two models showed similar prediction ability in the validation cohort.

Predicting Colorectal Cancer Survival Using Time-to-Event Machine Learning: Retrospective Cohort Study.

Machine learning (ML) methods have shown great potential in predicting colorectal cancer (CRC) survival. However, the ML models introduced thus far have mainly focused on binary outcomes and have not considered the time-to-event nature of this type of modeling. This study aims to evaluate the performance of ML approaches for modeling time-to-event survival data and develop transparent models for predicting CRC-specific survival. The data set used in this retrospective cohort study contains information on patients who were newly diagnosed with CRC between December 28, 2012, and December 27, 2019, at West China Hospital, Sichuan University. We assessed the performance of 6 representative ML models, including random survival forest (RSF), gradient boosting machine (GBM), DeepSurv, DeepHit, neural net-extended time-dependent Cox (or Cox-Time), and neural multitask logistic regression (N-MTLR) in predicting CRC-specific survival. Multiple imputation by chained equations method was applied to handle missing values in variables. Multivariable analysis and clinical experience were used to select significant features associated with CRC survival. Model performance was evaluated in stratified 5-fold cross-validation repeated 5 times by using the time-dependent concordance index, integrated Brier score, calibration curves, and decision curves. The SHapley Additive exPlanations method was applied to calculate feature importance. A total of 2157 patients with CRC were included in this study. Among the 6 time-to-event ML models, the DeepHit model exhibited the best discriminative ability (time-dependent concordance index 0.789, 95% CI 0.779-0.799) and the RSF model produced better-calibrated survival estimates (integrated Brier score 0.096, 95% CI 0.094-0.099), but these are not statistically significant. Additionally, the RSF, GBM, DeepSurv, Cox-Time, and N-MTLR models have comparable predictive accuracy to the Cox Proportional Hazards model in terms of discrimination and calibration. The calibration curves showed that all the ML models exhibited good 5-year survival calibration. The decision curves for CRC-specific survival at 5 years showed that all the ML models, especially RSF, had higher net benefits than default strategies of treating all or no patients at a range of clinically reasonable risk thresholds. The SHapley Additive exPlanations method revealed that R0 resection, tumor-node-metastasis staging, and the number of positive lymph nodes were important factors for 5-year CRC-specific survival. This study showed the potential of applying time-to-event ML predictive algorithms to help predict CRC-specific survival. The RSF, GBM, Cox-Time, and N-MTLR algorithms could provide nonparametric alternatives to the Cox Proportional Hazards model in estimating the survival probability of patients with CRC. The transparent time-to-event ML models help clinicians to more accurately predict the survival rate for these patients and improve patient outcomes by enabling personalized treatment plans that are informed by explainable ML models.

Deep Learning–Based Survival Analysis for Receiving a Steatotic Donor Liver Versus Waiting for a Standard Liver

An emerging strategy to expand the donor pool is the use of a steatotic donor liver (SDLs; ≥ 30% macrosteatosis on biopsy). With the obesity epidemic and prevalence of nonalcoholic fatty liver disease, SDLs have been reported in 59% of all deceased donors. Many potential candidates need to decide whether to accept an SDL offer or remain on the waitlist for a nonsteatotic donor liver (non-SDL). The objective of this study was to compare the survival of accepting an SDL vs using a non-SDL after waiting various times. Using data from the United States' organ procurement and transplantation network, deep survival learning predictive models were built to compare post-decision survival after accepting an SDL vs waiting for a non-SDL. The comparison subjects contain simulated 20,000 different scenarios of a candidate either accepting an SDL immediately or receiving a non-SDL after waiting various times. The research variables were selected using the LASSO-Cox and Random Survival Forest (RSF) models. The Cox proportional hazards and RSF models were also comparatively included for survival prediction. In addition, personalized survival curves for randomly selected candidates were generated. Deep survival learning outperformed Cox proportional hazards and RSF in predicting the survival of liver transplants. Among the simulations, 25% to 30% of scenarios demonstrated a higher 3-year survival post-decision for candidates accepting an SDL than waiting and receiving a non-SDL. The difference was only 1.43% in 3-year survival post-decision between accepting an SDL and waiting 260 days (mean waitlist time) for a non-SDL. As the number of days on the waitlist increases, the difference in survival between accepting SDLs and waiting for non-SDLs decreases. Appropriately used SDLs could expand the donor pool and relieve the candidates' unmet need for donor livers, which presents long-term survival benefits for recipients.

Machine learning model for predicting physical activity related bleeding risk in Chinese boys with haemophilia A

BackgroundPhysical activity is a crucial part of an active lifestyle for haemophiliac children. However, the fear of bleeds has been identified as barriers to participating physical activity for haemophiliac children even with prophylaxis. Lack of evidence and metrics driven by data is key problem. ObjectivesWe aim to develop machine learning models based on clinical data with multiple potential factors considered to predict risk of physical activity bleeding for haemophilia children with prophylaxis. MethodsFrom this cohort study, we collected information on 98 haemophiliac children with adequate prophylaxis (trough FVIII:C level > 1 %). The involved potential predictor variables include demographic information, treatment information, physical activity, joint evaluation, and pharmacokinetic parameters, etc. We applied CoxPH, Random Survival Forests (RSF) and DeepSurv to construct prediction models for the risk of bleeding during physical activities. All three survival analysis models were internally and externally validated. ResultsA total of 98 patients were enrolled in this study. Their median age was 7.9 (5.5, 10.2) years. The CoxPH, RSF and DeepSurv models' discriminative and calibration abilities were all high, and the RSF model had the best performance (Internal validation: C-index, 0.7648 ± 0.0139; Brier Score, 0.1098 ± 0.0015; External validation: C-index, 0.7260 ± 0.0154; Brier Score, 0.0930 ± 0.0018). The prediction curves demonstrated that the developed RSF model can distinguish the risks well between bleeding and non-bleeding patients, as well as patients with different levels of physical activity. Meanwhile, the feature importance analysis confirmed that physical activity bleeding was deduced by comprehensive effects of various factors, and the importance of different factors on bleeding outcome is discrepant. ConclusionsThis study revealed from the mechanism that it is necessary to incorporate multiple factors to accurately predict physical activity related bleeding risk. In clinical practice, the designed machine learning models can provide guidance for children with haemophilia A to positively participate in physical activity.

Prediction of survival in patients with infected pancreatic necrosis: a prospective cohort study.

Infected pancreatic necrosis (IPN) is a severe complication of acute pancreatitis, with mortality rates ranging from 15 to 35%. However, limited studies exist to predict the survival of IPN patients and nomogram has never been built. This study aimed to identify predictors of mortality, estimate conditional survival (CS), and develop a CS nomogram and logistic regression nomogram for real-time prediction of survival in IPN patients. A prospective cohort study was performed in 335 IPN patients consecutively enrolled at a large Chinese tertiary hospital from January 2011 to December 2022. The random survival forest method was first employed to identify the most significant predictors and capture clinically relevant nonlinear threshold effects. Instantaneous death risk and CS was first utilized to reveal the dynamic changes in the survival of IPN patients. A Cox model-based nomogram incorporating CS and a logistic regression-based nomogram were first developed and internally validated with a bootstrap method. The random survival forest model identified seven foremost predictors of mortality, including the number of organ failures, duration of organ failure, age, time from onset to first intervention, hemorrhage, bloodstream infection, and severity classification. Duration of organ failure and time from onset to first intervention showed distinct thresholds and nonlinear relationships with mortality. Instantaneous death risk reduced progressively within the first 30 days, and CS analysis indicated gradual improvement in real-time survival since diagnosis, with 90-day survival rates gradually increasing from 0.778 to 0.838, 0.881, 0.974, and 0.992 after surviving 15, 30, 45, 60, and 75 days, respectively. After further variables selection using step regression, five predictors (age, number of organ failures, hemorrhage, time from onset to first intervention, and bloodstream infection) were utilized to construct both the CS nomogram and logistic regression nomogram, both of which demonstrated excellent performance with 1000 bootstrap. Number of organ failures, duration of organ failure, age, time from onset to first intervention, hemorrhage, bloodstream infection, and severity classification were the most crucial predictors of mortality of IPN patients. The CS nomogram and logistic regression nomogram constructed by these predictors could help clinicians to predict real-time survival and optimize clinical decisions.

Personalized prognostic prediction tool for high-grade neuroendocrine cervical cancer: a SEER database analysis and single-center validation

PurposeCervical high-grade neuroendocrine carcinoma (CHGNEC) is a rare but highly aggressive cancer. The purpose of this study is to develop a prognostic nomogram that can accurately predict the outcomes for CHGNEC patients.MethodsWe analyzed clinical data from the Surveillance, Epidemiology, and End Results (SEER) database of CHGNEC patients, including small-cell neuroendocrine carcinoma (SCNEC) and large-cell neuroendocrine carcinoma (LCNEC). We investigated patient characteristics and prognosis, and developed a prognostic nomogram model for cancer-specific survival in CHGNEC patients. External validation was conducted using real clinical cases from our hospital.ResultsOur study included 306 patients from SEER database, with a mean age of 49.9 ± 15.5 years. Most of the patients had SCNEC (86.9%). Among them, 170 died from the disease, while 136 either survived or died from other causes. Our final predictive model identified age at diagnosis, stage 1 status, stage 4 status, T1, N0, and surgery of the primary site as independent prognostic factors for CHGNEC. We validated our model using a group of 16 CHGNEC patients who underwent surgery at our center. The external validation showed that the prognostic nomogram had excellent discriminative ability, with an area under the receiver operating characteristic curve (AUC) of 0.76 (95% CI 0.49–1.00) for the prediction of 3-year cancer-specific survival (CSS) and an AUC of 0.85 (95% CI 0.62–1.00) for the prediction of 5-years CSS. The random survival forest model achieved an AUC of 0.80 (95% CI 0.56–1.00) for 3-years CSS and 0.91 (95% CI 0.72–1.00) for 5-years CSS, indicating its adequacy in predicting outcomes for CHGNEC patients.ConclusionOur study provides an excellent nomogram for predicting the prognosis of CHGNEC patients. The prognostic nomogram can be a useful tool for clinicians in identifying high-risk patients and making personalized treatment decisions.

A machine learning-based integrated clinical model for predicting prognosis in atypical meningioma patients.

Atypical meningioma (AM) recurs in up to half of patients after surgical resection and may require adjuvant therapy to improve patient prognosis. Various clinicopathological features have been shown to have prognostic implications in AM, but an integrated prediction model is lacking. Thus, in this study, we aimed to develop and validate an integrated prognostic model for AM. A retrospective cohort of 528 adult AM patients surgically treated at our institution were randomly assigned to a training or validation group in a 7:3 ratio. Sixteen baseline demographic, clinical, and pathological parameters, progression-free survival (PFS), and overall survival (OS) were analysed. Sixty-five combinations of machine learning (ML) algorithms were used for model training and validation to predict tumour recurrence and patient mortality. The random survival forest (RSF) model was the best model for predicting recurrence and death. Primary or secondary tumour, Ki-67 index, extent of resection, tumour size, brain involvement, tumour necrosis, and age contributed significantly to the model. The C-index value of the RSF recurrence prediction model reached 0.8080. The AUCs for 1-, 3-, and 5-year PFS were 0.83, 0.82, and 0.86, respectively. The C-index value of the RSF death prediction model reached 0.8890. The AUCs for 3-year and 5-year OS were 0.88 and 0.89, respectively. A high-performing integrated RSF predictive model for AM recurrence and patient mortality was proposed that may guide therapeutic decision-making and long-term monitoring.

Multicomponent prediction of 2-year mortality and amputation in patients with diabetic foot using a random survival forest model: Uric acid, alanine transaminase, urine protein and platelet as important predictors.

The current methods for the predictionof mortality and amputation for inpatients with diabetic foot (DF) use onlyconventional, simple variables, which limitstheir performance. Here, we used a random survival forest (RSF) model and multicomponent variables to improve the prediction of mortality and amputation for these patients. We performed a retrospective cohort study of 175 inpatients with DF who were recruited between 2014 and 2021. Thirty-one predictors in six categories were considered as potential covariates. Seventy percent(n = 122) of the participants were randomly selected to constitute a training set, and 30% (n = 53) were assigned to a testing set. The RSF model was used to screen appropriate variables for their value as predictors of 2-year all-cause mortality and amputation, and a multicomponent prediction model was established. Model performance was evaluated using the area under the curve (AUC) and the Hosmer-Lemeshow test. The AUCs were compared using the Delong test. Seventeen variables were selected to predict mortality and 23 were selected to predict amputation. Uric acid and alanine transaminase were thetop two most usefulvariables for the prediction of mortality, whereas urine protein and platelet were the top variablesfor the prediction of amputation. The AUCs were 0.913 and 0.851for the prediction of mortality for the training and testing sets, respectively; and the equivalent AUCs were 0.963 and 0.893 for the prediction of amputation. There were no significant differences between the AUCs for the training and testing sets for both the mortality and amputation models. These models showed a good degree of fit. Thus, the RSF model can predict mortality and amputation in inpatients with DF. This multicomponent prediction model could help clinicians consider predictors of different dimensions to effectively prevent DF from clinical outcomes .

Screening strategies and dynamic risk prediction models for Alzheimer's disease

BackgroundCharacterizing the progression from Mild cognitive impairment (MCI) to Alzheimer's disease (AD) is essential for early AD prevention and targeted intervention. Our goal was to construct precise screening schemes for individuals with different risk of AD and to establish prognosis models for them. MethodsWe constructed a retrospective cohort by reviewing individuals with baseline diagnosis of MCI and at least one follow-up visits between November 2005 and May 2021. They were stratified into high-risk and low-risk groups with longitudinal cognitive trajectory. Then, we established a screening framework and obtained optimal screening strategies for two risk groups. Cox and random survival forest (RSF) models were developed for dynamic prognosis prediction. ResultsIn terms of screening strategies, the combination of Clinical Dementia Rating Sum of Boxes (CDRSB) and hippocampus volume was recommended for the high-risk MCI group, while the combination of Alzheimer's Disease Assessment Scale Cognitive 13 items (ADAS13) and FAQ was recommended for low-risk MCI group. The concordance index (C-index) of the Cox model for the high-risk group was 0.844 (95% CI: 0.815–0.873) and adjustments for demographic information and APOE ε4. The RSF model incorporating longitudinal ADAS13, FAQ, and demographic information and APOE ε4 performed for the low-risk group. ConclusionThis precise screening scheme will optimize allocation of medical resources and reduce the economic burden on individuals with low risk of MCI. Moreover, dynamic prognosis models may be helpful for early identification of individuals at risk and clinical decisions, which will promote the secondary prevention of AD.

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.

Prediction of lung papillary adenocarcinoma-specific survival using ensemble machine learning models

Accurate prognostic prediction is crucial for treatment decision-making in lung papillary adenocarcinoma (LPADC). The aim of this study was to predict cancer-specific survival in LPADC using ensemble machine learning and classical Cox regression models. Moreover, models were evaluated to provide recommendations based on quantitative data for personalized treatment of LPADC. Data of patients diagnosed with LPADC (2004–2018) were extracted from the Surveillance, Epidemiology, and End Results database. The set of samples was randomly divided into the training and validation sets at a ratio of 7:3. Three ensemble models were selected, namely gradient boosting survival (GBS), random survival forest (RSF), and extra survival trees (EST). In addition, Cox proportional hazards (CoxPH) regression was used to construct the prognostic models. The Harrell’s concordance index (C-index), integrated Brier score (IBS), and area under the time-dependent receiver operating characteristic curve (time-dependent AUC) were used to evaluate the performance of the predictive models. A user-friendly web access panel was provided to easily evaluate the model for the prediction of survival and treatment recommendations. A total of 3615 patients were randomly divided into the training and validation cohorts (n = 2530 and 1085, respectively). The extra survival trees, RSF, GBS, and CoxPH models showed good discriminative ability and calibration in both the training and validation cohorts (mean of time-dependent AUC: > 0.84 and > 0.82; C-index: > 0.79 and > 0.77; IBS: < 0.16 and < 0.17, respectively). The RSF and GBS models were more consistent than the CoxPH model in predicting long-term survival. We implemented the developed models as web applications for deployment into clinical practice (accessible through https://shinyshine-820-lpaprediction-model-z3ubbu.streamlit.app/). All four prognostic models showed good discriminative ability and calibration. The RSF and GBS models exhibited the highest effectiveness among all models in predicting the long-term cancer-specific survival of patients with LPADC. This approach may facilitate the development of personalized treatment plans and prediction of prognosis for LPADC.