Machine Learning Model for Predicting Postoperative Survival of Patients with Colorectal Cancer

Association of sarcopenia with all-cause and cause-specific mortality in cancer patients: development and validation of a 3-year and 5-year survival prediction model

BACKGROUND

Sarcopenia is a clinicopathological condition characterized by a decrease in muscle strength and muscle mass, playing a crucial role in the prognosis of cancer. Therefore, this study aims to investigate the association between sarcopenia and both all-cause mortality and cancer-specific mortality among cancer patients. Furthermore, we plan to develop risk prediction models using machine learning algorithms to predict 3-year and 5-year survival rates in cancer patients.

METHOD

This study included 1095 cancer patients from the National Health and Nutrition Examination Survey (NHANES) cohorts spanning 1999-2006 and 2011-2014. Initially, we used the Least Absolute Shrinkage and Selection Operator (LASSO)-Cox regression models for feature selection. Subsequently, we employed multivariable Cox regression models to investigate the association between sarcopenia and all-cause and cancer-specific mortality in cancer patients. We developed five machine learning algorithms, including Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), LightGBM, and XGBoost, to predict 3-year and 5-year survival rates and to perform risk stratification.

RESULTS

The multivariable COX regression model showed sarcopenia significantly increases the risk of all-cause mortality (HR = 1.33, 95%CI:1.05, 1.70, P = 0.0194) and cancer-specific mortality (HR = 1.67, 95%CI:1.09, 2.55, P = 0.0176) in cancer patients. Among the five machine learning algorithms developed, the LightGBM model demonstrated strong performance in the 3-year and 5-year survival prediction tasks, making it the optimal model selection. Decision curve analysis and Kaplan-Meier curves further confirmed our model's ability to identify high-risk individuals effectively.

CONCLUSIONS

Sarcopenia significantly increases the risk of mortality in cancer patients. We developed a survival prediction model for cancer patients that effectively identifies high-risk individuals, thereby providing a foundation for personalized survival assessment.

Personalized surveillance in colorectal cancer: Integrating circulating tumor DNA and artificial intelligence into post-treatment follow-up

Given the growing burden of colorectal cancer (CRC) as a global health challenge, it becomes imperative to focus on strategies that can mitigate its impact. Post-treatment surveillance has emerged as essential for early detection of recurrence, significantly improving patient outcomes. However, intensive surveillance strategies have shown mixed results compared to less intensive methods, emphasizing the necessity for personalized, risk-adapted approaches. The observed suboptimal adherence to existing surveillance protocols underscores the urgent need for more tailored and efficient strategies. In this context, circulating tumor DNA (ctDNA) emerges as a promising biomarker with significant potential to revolutionize post-treatment surveillance, demonstrating high specificity [0.95, 95% confidence interval (CI): 0.91-0.97] and robust diagnostic odds (37.6, 95%CI: 20.8-68.0) for recurrence detection. Furthermore, artificial intelligence and machine learning models integrating patient-specific and tumor features can enhance risk stratification and optimize surveillance strategies. The reported area under the receiver operating characteristic curve, measuring artificial intelligence model performance in predicting CRC recurrence, ranged from 0.581 and 0.593 at the lowest to 0.979 and 0.978 at the highest in training and validation cohorts, respectively. Despite this promise, addressing cost, accessibility, and extensive validation remains crucial for equitable integration into clinical practice.

Externally validated and clinically useful machine learning algorithms to support patient-related decision-making in oncology: a scoping review

BACKGROUND

This scoping review systematically maps externally validated machine learning (ML)-based models in cancer patient care, quantifying their performance, and clinical utility, and examining relationships between models, cancer types, and clinical decisions. By synthesizing evidence, this study identifies, strengths, limitations, and areas requiring further research.

METHODS

The review followed the Joanna Briggs Institute's methodology, Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines, and the Population, Concept, and Context mnemonic. Searches were conducted across Embase, IEEE Xplore, PubMed, Scopus, and Web of Science (January 2014-September 2022), targeting English-language quantitative studies in Q1 journals (SciMago Journal and Country Ranking > 1) that used ML to evaluate clinical outcomes for human cancer patients with commonly available data. Eligible models required external validation, clinical utility assessment, and performance metric reporting. Studies involving genetics, synthetic patients, plants, or animals were excluded. Results were presented in tabular, graphical, and descriptive form.

RESULTS

From 4023 deduplicated abstracts and 636 full-text reviews, 56 studies (2018-2022) met the inclusion criteria, covering diverse cancer types and applications. Convolutional neural networks were most prevalent, demonstrating high performance, followed by gradient- and decision tree-based algorithms. Other algorithms, though underrepresented, showed promise. Lung and digestive system cancers were most frequently studied, focusing on diagnosis and outcome predictions. Most studies were retrospective and multi-institutional, primarily using image-based data, followed by text-based and hybrid approaches. Clinical utility assessments involved 499 clinicians and 12 tools, indicating improved clinician performance with AI assistance and superior performance to standard clinical systems.

DISCUSSION

Interest in ML-based clinical decision-making has grown in recent years alongside increased multi-institutional collaboration. However, small sample sizes likely impacted data quality and generalizability. Persistent challenges include limited international validation across ethnicities, inconsistent data sharing, disparities in validation metrics, and insufficient calibration reporting, hindering model comparison reliability.

CONCLUSION

Successful integration of ML in oncology decision-making requires standardized data and methodologies, larger sample sizes, greater transparency, and robust validation and clinical utility assessments.

OTHER

Financed by FCT-Fundação para a Ciência e a Tecnologia (Portugal, project LA/P/0063/2020, grant 2021.09040.BD) as part of CSS's Ph.D. This work was not registered.

Development and validation of a biomarker-based prediction model for metastasis in patients with colorectal cancer: Application of machine learning algorithms

Objective

The purpose of the current study was to develop and validate a biomarker-based prediction model for metastasis in patients with colorectal cancer (CRC).

Methods

Two datasets, GSE68468 and GSE41568, were retrieved from the Gene Expression Omnibus (GEO) database. In the GSE68468 dataset, key biomarkers were identified through a screening process involving differential expression analysis, redundancy analysis, and recursive feature elimination technique. Subsequently, the prediction model was developed and internally validated using five machine learning (ML) algorithms including lasso and elastic-net regularized generalized linear model (glmnet), k-nearest neighbors (kNN), support vector machine (SVM) with Radial Basis Function Kernel, random forest (RF), and eXtreme Gradient Boosting (XGBoost). The predictive performance of the algorithm with the highest accuracy was then externally validated on the GSE41568 dataset.

Results

Among 22,283 registered genes in the GSE68468 dataset, the screening process identified 16 key genes including MMP3, CCDC102B, CDH2, SCGB1A1, KRT7, CYP1B1, LAMC3, ALB, DIXDC1, VWF, MMP1, CYP4B1, NKX3-2, TMEM158, GADD45B, SERPINA1 and these genes were used to build the prediction model. On the internal validation dataset, the prediction performance of five ML algorithms was as follows; RF (accuracy = 0.97 and kappa = 0.91), XGBoost (0.93, 0.81), kNN (0.93, 0.81), glmnet (0.93, 0.82) and SVM (0.92, 0.80). Top five biomarkers were MMP3, CCDC102B, CDH2, VWF and MMP1. The RF model exhibited an accuracy of 0.97, a kappa value of 0.92, and an area under the curve (AUC) of 0.99 in the external validation dataset.

Conclusion

The results of this study have identified biomarkers through ML algorithms which help to identify patients with CRC prone to metastasis.

Predicting Factors Affecting Survival Rate in Patients Undergoing On-Pump Coronary Artery Bypass Graft Surgery Using Machine Learning Methods: A Systematic Review

Background and Aim

Coronary artery bypass grafting (CABG) is a key treatment for coronary artery disease, but accurately predicting patient survival after the procedure presents significant challenges. This study aimed to systematically review articles using machine learning techniques to predict patient survival rates and identify factors affecting these rates after CABG surgery.

Methods

From January 1, 2015, to January 20, 2024, a comprehensive literature search was conducted across PubMed, Scopus, IEEE Xplore, and Web of Science. The review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Inclusion criteria included studies that evaluated survival rates and predictors associated with CABG patients during the specified period.

Results

After eliminating duplicates, a total of 1330 articles were identified. Following a systematic screening, 24 studies met the inclusion criteria. Our findings revealed 43 distinct factors influencing survival rates in patients undergoing CABG. Notably, five factors-age, ejection fraction, diabetes mellitus, a history of cerebrovascular disease or accidents, and renal function-were consistently identified across multiple studies as significant predictors of postsurgical survival.

Conclusion

This systematic review identifies key factors influencing survival rates after CABG surgery and highlights the role of machine learning in improving predictive accuracy. By identifying high-risk patients through these key factors, our findings offer practical insights for healthcare providers, enhancing patient management and customizing therapeutic strategies after CABG. This study significantly enhances existing literature by combining machine learning techniques with clinical factors, thereby improving the understanding of patient outcomes in CABG surgery.

Light gradient boost tree classifier predictions on appendicitis with periodontal disease from biochemical and clinical parameters

INTRODUCTION

Untreated periodontitis significantly increases the risk of tooth loss, often delaying treatment due to asymptomatic phases. Recent studies have increasingly associated poor dental health with conditions such as rheumatoid arthritis, diabetes, obesity, pneumonia, cardiovascular disease, and renal illness. Despite these connections, limited research has investigated the relationship between appendicitis and periodontal disease. This study aims to predict appendicitis in patients with periodontal disease using biochemical and clinical parameters through the application of a light gradient boost tree classifier.

METHODS

Data from 125 patient records at Saveetha Institute of Dental College and Medical College were pre-processed and analyzed. We utilized data preprocessing techniques, feature selection methods, and model development approaches to estimate the risk of appendicitis in patients with periodontitis. Both Random Forest and Light Gradient Boosting algorithms were evaluated for accuracy using confusion matrices to assess their predictive performance.

RESULTS

The Random Forest model achieved an accuracy of 94%, demonstrating robust predictive capability in this context. In contrast, the Light Gradient Boost algorithms achieved a significantly higher accuracy of 98%, underscoring their superior predictive efficiency. This substantial difference highlights the importance of algorithm selection and optimization in developing reliable predictive models. The higher accuracy of Light Gradient Boost algorithms suggests effective minimization of prediction errors and improved differentiation between appendicitis with periodontitis and healthy states. Our study identifies age, white blood cell count, and symptom duration as pivotal predictors for detecting concurrent periodontitis in acute appendicitis cases.

CONCLUSIONS

The newly developed prediction model introduces a novel and promising approach, providing valuable insights into distinguishing between periodontitis and acute appendicitis. These findings highlight the potential to improve diagnostic accuracy and support informed clinical decision-making in patients presenting with both conditions, offering new avenues for optimizing patient care strategies.

Precision prognosis of colorectal cancer: a multi-tiered model integrating microsatellite instability genes and clinical parameters

Background

Prognostic assessment for colorectal cancer (CRC) displays substantial heterogeneity, as reliance solely on traditional TNM staging falls short of achieving precise individualized predictions. The integration of diverse biological information sources holds the potential to enhance prognostic accuracy.

Objective

To establish a comprehensive multi-tiered precision prognostic evaluation system for CRC by amalgamating gene expression profiles, clinical characteristics, and tumor microsatellite instability (MSI) status in CRC patients.

Methods

We integrated genomic data, clinical information, and survival follow-up data from 483 CRC patients obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. MSI-related gene modules were identified using differential expression analysis and Weighted Gene Co-expression Network Analysis (WGCNA). Three prognostic models were constructed: MSI-Related Gene Prognostic Model (Model I), Clinical Prognostic Model (Model II), and Integrated Multi-Layered Prognostic Model (Model III) by combining clinical features. Model performance was assessed and compared using Receiver Operating Characteristic (ROC) curves, Kaplan-Meier analysis, and other methods.

Results

Six MSI-related genes were selected for constructing Model I (AUC = 0.724); Model II used two clinical features (AUC = 0.684). Compared to individual models, the integrated Model III exhibited superior performance (AUC = 0.825) and demonstrated good stability in an independent dataset (AUC = 0.767).

Conclusion

This study successfully developed and validated a comprehensive multi-tiered precision prognostic assessment model for CRC, providing an effective tool for personalized medical management of CRC.

SAGL: A self-attention-based graph learning framework for predicting survival of colorectal cancer patients

BACKGROUND AND OBJECTIVE

Colorectal cancer (CRC) is one of the most commonly diagnosed cancers worldwide. The accurate survival prediction for CRC patients plays a significant role in the formulation of treatment strategies. Recently, machine learning and deep learning approaches have been increasingly applied in cancer survival prediction. However, most existing methods inadequately represent and leverage the dependencies among features and fail to sufficiently mine and utilize the comorbidity patterns of CRC. To address these issues, we propose a self-attention-based graph learning (SAGL) framework to improve the postoperative cancer-specific survival prediction for CRC patients.

METHODS

We present a novel method for constructing dependency graph (DG) to reflect two types of dependencies including comorbidity-comorbidity dependencies and the dependencies between features related to patient characteristics and cancer treatments. This graph is subsequently refined by a disease comorbidity network, which offers a holistic view of comorbidity patterns of CRC. A DG-guided self-attention mechanism is proposed to unearth novel dependencies beyond what DG offers, thus augmenting CRC survival prediction. Finally, each patient will be represented, and these representations will be used for survival prediction.

RESULTS

The experimental results show that SAGL outperforms state-of-the-art methods on a real-world dataset, with the receiver operating characteristic curve for 3- and 5-year survival prediction achieving 0.849±0.002 and 0.895±0.005, respectively. In addition, the comparison results with different graph neural network-based variants demonstrate the advantages of our DG-guided self-attention graph learning framework.

CONCLUSIONS

Our study reveals that the potential of the DG-guided self-attention in optimizing feature graph learning which can improve the performance of CRC survival prediction.

Machine learning for predicting colon cancer recurrence

INTRODUCTION

Colorectal cancer (CRC) is a global public health concern, ranking among the most commonly diagnosed malignancies worldwide. Despite advancements in treatment modalities, the specter of CRC recurrence remains a significant challenge, demanding innovative solutions for early detection and intervention. The integration of machine learning into oncology offers a promising avenue to address this issue, providing data-driven insights and personalized care.

METHODS

This retrospective study analyzed data from 396 patients who underwent surgical procedures for colon cancer (CC) between 2010 and 2021. Machine learning algorithms were employed to predict CC recurrence, with a focus on demographic, clinicopathological, and laboratory characteristics. A range of evaluation metrics, including AUC (Area Under the Receiver Operating Characteristic), accuracy, recall, precision, and F1 scores, assessed the performance of machine learning algorithms.

RESULTS

Significant risk factors for CC recurrence were identified, including sex, carcinoembryonic antigen (CEA) levels, tumor location, depth, lymphatic and venous invasion, and lymph node involvement. The CatBoost Classifier demonstrated exceptional performance, achieving an AUC of 0.92 and an accuracy of 88 % on the test dataset. Feature importance analysis highlighted the significance of CEA levels, albumin levels, N stage, weight, platelet count, height, neutrophil count, lymphocyte count, and gender in determining recurrence risk.

DISCUSSION

The integration of machine learning into healthcare, exemplified by this study's findings, offers a pathway to personalized patient risk stratification and enhanced clinical decision-making. Early identification of individuals at risk of CC recurrence holds the potential for more effective therapeutic interventions and improved patient outcomes.

CONCLUSION

Machine learning has the potential to revolutionize our approach to CC recurrence prediction, emphasizing the synergy between medical expertise and cutting-edge technology in the fight against cancer. This study represents a vital step toward precision medicine in CC management, showcasing the transformative power of data-driven insights in oncology.

The development and validation of automated machine learning models for predicting lymph node metastasis in Siewert type II T1 adenocarcinoma of the esophagogastric junction

Background

Lymph node metastasis (LNM) is considered an essential prognosis factor for adenocarcinoma of the esophagogastric junction (AEG), which also affects the treatment strategies of AEG. We aimed to evaluate automated machine learning (AutoML) algorithms for predicting LNM in Siewert type II T1 AEG.

Methods

A total of 878 patients with Siewert type II T1 AEG were selected from the Surveillance, Epidemiology, and End Results (SEER) database to develop the LNM predictive models. The patients from two hospitals in Suzhou were collected as the test set. We applied five machine learning algorithms to develop the LNM prediction models. The performance of predictive models was assessed using various metrics including accuracy, sensitivity, specificity, the area under the curve (AUC), and receiver operating characteristic (ROC) curve.

Results

Patients with LNM exhibited a higher proportion of male individuals, a poor degree of differentiation, and submucosal infiltration, with statistical differences. The deep learning (DL) model demonstrated relatively good accuracy (0.713) and sensitivity (0.868) among the five models. Moreover, the DL model achieved the highest AUC (0.781) and sensitivity (1.000) in the test set.

Conclusion

The DL model showed good predictive performance among five AutoML models, indicating the advantage of AutoML in modeling LNM prediction in patients with Siewert type II T1 AEG.

New perspectives on cancer clinical research in the era of big data and machine learning

In the 21st century, the development of medical science has entered the era of big data, and machine learning has become an essential tool for mining medical big data. The establishment of the SEER database has provided a wealth of epidemiological data for cancer clinical research, and the number of studies based on SEER and machine learning has been growing in recent years. This article reviews recent research based on SEER and machine learning and finds that the current focus of such studies is primarily on the development and validation of models using machine learning algorithms, with the main directions being lymph node metastasis prediction, distant metastasis prediction, and prognosis-related research. Compared to traditional models, machine learning algorithms have the advantage of stronger adaptability, but also suffer from disadvantages such as overfitting and poor interpretability, which need to be weighed in practical applications. At present, machine learning algorithms, as the foundation of artificial intelligence, have just begun to emerge in the field of cancer clinical research. The future development of oncology will enter a more precise era of cancer research, characterized by larger data, higher dimensions, and more frequent information exchange. Machine learning is bound to shine brightly in this field.

Classification and Diagnostic Prediction of Colorectal Cancer Mortality Based on Machine Learning Algorithms: A Multicenter National Study

INTRODUCTION

Colorectal cancer (CRC) ranks as the second leading cause of cancer-related deaths. This study aimed to predict survival outcomes of CRC patients using machine learning (ML) methods.

MATERIAL AND METHODS

A retrospective analysis included 1853 CRC patients admitted to three prominent tertiary hospitals in Iran from October 2006 to July 2019. Six ML methods, namely logistic regression (LR), Naïve Bayes (NB), Support Vector Machine (SVM), Neural Network (NN), Decision Tree (DT), and Light Gradient Boosting Machine (LGBM), were developed with 10-fold cross-validation. Feature selection employed the Random Forest method based on mean decrease GINI criteria. Model performance was assessed using Area Under the Curve (AUC).

RESULTS

Time from diagnosis, age, tumor size, metastatic status, lymph node involvement, and treatment type emerged as crucial predictors of survival based on mean decrease GINI. The NB (AUC = 0.70, 95% Confidence Interval [CI] 0.65-0.75) and LGBM (AUC = 0.70, 95% CI 0.65-0.75) models achieved the highest predictive AUC values for CRC patient survival.

CONCLUSIONS

This study highlights the significance of variables including time from diagnosis, age, tumor size, metastatic status, lymph node involvement, and treatment type in predicting CRC survival. The NB model exhibited optimal efficacy in mortality prediction, maintaining a balanced sensitivity and specificity. Policy recommendations encompass early diagnosis and treatment initiation for CRC patients, improved data collection through digital health records and standardized protocols, support for predictive analytics integration in clinical decisions, and the inclusion of identified prognostic variables in treatment guidelines to enhance patient outcomes.

Using machine learning algorithms to predict 28-day mortality in critically ill elderly patients with colorectal cancer

OBJECTIVE

To predict the 28-day mortality of critically ill, elderly patients with colorectal cancer (CRC) using five machine learning approaches.

METHODS

Data were extracted from the eICU Collaborative Research Database (eICU-CRD) (version 2.0) for a training cohort and from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) and Wuhan Union hospital for validation cohorts. Clinical information (i.e., demographics; initial laboratory tests; vital signs; outcomes) were collected. Five machine learning algorithms (LightGBM, decision tree, XGBoost, random forest, and ensemble model) and a logistic regression were applied for the prediction of 28-day mortality.

RESULTS

Overall, 693 patients were included from the eICU cohort, 181 patients from the MIMIC-IV cohort and 95 from the Wuhan Union cohort. Among the six machine learning models, the ensemble model exhibited the best predictive ability (AUC, 0.86), followed by random forest (AUC, 0.83) and LightGBM (AUC, 0.82) in the training cohort. The models also obtained the good predictive performance for the 28-day mortality in the validation cohorts.

CONCLUSIONS

We showed that machine learning algorithms can be used for the 28-day mortality prediction in critically ill, elderly patients with CRC.

Clinicopathologic and prognostic factors of patients with T3/T4 colorectal signet ring cell carcinoma: a population-based study

BACKGROUND

To evaluate cancer-specific survival (CSS) and construct a nomogram to predict the CSS of patients with colorectal signet ring cell carcinoma (SRCC).

METHODS

The data for patients with colorectal SRCC from 2000 to 2019 was identified from Surveillance, Epidemiology, and End Results (SEER) database. Propensity Score Matching (PSM) was used to minimize bias between SRCC and adenocarcinoma patients. Kaplan-Meier method and log-rank test were used to estimate the CSS. A nomogram was constructed based on the independent prognostic factors identified by univariate and multivariate Cox proportional hazards regression analyses. The model was evaluated by receiver operating characteristic (ROC) curves and calibration plots.

RESULTS

Poor CSS was more common in patients with colorectal SRCC, especially in patients with T4/N2 stage, tumor size > 80 mm, grade III-IV, and chemotherapy. Age, T/N stage, and tumor size > 80 mm were identified as independent prognostic indicators. And a prognostic nomogram was constructed and validated as an accurate model for the CSS of patients with colorectal SRCC by ROC curves and calibration plots.

CONCLUSION

Patients with colorectal SRCC have a poor prognosis. And the nomogram was expected to be effective in predicting the survival of patients with colorectal SRCC.

Statistical and machine learning approaches to predict the necessity for computed tomography in children with mild traumatic brain injury

BACKGROUND

Minor head trauma in children is a common reason for emergency department visits, but the risk of traumatic brain injury (TBI) in those children is very low. Therefore, physicians should consider the indication for computed tomography (CT) to avoid unnecessary radiation exposure to children. The purpose of this study was to statistically assess the differences between control and mild TBI (mTBI). In addition, we also investigate the feasibility of machine learning (ML) to predict the necessity of CT scans in children with mTBI.

METHODS AND FINDINGS

The study enrolled 1100 children under the age of 2 years to assess pre-verbal children. Other inclusion and exclusion criteria were per the PECARN study. Data such as demographics, injury details, medical history, and neurological assessment were used for statistical evaluation and creation of the ML algorithm. The number of children with clinically important TBI (ciTBI), mTBI on CT, and controls was 28, 30, and 1042, respectively. Statistical significance between the control group and clinically significant TBI requiring hospitalization (csTBI: ciTBI+mTBI on CT) was demonstrated for all nonparametric predictors except severity of the injury mechanism. The comparison between the three groups also showed significance for all predictors (p<0.05). This study showed that supervised ML for predicting the need for CT scan can be generated with 95% accuracy. It also revealed the significance of each predictor in the decision tree, especially the "days of life."

CONCLUSIONS

These results confirm the role and importance of each of the predictors mentioned in the PECARN study and show that ML could discriminate between children with csTBI and the control group.

Machine learning for prediction of in-hospital mortality in lung cancer patients admitted to intensive care unit

BACKGROUNDS

The in-hospital mortality in lung cancer patients admitted to intensive care unit (ICU) is extremely high. This study intended to adopt machine learning algorithm models to predict in-hospital mortality of critically ill lung cancer for providing relative information in clinical decision-making.

METHODS

Data were extracted from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) for a training cohort and data extracted from the Medical Information Mart for eICU Collaborative Research Database (eICU-CRD) database for a validation cohort. Logistic regression, random forest, decision tree, light gradient boosting machine (LightGBM), eXtreme gradient boosting (XGBoost), and an ensemble (random forest+LightGBM+XGBoost) model were used for prediction of in-hospital mortality and important feature extraction. The AUC (area under receiver operating curve), accuracy, F1 score and recall were used to evaluate the predictive performance of each model. Shapley Additive exPlanations (SHAP) values were calculated to evaluate feature importance of each feature.

RESULTS

Overall, there were 653 (24.8%) in-hospital mortality in the training cohort, and 523 (21.7%) in-hospital mortality in the validation cohort. Among the six machine learning models, the ensemble model achieved the best performance. The top 5 most influential features were the sequential organ failure assessment (SOFA) score, albumin, the oxford acute severity of illness score (OASIS) score, anion gap and bilirubin in random forest and XGBoost model. The SHAP summary plot was used to illustrate the positive or negative effects of the top 15 features attributed to the XGBoost model.

CONCLUSION

The ensemble model performed best and might be applied to forecast in-hospital mortality of critically ill lung cancer patients, and the SOFA score was the most important feature in all models. These results might offer valuable and significant reference for ICU clinicians' decision-making in advance.

Application of machine learning approaches to predict the 5-year survival status of patients with esophageal cancer

Background

Accurate prognostic estimation for esophageal cancer (EC) patients plays an important role in the process of clinical decision-making. The objective of this study was to develop an effective model to predict the 5-year survival status of EC patients using machine learning (ML) algorithms.

Methods

We retrieved the information of patients diagnosed with EC between 2010 and 2015 from the Surveillance, Epidemiology, and End Results (SEER) Program, including 24 features. A total of 8 ML models were applied to the selected dataset to classify the EC patients in terms of 5-year survival status, including 3 newly developed gradient boosting models (GBM), XGBoost, CatBoost, and LightGBM, 2 commonly used tree-based models, gradient boosting decision trees (GBDT) and random forest (RF), and 3 other ML models, artificial neural networks (ANN), naive Bayes (NB), and support vector machines (SVM). A 5-fold cross-validation was used in model performance measurement.

Results

After excluding records with missing data, the final study population comprised 10,588 patients. Feature selection was conducted based on the χ² test, however, the experiment results showed that the complete dataset provided better prediction of outcomes than the dataset with removal of non-significant features. Among the 8 models, XGBoost had the best performance [area under the receiver operating characteristic (ROC) curve (AUC): 0.852 for XGBoost, 0.849 for CatBoost, 0.850 for LightGBM, 0.846 for GBDT, 0.838 for RF, 0.844 for ANN, 0.833 for NB, and 0.789 for SVM]. The accuracy and logistic loss of XGBoost were 0.875 and 0.301, respectively, which were also the best performances. In the XGBoost model, the SHapley Additive exPlanations (SHAP) value was calculated and the result indicated that the four features: reason no cancer-directed surgery, Surg Prim Site, age, and stage group had the greatest impact on predicting the outcomes.

Conclusions

The XGBoost model and the complete dataset can be used to construct an accurate prognostic model for patients diagnosed with EC which may be applicable in clinical practice in the future.