How to use the Surveillance, Epidemiology, and End Results (SEER) data: research design and methodology

Che, Wen-Qiang; Li, Yuan-Jie; Tsang, Chi-Kwan; Wang, Yu-Jiao; Chen, Zheng; Wang, Xiang-Yu; Xu, An-Ding; Lyu, Jun

doi:10.1186/s40779-023-00488-2

Cited by 23 publications

(11 citation statements)

References 107 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data collected by the SEER program are accurate and complete on all cancers and in all regions. It employs a continuous quality control and improvement program that minimizes and corrects errors and ensures the data’s high quality 55 . The glioblastoma patients’ data between 2007 and 2016 was picked from SEER for this study.…”

Section: Methodsmentioning

confidence: 99%

Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques

Babaei Rikan,

Sorayaie Azar,

Naemi

et al. 2024

Sci Rep

View full text Add to dashboard Cite

In this study, we utilized data from the Surveillance, Epidemiology, and End Results (SEER) database to predict the glioblastoma patients’ survival outcomes. To assess dataset skewness and detect feature importance, we applied Pearson's second coefficient test of skewness and the Ordinary Least Squares method, respectively. Using two sampling strategies, holdout and five-fold cross-validation, we developed five machine learning (ML) models alongside a feed-forward deep neural network (DNN) for the multiclass classification and regression prediction of glioblastoma patient survival. After balancing the classification and regression datasets, we obtained 46,340 and 28,573 samples, respectively. Shapley additive explanations (SHAP) were then used to explain the decision-making process of the best model. In both classification and regression tasks, as well as across holdout and cross-validation sampling strategies, the DNN consistently outperformed the ML models. Notably, the accuracy were 90.25% and 90.22% for holdout and five-fold cross-validation, respectively, while the corresponding R2 values were 0.6565 and 0.6622. SHAP analysis revealed the importance of age at diagnosis as the most influential feature in the DNN's survival predictions. These findings suggest that the DNN holds promise as a practical auxiliary tool for clinicians, aiding them in optimal decision-making concerning the treatment and care trajectories for glioblastoma patients.

show abstract

Section: Methodsmentioning

confidence: 99%

Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques

Babaei Rikan,

Sorayaie Azar,

Naemi

et al. 2024

Sci Rep

View full text Add to dashboard Cite

show abstract

“…This was a retrospective cohort study predicting the survival outcomes of patients with GBM and identifying the patients’ ITE to determine whether an individual is better suited to receive GTR or SpTR with DL models. All participants included in this study were selected from the Surveillance, Epidemiology, and End Results 18 (SEER 18) database, which tracks patients with cancer from 18 regions of the United States, and the population in SEER 18 represents approximately 27.8% of the US population ( 20 ). This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines ( 21 ).…”

Section: Methodsmentioning

confidence: 99%

Individualized survival prediction and surgery recommendation for patients with glioblastoma

Zhu,

Wang,

Jing

et al. 2024

Front. Med.

View full text Add to dashboard Cite

BackgroundThere is a lack of individualized evidence on surgical choices for glioblastoma (GBM) patients.AimThis study aimed to make individualized treatment recommendations for patients with GBM and to determine the importance of demographic and tumor characteristic variables in the selection of extent of resection.MethodsWe proposed Balanced Decision Ensembles (BDE) to make survival predictions and individualized treatment recommendations. We developed several DL models to counterfactually predict the individual treatment effect (ITE) of patients with GBM. We divided the patients into the recommended (Rec.) and anti-recommended groups based on whether their actual treatment was consistent with the model recommendation.ResultsThe BDE achieved the best recommendation effects (difference in restricted mean survival time (dRMST): 5.90; 95% confidence interval (CI), 4.40–7.39; hazard ratio (HR): 0.71; 95% CI, 0.65–0.77), followed by BITES and DeepSurv. Inverse probability treatment weighting (IPTW)-adjusted HR, IPTW-adjusted OR, natural direct effect, and control direct effect demonstrated better survival outcomes of the Rec. group.ConclusionThe ITE calculation method is crucial, as it may result in better or worse recommendations. Furthermore, the significant protective effects of machine recommendations on survival time and mortality indicate the superiority of the model for application in patients with GBM. Overall, the model identifies patients with tumors located in the right and left frontal and middle temporal lobes, as well as those with larger tumor sizes, as optimal candidates for SpTR.

show abstract

“…Therefore, we performed a comprehensive analysis on a large registry to evaluate the frequency and likelihood of SPM rates and risks in MM patients, as well as to investigate the contributing risk factors. The Surveillance, Epidemiology, and End Results (SEER) program offers detailed population-based data on cancer diagnosis stage and patient outcomes, including survival rates [10] . The primary aim of this study was to analyze SEER database and determine the occurrence of additional malignancies in individuals with MM, as well as identify the associated survival risk factors.…”

Section: Introductionmentioning

confidence: 99%

Investigating additional malignancies rates and overall survival of multiple myeloma patients: a SEER database study

Dong,

Ye,

Liu

2024

Preprint

View full text Add to dashboard Cite

Objective The development of effective treatments for multiple myeloma (MM) has resulted in longer surviving of patients. However, these patients are at risk of developing additional malignancies. This study aimed to investigate rates of additional malignancies in MM patients and the median of latency. Findings from this study are expected to improve patient outcomes. Methods Data of MM patients with additional malignancies between 1992 and 2020 was extracted from SEER. The developed of different additional systems and malignancies were analyzed, including solid tumors and hematology. A standardized incidence ratio (SIR) for second primary malignancies (SPMs) was calculated to determine the MM risk. Cox regressions were performed to predict the risk factors associated with MM in patients with additional malignancies. Two nomograms were developed to determine the patient's probability prognosis. Results Regarding different additional malignancies, the most common types of solid tumors of prior malignancies (PMs) and SPMs are prostate cancer. Non-Hodgkin's lymphoma (NHL) as an additional malignancy of MM is the most frequent types of hematology. To construct the diagnostic nomogram, 3676 MM patients with SPMs were included. The nomogram, consisting of six independent risk factors, demonstrated high accuracy with a 3-year survival area under the curve (AUC) of 0.94400 and a 5-year survival AUC of 0.94377. Moreover, 1663 MM patients with PMs were analyzed. The diagnostic nomogram, comprising four independent risk factors, showed moderate accuracy (P < 0.05) with a 3-year survival AUC of 0.62200 and a 5-year survival AUC of 0.61743. Conclusion The present analysis and nomograms are expected to help clinicians in assessing individual survival outcomes of additional malignancies and implementing personalized clinical decisions. Several factors can decrease the survival rate of MM patients with SPMs including bones and joints in solid tumors, hematopoietic system disorders, occurrence of third or subsequent primary malignancies, and advanced age. During the assessment of MM patients with PMs, various parameters should be considered such as advanced age, presence of non-Hodgkin lymphoma, and development of third or subsequent primary malignancies.

show abstract

How to use the Surveillance, Epidemiology, and End Results (SEER) data: research design and methodology

Cited by 23 publications

References 107 publications

Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques

Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques

Individualized survival prediction and surgery recommendation for patients with glioblastoma

Investigating additional malignancies rates and overall survival of multiple myeloma patients: a SEER database study

Contact Info

Product

Resources

About