Deep learning models for predicting the survival of patients with hepatocellular carcinoma based on a surveillance, epidemiology, and end results (SEER) database analysis

Shoucheng Wang,Mingyi Shao,Yu Fu,Ruixia Zhao,Yunfei Xing,Liujie Zhang,Yang Xu
DOI: https://doi.org/10.1038/s41598-024-63531-9
IF: 4.6
2024-06-11
Scientific Reports
Abstract:Hepatocellular carcinoma (HCC) is a common malignancy with poor survival and requires long-term follow-up. Hence, we collected information on patients with Primary Hepatocellular Carcinoma in the United States from the Surveillance, Epidemiology, and EndResults (SEER) database. We used this information to establish a deep learning with a multilayer neural network (the NMTLR model) for predicting the survival rate of patients with Primary Hepatocellular Carcinoma. HCC patients pathologically diagnosed between January 2011 and December 2015 in the SEER (Surveillance, Epidemiology, and End Results) database of the National Cancer Institute of the United States were selected as study subjects. We utilized two deep learning-based algorithms (DeepSurv and Neural Multi-Task Logistic Regression [NMTLR]) and a machine learning-based algorithm (Random Survival Forest [RSF]) for model training. A multivariable Cox Proportional Hazards (CoxPH) model was also constructed for comparison. The dataset was randomly divided into a training set and a test set in a 7:3 ratio. The training dataset underwent hyperparameter tuning through 1000 iterations of random search and fivefold cross-validation. Model performance was assessed using the concordance index (C-index), Brier score, and Integrated Brier Score (IBS). The accuracy of predicting 1-year, 3-year, and 5-year survival rates was evaluated using Receiver Operating Characteristic (ROC) curves, calibration plots, and Area Under the Curve (AUC). The primary outcomes were the 1-year, 3-year, and 5-year overall survival rates. Models were developed using DeepSurv, NMTLR, RSF, and Cox Proportional Hazards regression. Model differentiation was evaluated using the C-index, calibration with concordance plots, and risk stratification capability with the log-rank test. The study included 2197 HCC patients, randomly divided into a training cohort (70%, n = 1537) and a testing cohort (30%, n = 660). Clinical characteristics between the two cohorts showed no significant statistical difference (p > 0.05). The deep learning models outperformed both RSF and CoxPH models, with C-indices of 0.735 (NMTLR) and 0.731 (DeepSurv) in the test dataset. The NMTLR model demonstrated enhanced accuracy and well-calibrated survival estimates, achieving an Area Under the Curve (AUC) of 0.824 for 1-year survival predictions, 0.813 for 3-year, and 0.803 for 5-year survival rates. This model's superior calibration and discriminative ability enhance its utility for clinical prognostication in Primary Hepatocellular Carcinoma. We deployed the NMTLR model as a web application for clinical practice. The NMTLR model have potential advantages over traditional linear models in prognostic assessment and treatment recommendations. This novel analytical approach may provide reliable information on individual survival and treatment recommendations for patients with primary liver cancer.
multidisciplinary sciences
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to improve the accuracy of predicting the survival rate of patients with hepatocellular carcinoma (HCC). Specifically, the researchers used the detailed clinical data in the Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute in the United States to construct a multi - layer neural network model based on deep learning (NMTLR model) to predict the survival rate of HCC patients. By comparing with the traditional Random Survival Forest (RSF) and Cox proportional - hazards model, the performance of these models in predicting the 1 - year, 3 - year, and 5 - year survival rates was evaluated. ### Research Background - **Hepatocellular carcinoma (HCC)** is a common malignant tumor with a poor survival rate and requires long - term follow - up. - **SEER database** collects data from 18 regional cancer registries in the United States, covering about 28% of the US population, providing a representative and diverse sample base and enhancing the generalizability of the research results. - **Limitations of existing models**: Traditional linear models (such as the Cox proportional - hazards model) assume that the influence of each predictor at different follow - up times is the same, ignoring the differences of individual patients at different time points. In addition, these models adopt linear assumptions and do not consider non - linear analysis in real - world clinical aspects. ### Research Methods - **Data source**: Extract patient data with a pathological diagnosis of primary hepatocellular carcinoma between 2010 and 2015 from the SEER database. - **Model construction**: Use four algorithms (Cox proportional - hazards model, DeepSurv, NMTLR, RSF) for model training, and optimize hyper - parameters through random search and five - fold cross - validation. - **Performance evaluation**: Use indicators such as the concordance index (C - index), Brier score, integrated Brier score (IBS), receiver operating characteristic curve (ROC curve), and area under the curve (AUC) to evaluate model performance. ### Main Findings - **Model performance**: The C - index of the NMTLR model on the test set is 0.7353, which is significantly better than that of the Cox proportional - hazards model (C - index of 0.6837) and other machine - learning models. - **Calibration and discrimination ability**: The NMTLR model shows the highest AUC values in predicting the 1 - year, 3 - year, and 5 - year survival rates, which are 0.824, 0.813, and 0.803 respectively, showing its superiority in survival prediction. - **Practical application**: The best model (NMTLR) is deployed as an online calculator for clinicians to use in real - time to improve the efficiency and accuracy of clinical decision - making. ### Conclusion This study uses deep - learning technology to process and analyze large - scale cancer data, aiming to provide more accurate survival predictions for patients with hepatocellular carcinoma, thereby providing a scientific basis for clinical decision - making. The NMTLR model performs well on multiple evaluation indicators and has high practical value.