Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques
Abstract 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 Ordin...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-024-53006-2 |
| _version_ | 1797274702932082688 |
|---|---|
| author | Samin Babaei Rikan Amir Sorayaie Azar Amin Naemi Jamshid Bagherzadeh Mohasefi Habibollah Pirnejad Uffe Kock Wiil |
| author_facet | Samin Babaei Rikan Amir Sorayaie Azar Amin Naemi Jamshid Bagherzadeh Mohasefi Habibollah Pirnejad Uffe Kock Wiil |
| author_sort | Samin Babaei Rikan |
| collection | DOAJ |
| description | Abstract 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. |
| first_indexed | 2024-03-07T15:02:04Z |
| format | Article |
| id | doaj.art-1ca083ad63994163a750984fcf1ca7ba |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-07T15:02:04Z |
| publishDate | 2024-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-1ca083ad63994163a750984fcf1ca7ba2024-03-05T19:04:25ZengNature PortfolioScientific Reports2045-23222024-01-0114111210.1038/s41598-024-53006-2Survival prediction of glioblastoma patients using modern deep learning and machine learning techniquesSamin Babaei Rikan0Amir Sorayaie Azar1Amin Naemi2Jamshid Bagherzadeh Mohasefi3Habibollah Pirnejad4Uffe Kock Wiil5Department of Computer Engineering, Urmia UniversityDepartment of Computer Engineering, Urmia UniversitySDU Health Informatics and Technology, The Maersk Mc-Kinney Moller Institute, University of Southern DenmarkDepartment of Computer Engineering, Urmia UniversityErasmus School of Health Policy and Management (ESHPM), Erasmus University RotterdamSDU Health Informatics and Technology, The Maersk Mc-Kinney Moller Institute, University of Southern DenmarkAbstract 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.https://doi.org/10.1038/s41598-024-53006-2 |
| spellingShingle | Samin Babaei Rikan Amir Sorayaie Azar Amin Naemi Jamshid Bagherzadeh Mohasefi Habibollah Pirnejad Uffe Kock Wiil Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques Scientific Reports |
| title | Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques |
| title_full | Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques |
| title_fullStr | Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques |
| title_full_unstemmed | Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques |
| title_short | Survival prediction of glioblastoma patients using modern deep learning and machine learning techniques |
| title_sort | survival prediction of glioblastoma patients using modern deep learning and machine learning techniques |
| url | https://doi.org/10.1038/s41598-024-53006-2 |
| work_keys_str_mv | AT saminbabaeirikan survivalpredictionofglioblastomapatientsusingmoderndeeplearningandmachinelearningtechniques AT amirsorayaieazar survivalpredictionofglioblastomapatientsusingmoderndeeplearningandmachinelearningtechniques AT aminnaemi survivalpredictionofglioblastomapatientsusingmoderndeeplearningandmachinelearningtechniques AT jamshidbagherzadehmohasefi survivalpredictionofglioblastomapatientsusingmoderndeeplearningandmachinelearningtechniques AT habibollahpirnejad survivalpredictionofglioblastomapatientsusingmoderndeeplearningandmachinelearningtechniques AT uffekockwiil survivalpredictionofglioblastomapatientsusingmoderndeeplearningandmachinelearningtechniques |