GNN-surv: Discrete-Time Survival Prediction Using Graph Neural Networks
Survival prediction models play a key role in patient prognosis and personalized treatment. However, their accuracy can be improved by incorporating patient similarity networks, which uncover complex data patterns. Our study uses Graph Neural Networks (GNNs) to enhance discrete-time survival predict...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/10/9/1046 |
| _version_ | 1797581253602443264 |
|---|---|
| author | So Yeon Kim |
| author_facet | So Yeon Kim |
| author_sort | So Yeon Kim |
| collection | DOAJ |
| description | Survival prediction models play a key role in patient prognosis and personalized treatment. However, their accuracy can be improved by incorporating patient similarity networks, which uncover complex data patterns. Our study uses Graph Neural Networks (GNNs) to enhance discrete-time survival predictions (GNN-surv) by leveraging relationships in these networks. We build these networks using cancer patients’ genomic and clinical data and train various GNN models on them, integrating Logistic Hazard and PMF survival models. GNN-surv models exhibit superior performance in survival prediction across two urologic cancer datasets, outperforming traditional MLP models. They maintain robustness and effectiveness under varying graph construction hyperparameter <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>μ</mi></semantics></math></inline-formula> values, with performance boosts of up to 14.6% and 7.9% in the time-dependent concordance index and reductions in the integrated brier score of 26.7% and 24.1% in the BLCA and KIRC datasets, respectively. Notably, these models also maintain their effectiveness across three different types of GNN models, suggesting potential adaptability to other cancer datasets. The superior performance of our GNN-surv models underscores their wide applicability in the fields of oncology and personalized medicine, providing clinicians with a more accurate tool for patient prognosis and personalized treatment planning. Future studies can further optimize these models by incorporating other survival models or additional data modalities. |
| first_indexed | 2024-03-10T23:01:44Z |
| format | Article |
| id | doaj.art-b63e182f778e4a50acb5a5207b6fb692 |
| institution | Directory Open Access Journal |
| issn | 2306-5354 |
| language | English |
| last_indexed | 2024-03-10T23:01:44Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Bioengineering |
| spelling | doaj.art-b63e182f778e4a50acb5a5207b6fb6922023-11-19T09:36:55ZengMDPI AGBioengineering2306-53542023-09-01109104610.3390/bioengineering10091046GNN-surv: Discrete-Time Survival Prediction Using Graph Neural NetworksSo Yeon Kim0Department of Artificial Intelligence, Ajou University, Suwon 16499, Republic of KoreaSurvival prediction models play a key role in patient prognosis and personalized treatment. However, their accuracy can be improved by incorporating patient similarity networks, which uncover complex data patterns. Our study uses Graph Neural Networks (GNNs) to enhance discrete-time survival predictions (GNN-surv) by leveraging relationships in these networks. We build these networks using cancer patients’ genomic and clinical data and train various GNN models on them, integrating Logistic Hazard and PMF survival models. GNN-surv models exhibit superior performance in survival prediction across two urologic cancer datasets, outperforming traditional MLP models. They maintain robustness and effectiveness under varying graph construction hyperparameter <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>μ</mi></semantics></math></inline-formula> values, with performance boosts of up to 14.6% and 7.9% in the time-dependent concordance index and reductions in the integrated brier score of 26.7% and 24.1% in the BLCA and KIRC datasets, respectively. Notably, these models also maintain their effectiveness across three different types of GNN models, suggesting potential adaptability to other cancer datasets. The superior performance of our GNN-surv models underscores their wide applicability in the fields of oncology and personalized medicine, providing clinicians with a more accurate tool for patient prognosis and personalized treatment planning. Future studies can further optimize these models by incorporating other survival models or additional data modalities.https://www.mdpi.com/2306-5354/10/9/1046discrete survival modelGraph Neural Networkspatient similarity networksurvival predictiontime-to-event prediction |
| spellingShingle | So Yeon Kim GNN-surv: Discrete-Time Survival Prediction Using Graph Neural Networks Bioengineering discrete survival model Graph Neural Networks patient similarity network survival prediction time-to-event prediction |
| title | GNN-surv: Discrete-Time Survival Prediction Using Graph Neural Networks |
| title_full | GNN-surv: Discrete-Time Survival Prediction Using Graph Neural Networks |
| title_fullStr | GNN-surv: Discrete-Time Survival Prediction Using Graph Neural Networks |
| title_full_unstemmed | GNN-surv: Discrete-Time Survival Prediction Using Graph Neural Networks |
| title_short | GNN-surv: Discrete-Time Survival Prediction Using Graph Neural Networks |
| title_sort | gnn surv discrete time survival prediction using graph neural networks |
| topic | discrete survival model Graph Neural Networks patient similarity network survival prediction time-to-event prediction |
| url | https://www.mdpi.com/2306-5354/10/9/1046 |
| work_keys_str_mv | AT soyeonkim gnnsurvdiscretetimesurvivalpredictionusinggraphneuralnetworks |