Comparison of models for stroke-free survival prediction in patients with CADASIL
Abstract Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy, which is caused by mutations of the NOTCH3 gene, has a large heterogeneous progression, presenting with declines of various clinical scores and occurrences of various clinical event. To help assess d...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-12-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-023-49552-w |
| _version_ | 1797388341396635648 |
|---|---|
| author | Henri Chhoa Hugues Chabriat Sylvie Chevret Lucie Biard |
| author_facet | Henri Chhoa Hugues Chabriat Sylvie Chevret Lucie Biard |
| author_sort | Henri Chhoa |
| collection | DOAJ |
| description | Abstract Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy, which is caused by mutations of the NOTCH3 gene, has a large heterogeneous progression, presenting with declines of various clinical scores and occurrences of various clinical event. To help assess disease progression, this work focused on predicting the composite endpoint of stroke-free survival time by comparing the performance of Cox proportional hazards regression to that of machine learning models using one of four feature selection approaches applied to demographic, clinical and magnetic resonance imaging observational data collected from a study cohort of 482 patients. The quality of the modeling process and the predictive performance were evaluated in a nested cross-validation procedure using the time-dependent Brier Score and AUC at 5 years from baseline, the former measuring the overall performance including calibration and the latter highlighting the discrimination ability, with both metrics taking into account the presence of right-censoring. The best model for each metric was the componentwise gradient boosting model with a mean Brier score of 0.165 and the random survival forest model with a mean AUC of 0.773, both combined with the LASSO feature selection method. |
| first_indexed | 2024-03-08T22:38:21Z |
| format | Article |
| id | doaj.art-6198d8d459a44878bcca05f2d0fa8b02 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-08T22:38:21Z |
| publishDate | 2023-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-6198d8d459a44878bcca05f2d0fa8b022023-12-17T12:17:45ZengNature PortfolioScientific Reports2045-23222023-12-0113111010.1038/s41598-023-49552-wComparison of models for stroke-free survival prediction in patients with CADASILHenri Chhoa0Hugues Chabriat1Sylvie Chevret2Lucie Biard3ECSTRRA Team, Université Paris Cité, UMR1153, INSERMCentre NeuroVasculaire Translationnel - Centre de Référence CERVCO, DMU NeuroSciences, Hôpital Lariboisière, GHU APHP-Nord, Université Paris CitéECSTRRA Team, Université Paris Cité, UMR1153, INSERMECSTRRA Team, Université Paris Cité, UMR1153, INSERMAbstract Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy, which is caused by mutations of the NOTCH3 gene, has a large heterogeneous progression, presenting with declines of various clinical scores and occurrences of various clinical event. To help assess disease progression, this work focused on predicting the composite endpoint of stroke-free survival time by comparing the performance of Cox proportional hazards regression to that of machine learning models using one of four feature selection approaches applied to demographic, clinical and magnetic resonance imaging observational data collected from a study cohort of 482 patients. The quality of the modeling process and the predictive performance were evaluated in a nested cross-validation procedure using the time-dependent Brier Score and AUC at 5 years from baseline, the former measuring the overall performance including calibration and the latter highlighting the discrimination ability, with both metrics taking into account the presence of right-censoring. The best model for each metric was the componentwise gradient boosting model with a mean Brier score of 0.165 and the random survival forest model with a mean AUC of 0.773, both combined with the LASSO feature selection method.https://doi.org/10.1038/s41598-023-49552-w |
| spellingShingle | Henri Chhoa Hugues Chabriat Sylvie Chevret Lucie Biard Comparison of models for stroke-free survival prediction in patients with CADASIL Scientific Reports |
| title | Comparison of models for stroke-free survival prediction in patients with CADASIL |
| title_full | Comparison of models for stroke-free survival prediction in patients with CADASIL |
| title_fullStr | Comparison of models for stroke-free survival prediction in patients with CADASIL |
| title_full_unstemmed | Comparison of models for stroke-free survival prediction in patients with CADASIL |
| title_short | Comparison of models for stroke-free survival prediction in patients with CADASIL |
| title_sort | comparison of models for stroke free survival prediction in patients with cadasil |
| url | https://doi.org/10.1038/s41598-023-49552-w |
| work_keys_str_mv | AT henrichhoa comparisonofmodelsforstrokefreesurvivalpredictioninpatientswithcadasil AT hugueschabriat comparisonofmodelsforstrokefreesurvivalpredictioninpatientswithcadasil AT sylviechevret comparisonofmodelsforstrokefreesurvivalpredictioninpatientswithcadasil AT luciebiard comparisonofmodelsforstrokefreesurvivalpredictioninpatientswithcadasil |