NCT04092543

Validation of DRAGON Versus a Simplified DRAGON/Machine Learning

Prediction of Outcome After Acute Stroke: Validation of CT-DRAGON Versus a Simplified CT-DRAGON Score

• Status: Unknown
• Phase: —
• Study type: Observational
• Enrollment: 700 (estimated)
• Sponsor: Ziekenhuis Oost-Limburg · Academic / Other
• Sex: All
• Age: —
• Healthy volunteers: Not accepted

Summary

The CT-DRAGON score can predict long-term functional outcome after acute stroke treated by thrombolysis. However, implementation in clinical practice is hampered by a lack of validation in the broad spectrum of stroke patients undergoing thrombectomy, whether or not in combination with thrombolysis or conservative treatment. Furthermore, the CT-DRAGON score considers multiple items, which are not always readily available in every setting. This study aims to investigate whether either a simplified version of the CT-DRAGON score with only three clinical items or a machine learning technique could be as powerful and more feasible.

Detailed description

The investigators aim to validate the CT-DRAGON score in all ischaemic stroke localisations and for all treatment options, including a conservative treatment policy. The predictability will then be compared with on the one hand simplified prognostic models that include only a selective set of highly predictive parameters that have already been described in the literature, such as patient age, National Institutes of Health Stroke Scale (NIHSS) and pre-stroke modified Rankin Scale (mRS). On the other hand, machine learning techniques, that incorporate a large set of variables and have recently shown some promising results, will also be applied to predict long-term outcome after ischaemic stroke.

Conditions

• Stroke, Acute

Timeline

• Start date: 2019-03-01
• Primary completion: 2021-08-01
• Completion: 2021-11-01
• First posted: 2019-09-17
• Last updated: 2019-09-17

Locations

1 site across 1 country: Belgium