Select Assessment Configuration
Choose the trained model set you wish to use for this patient evaluation.
Loading available configurations…
An evidence-based tool that estimates a patient's risk of post-operative complications from routinely available pre-operative variables. Complete a short structured form and receive calibrated, model-derived probabilities to support — not replace — your clinical judgement.
Each estimate is produced by a model trained and validated specifically for that outcome, then expressed as an individual probability for the patient you describe.
Probability of any complication occurring within 30 days of surgery.
Probability of a decline in Karnofsky Performance Status by hospital discharge.
Probability of new post-operative neurological deficits attributable to surgery.
Probability of a severe or life-threatening post-operative complication.
The predictions draw on a curated registry of operated skull base meningiomas. For every outcome, several model families are trained and compared, and the best-performing model is retained for use in this tool.
You enter pre-operative variables — demographics, functional status, imaging findings and anatomical involvement — through a guided form.
Multiple algorithms are trained per outcome with temporal validation, and ranked by discrimination (ROC-AUC) before the strongest is selected.
The model returns an individual probability, shown against the operating decision threshold chosen during validation.
Results are presented as Low / Moderate / High bands to aid communication, always alongside the underlying percentage.
Choose a model configuration, describe the patient, and review the predicted risks.
Choose the trained model set you wish to use for this patient evaluation.
Hidden configurations are preserved on disk but not shown on the main screen.