Wallerian Degeneration Prognostic Model: Clinical and Imaging Data for 285 Patients

285 Wallerian degeneration patient records from Xinhua Hospital affiliated with Dalian University, with a median age of 63 years. Zhiqi Yu developed and validated a machine learning model to predict poor prognosis based on clinical indicators and imaging data. The dataset was last updated on 2026-05-26.

Use Cases

• Predicting poor prognosis in Wallerian degeneration patients based on NIHSS score and atrial fibrillation status.
• Evaluating feature importance for neurological outcomes using SHAP analysis on clinical and imaging predictors.
• Training machine learning models for risk assessment using variables like diabetes, hypertension, and hyperlipidemia.
• Comparing the performance of different ML algorithms (e.g., AdaBoost) for clinical prediction tasks.

Strengths

• Includes 285 patient records with 161 (56.49%) having a poor prognosis outcome.
• Identifies 10 candidate predictors for poor prognosis, with seven showing statistical significance.
• The best-performing AdaBoost model achieved an AUC of 0.880, indicating strong discriminative performance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is known (285), but the specific number of features per row is unknown.
• Data is sourced from a single hospital, which may reflect geographic or institutional bias.

Provenance

Source

Xinhua Hospital, affiliated with Dalian University.

Collection Method

Clinical patient data were randomly split into a training set and a validation set in a 7/3 ratio.

Freshness

Last updated 2026-05-26 04:39:05; freshness should be verified.