Wallerian Degeneration Patient Data for Predicting Adverse Prognosis

285 patient records from Xinhua Hospital were used to develop a machine learning model predicting poor prognosis in Wallerian degeneration. The dataset includes clinical indicators like NIHSS score, diabetes, and hypertension, alongside imaging data of brain regions. An AdaBoost model achieved an AUC of 0.880, with SHAP analysis highlighting NIHSS score, atrial fibrillation, and hypertension as key predictors.

Use Cases

• Predicting poor prognosis in Wallerian degeneration patients based on clinical and imaging features.
• Evaluating feature importance for neurological outcomes using SHAP analysis.
• Comparing the performance of eight machine learning models, including AdaBoost, for clinical risk assessment.
• Identifying high-risk patient subgroups for early intervention based on predictors like NIHSS score and atrial fibrillation.

Strengths

• Includes data from 285 patients with a median age of 63 years.
• Model development used a 7/3 train-validation split and evaluated eight algorithms.
• The best-performing AdaBoost model achieved an AUC of 0.880.
• Identified ten candidate predictors, with seven showing statistical significance.

Limitations

• Row count and column-level documentation are unknown, requiring manual inspection after download.
• Data is from a single hospital, which may limit generalizability.
• The dataset is small at 72.3 KB, indicating limited scope.

Provenance

Source

Xinhua Hospital, affiliated with Dalian University.

Collection Method

Clinical patient data were randomly split into training and validation sets.

Freshness

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