Blood Biomarker Profiles for Spastic Cerebral Palsy Analysis

A retrospective study analyzes blood biomarker features from 330 children with spastic cerebral palsy and 150 healthy controls. It applies a machine-learning framework to identify 7 biomarkers independently associated with the condition and stratifies patients by Gross Motor Function Classification System (GMFCS) severity. The dataset includes derived systemic immune–inflammation indices and serum biochemical parameters.

Use Cases

• Train a logistic regression model to classify spastic cerebral palsy using the 7 identified biomarkers (MPV, CHO, DBIL, PDW, BASO%, GLB, MCHC).
• Apply SHAP analysis to interpret the feature importance of blood parameters like ALT, WBC, and ALP in predicting severe vs. mild GMFCS subgroups.
• Build a predictive nomogram for disease discrimination based on the AUC performance metrics (e.g., 0.972) reported for the biomarker panel.
• Analyze the nonlinear associations between key indicators and outcome risk using the restricted cubic splines (RCS) methodology described.

Strengths

• Includes data from 480 total subjects (330 patients, 150 controls), providing a substantive clinical cohort.
• Model validation reported a high AUC of 0.972 for discriminating SCP from controls in an independent test set.
• Applies an interpretable machine-learning framework (LASSO, SHAP) for feature selection and importance ranking.

Limitations

• The dataset is small in file size (1017.1 KB) and is contained within a PDF, limiting direct computational analysis of raw tabular data.
• Subgroup analysis for severity (n=99 in test set) showed modest discriminative ability (AUC 0.717), indicating exploratory findings.
• Geographic origin and specific time range of the retrospective study are not specified, limiting generalizability.

Provenance

Source

figshare, authored by Yanjun Mo.

Collection Method

Retrospective study collecting complete blood count and serum biochemical parameters from enrolled children.

Freshness

Last updated March 2026.