Clinical Features for ECT Response in Adolescent MDD Patients

Bingyang Zha's dataset from 2026 contains clinical data from 503 adolescent patients with major depressive disorder (MDD) who underwent electroconvulsive therapy (ECT). It was used to build a machine learning model identifying baseline factors associated with poor treatment response. The simplified model uses two features: neutrophil-to-platelet ratio (NPR) and pre-treatment HAMD score.

Use Cases

• Predicting poor ECT response in adolescent MDD patients based on baseline NPR and HAMD scores.
• Training machine learning models for clinical stratification using routine admission data.
• Analyzing the relationship between ECT session completion and treatment response outcomes.
• Interpreting feature importance for clinical prediction models using SHAP analysis.

Strengths

• Dataset includes 503 patient records, providing a substantial sample size for analysis.
• Model achieved an AUC of 0.731 on the testing set using only two features, suggesting a concise predictive signal.
• Data is licensed under CC-BY-4.0, allowing for open sharing and reuse.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect temporal or source bias inherent to the retrospective, single-study design.

Provenance

Source

figshare

Collection Method

Retrospective clinical study.

Freshness

Last updated 2026-05-07 05:43:59; freshness should be verified.