Data Sheet 1_Predicting response to immunochemotherapy in EGFR-mutant lung adenocarcinoma

A retrospective multicenter study of 475 patients from two medical centers, allocated to training (332) and external validation (143) cohorts. The dataset was created by Shuai Qie and last updated in June 2026. It aims to predict response to immunochemotherapy in EGFR-mutant lung adenocarcinoma patients after third-generation TKI resistance.

Use Cases

• Training machine learning models to predict objective response rate (ORR) based on combined clinical and radiomic features.
• Developing habitat imaging models to characterize intratumoral heterogeneity from CT scans.
• Stratifying patients into risk groups for overall survival analysis using combined model predictions.
• Validating radiomics-based predictive performance through external testing across different medical centers.

Strengths

• Dataset includes 475 patient records from a multicenter study, with a defined split for training (332) and external validation (143).
• The combined predictive model achieved an area under the curve (AUC) of 0.904 in training and 0.890 in validation.
• Model performance was statistically validated, with significant survival differences between risk groups (training HR=3.688, validation HR=2.823).

Limitations

• Row count and specific column-level documentation are unknown, limiting suitability assessment.
• Data is from a retrospective study; prospective validation is noted as warranted.
• The dataset is small at 2.2 MB, which may limit the complexity of models it can support.

Provenance

Source

figshare

Collection Method

Retrospective multicenter study using computed tomography (CT) scans and clinical data.

Freshness

Last updated 2026-06-02 05:40:36.