MRI-Based Habitat Radiomics for Predicting Breast Cancer Treatment Response

A multicenter study of 336 women with node-positive breast cancer from two institutions. The dataset was used to develop a nomogram integrating MRI radiomic features from tumor habitats and clinicopathologic variables to predict axillary pathological complete response after neoadjuvant therapy. The model achieved an AUC of 0.845 in the training cohort and was validated externally.

Use Cases

• Predicting axillary pathological complete response based on MRI radiomic features.
• Developing a clinical nomogram based on integrated radiomics and clinicopathologic variables.
• Evaluating model performance using ROC analysis, calibration, and decision curve analysis.
• Stratifying axillary risk in breast cancer patients after neoadjuvant therapy.

Strengths

• Multicenter design involving two institutions.
• Model performance was validated in independent external test cohorts with AUCs around 0.754-0.755.
• The nomogram model achieved an AUC of 0.845 in the training cohort.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is small, at 677.1 KB.

Provenance

Source

figshare

Collection Method

Retrospective analysis of patients who underwent pretreatment breast MRI.

Freshness

Last updated 2026-05-13 04:45:05; freshness should be verified.

Geography

Multicenter study; specific geography not stated.