Data Sheet 3_Integrating CT radiomics and transcriptomics: a biologically-informed machine

A dataset of 161 advanced laryngeal cancer patients used to build a machine learning model for predicting chemotherapy response. The data includes 1,321 extracted radiomics features from pre-treatment CT images, a constructed radiomics score, and clinical features. The model was developed by Xin Xiong and published on figshare in April 2026.

Use Cases

• Training machine learning models to predict chemotherapy response based on radiomics features and clinical predictors.
• Validating radiomics feature selection methods for cancer imaging analysis.
• Conducting SHAP analysis to interpret model predictions for clinical decision support.
• Benchmarking model performance (e.g., AUC, accuracy) in medical prediction tasks.

Strengths

• Includes data from 161 patient cases, providing a clinical study foundation.
• Model performance is documented with specific metrics, including an AUC of 0.810 in an external test set.
• The dataset is focused on a specific, clinically challenging prediction task for advanced laryngeal cancer.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the underlying feature data is unknown, which may limit suitability assessment.
• The dataset is small at 31.5 KB, indicating limited scope or a summary of results rather than raw patient-level data.

Provenance

Source

figshare, author Xin Xiong.

Collection Method

Retrospective analysis of patient data; radiomics features extracted from pre-treatment CT images.

Time Range

null

Freshness

Last updated 2026-04-15 04:36:57; freshness should be verified.

Geography

null