Data Sheet 1_Prediction of cervical stromal invasion using ultrasound radiomics: from conv

294 patients with pathologically confirmed endometrial cancer were analyzed to develop a radiomics prediction model. The dataset likely contains extracted radiomics features and clinical predictors used to build models for preoperative assessment. Author Xiaoli Peng published the data on figshare under a CC-BY-4.0 license in May 2026.

Use Cases

• Training machine learning models for cervical stromal invasion prediction based on extracted ultrasound radiomics features.
• Benchmarking clinical prediction models against a radiomics-only model based on reported AUC performance metrics.
• Analyzing the correlation between clinical predictors (like CA-125 and tumor size) and radiomics features for endometrial cancer.
• Developing combined nomogram models for personalized treatment planning based on the described radiomic and clinical scores.

Strengths

• Dataset is derived from a study of 294 patients with pathologically confirmed endometrial cancer.
• The radiomics model described achieved a high AUC of 0.906 in the test cohort.
• Data includes 837 extracted radiomics features and six identified clinical predictors.

Limitations

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

Provenance

Source

figshare

Collection Method

Retrospective analysis of ultrasound images from patients; regions of interest were manually segmented and features extracted.

Freshness

Last updated 2026-05-07 04:45:31; freshness should be verified.