Prostate Cancer Radiomics Data for Surgical Decision Support

Fabian Tollens published a dataset on 2026-05-29 from a study of 249 patients undergoing radical prostatectomy. The data includes preoperative multiparametric MRI, clinical, laboratory, and pathological information used to build machine learning models for predicting surgical outcomes. The study reports model performance using area under the curve (AUC) metrics for predicting extracapsular extension, nerve-sparing approach decisions, and positive surgical margin risk.

Use Cases

• Training models to predict extracapsular extension based on multimodal clinical and imaging data.
• Evaluating the incremental value of radiomics features versus standard imaging parameters for surgical margin prediction.
• Developing a clinical decision support prototype for nerve-sparing surgery planning.
• Validating machine learning models for prostate cancer risk stratification in a prospective clinical setting.

Strengths

• Includes data from 249 patients (176 retrospective, 73 prospective), providing a basis for validation.
• Reports specific model performance metrics, such as an AUC of 0.90 for ECE prediction with imaging parameters.
• Data is multimodal, harmonizing MRI, clinical, laboratory, and pathological information.
• Prospective external validation was performed, suggesting tested generalizability.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 61.3 KB, indicating limited raw data or a summary document.

Provenance

Source

Fabian Tollens via figshare.

Collection Method

Single-center analysis of patients undergoing radical prostatectomy; data harmonized and imported into the International Radiomics Platform.

Freshness

Last updated 2026-05-29 04:49:04; freshness should be verified.