https://atlas.rsna.org/schemas/2025-11/model.json

WAW-TACE

https://zenodo.org/records/12741586

1.0

lp.ude.muw@kintrab.fotzsyzrk

Supported by the Medical University of Warsaw and the Warsaw University of Technology under collaborative grant “Integra WUM-PW” (no. 1W12/INTEGRA.1.6/N/23) and carried out with the support of the Laboratory of Bioinformatics and Computational Genomics and the High Performance Computing Center of the Faculty of Mathematics and Information Science Warsaw University of Technology.

Approved by the Bioethics Committee at the Medical University of Warsaw (decision no. AKBE/41/2024).

Dataset available at https://zenodo.org/records/12741586 (see also data sharing statement link https://zenodo.org/records/11063785).

Resource to enable development and validation of AI models for HCC detection, segmentation, prognostication (e.g., treatment response, survival) and related research.

Research use: patients with unresectable hepatocellular carcinoma treated with transarterial chemoembolization; multiphase pre-TACE abdominal CT with lesion and organ segmentations, radiomics features, and extensive clinical data.

Baseline multiphase abdominal CT (precontrast, late arterial, portal venous, delayed); associated clinical variables; segmentation masks (HCC lesions and 104 anatomic structures per phase).

Data intended for scientific research purposes only; refer to accompanying documentation (clinical_data, ct_scans, tumor_masks, organ_masks, radiomics_data, ct_hcc_metadata) for variable definitions and imaging parameters.

Only baseline imaging provided; not all patients have all CT phases; imaging from different CT systems with varying section thickness; no CT preprocessing or resampling prior to radiomics extraction; response assessments and PFS were based on reevaluation by one author following LI-RADS treatment response algorithm.