MRI-Based Clinicoradiologic Model for Predicting MT-HCC with Postoperative Prognosis

A retrospective analysis of 338 hepatocellular carcinoma (HCC) patients from two centers, conducted by Yiman Li and last updated in May 2026. The dataset supports a model integrating alpha-fetoprotein (AFP) levels and MRI features to predict microvascular invasion-positive and tertiary lymphoid structures-positive HCC (MT-HCC). It was used to explore associations with recurrence-free survival and postoperative targeted therapy and immunotherapy (TKI-ICI) exposure.

Use Cases

• Training a binary classifier to predict MT-HCC status based on integrated clinicoradiologic features.
• Analyzing the relationship between MRI features (e.g., tumor capsule, hemorrhage) and pathological outcomes.
• Investigating differential recurrence-free survival patterns in patient subgroups defined by postoperative therapy exposure.

Strengths

• Model performance was validated on a separate cohort of 101 patients.
• The developed model achieved an area under the curve (AUC) of 0.831 in the validation cohort.
• Data is derived from a multi-center study involving 338 patient records.

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 very small at 33.2 KB, indicating limited raw data scope.

Provenance

Source

figshare, author Yiman Li.

Collection Method

Retrospective analysis of patient medical records and MRI assessments.

Time Range

null

Freshness

Last updated 2026-05-08 05:51:27; freshness should be verified.

Geography

null