Table 1_Integrated model based on ultrasound attenuation and metabolic biomarkers for noni

A dataset of 172 participants from a single-center study evaluating models for noninvasive assessment of hepatic fat fraction categories in metabolic dysfunction-associated steatotic liver disease (MASLD). The data includes ultrasound attenuation values and serum levels of ALT, AST, and ferritin, referenced against quantitative computed tomography categories. The dataset was authored by Zejun Ma and last updated in May 2026.

Use Cases

• Comparing the diagnostic performance of ultrasound attenuation versus metabolic biomarkers for detecting hepatic steatosis.
• Training or validating integrated models that combine imaging and laboratory data for liver fat categorization.
• Analyzing model performance across demographic subgroups such as females, younger participants, and those with higher BMI.

Strengths

• Includes data from 172 participants with quantitative computed tomography (QCT) reference labels.
• Reports specific model performance metrics, such as an AUC of 0.847 for the USAT-only model.
• Uses internal validation methods including five-fold stratified cross-validation and bootstrap analysis.

Limitations

• Row count and column-level documentation are unknown, limiting suitability assessment.
• The study is single-center and focused on a Chinese health-examination cohort, which may limit generalizability.
• Multiclass categorization, particularly for the intermediate Category 2 (n=31), showed weak discrimination.

Provenance

Source

Zejun Ma via figshare.

Collection Method

Data collected from a clinical study enrolling participants categorized by QCT, with USAT and serum biomarker measurements.

Freshness

Last updated 2026-05-29 04:53:49; freshness should be verified.

Geography

Likely China, based on the description of a single-center Chinese health-examination cohort.