Semi-Supervised Learning for Hepatocellular Carcinoma Genomic Biomarker Prediction
September 16, 2026
A new multi-stage dataset for liver cancer classification was constructed using XGBoost and semi-supervised learning on genomic biomarkers. The dataset contains 770 patient samples across five classes, utilizing 11,150 gene expression features per sample.
HOW THIS AFFECTS YOU
●
researcherYou can leverage this semi-supervised approach to train on limited publicly available genomic data.
●
healthThis provides a more robust genomic dataset for training automated liver cancer diagnostic tools.