Sign in Subscribe
Machine learning

Breakthrough in Liver Cancer Diagnosis Using Machine Learning

Strategic Insights

— 1 min read

Researchers from Technical University Dresden (TU Dresden) have developed a machine learning-based approach for accurate intraoperative liver tumor classification, potentially improving surgical outcomes and reducing recurrence risks. According to the study, the method uses fiber-based attenuated total reflection infrared (ATR IR) spectroscopy to analyze fresh liver tissue from surgical patients, providing a rapid and objective diagnosis. The researchers claim that this label-free biochemical approach can distinguish between normal liver tissue, hepatocellular carcinoma (HCC), cholangiocellular carcinoma (CCC), and metastases with high accuracy.

Key Takeaways:

  • The study assessed the effectiveness of fiber-based ATR IR spectroscopy combined with supervised machine learning for intraoperative liver tumor classification in 69 surgical patients.
  • The proposed approach accurately classified normal liver tissue and tumor subtypes (HCC, CCC, metastases) with a sensitivity of 0.89, specificity of 0.92, and accuracy of 0.90.
  • The method distinguished HCC from CCC and metastases based on differences in glycogen content and structural compactness of tumor tissue.
  • The three-group classification of tumor subtypes yielded an average accuracy of 0.90.
  • The study concluded that this label-free biochemical approach may enhance surgical precision and reduce recurrence risks across the full range of solid tumor entities.

Statistics:

  • 69 surgical patients were analyzed using the proposed approach.
  • The sensitivity of the proposed approach was 0.89.
  • The specificity of the proposed approach was 0.92.
  • The accuracy of the proposed approach was 0.90.
  • The average accuracy of the three-group classification of tumor subtypes was 0.90.

Sources:

  • NewsRx. Study Data from Technical University Dresden (TU Dresden) Update Understanding of Machine Learning (algorithm-based Intraoperative Diagnosis of Liver Tumors Using Infrared Spectroscopy). Health & Medicine Week. July 18, 2025; p 988.
  • Technical University Dresden (TU Dresden). Faculty of Medicine, Dept. of Visceral Thoracic & Vascular Surgery.
  • Scientific Reports. (2025;15(1)).
  • Nature Portfolio. Heidelberger Platz 3, Berlin, 14197, Germany.

Read more