JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

A CT-based deep learning-driven tool for automatic liver tumor detection and delineation in patients with cancer

  • 0Radiomics Group, Vall d'Hebron Institute of Oncology (VHIO), 08035 Barcelona, Spain.
Cell Reports. Medicine +

|

March 21, 2025

|

March 21, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

A new automated system, SALSA, accurately detects and quantifies liver tumors on CT scans. This tool shows promise for improving cancer diagnosis, staging, and treatment response evaluation.

Area Of Science

  • Medical Imaging
  • Artificial Intelligence in Oncology
  • Radiology

Background

  • Accurate liver tumor identification and quantification are critical for cancer patient management, impacting diagnosis, prognosis, and therapy assessment.
  • Existing methods for liver tumor analysis can be time-consuming and subject to inter-observer variability.

Purpose Of The Study

  • To introduce SALSA (system for automatic liver tumor segmentation and detection), a fully automated tool for liver tumor detection and delineation.
  • To evaluate SALSA's performance in accuracy, tumor identification, and volume quantification compared to state-of-the-art models and expert radiologists.

Main Methods

  • Development of SALSA using a dataset of 1,598 computed tomography (CT) scans with 4,908 liver tumors.
  • Validation of SALSA's detection and segmentation capabilities on external cohorts.
  • Assessment of tumor volume quantification's prognostic value.

Main Results

  • SALSA achieved high patient-wise detection precision (99.65%) and lesion-level precision (81.72%).
  • The system demonstrated a Dice Similarity Coefficient (DSC) of 0.760, surpassing state-of-the-art models and inter-radiologist agreement.
  • Automatic tumor volume quantification by SALSA showed significant prognostic value across various solid tumors (p = 0.028).

Conclusions

  • SALSA offers superior accuracy in liver tumor detection and volume quantification.
  • The automated system has the potential to serve as a medical device for cancer detection, staging, and response evaluation.
  • SALSA's performance indicates its utility in enhancing clinical decision-making for cancer patients.

Keywords: