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

Journal: Cell reports. Medicine
PMID:

Abstract

Liver tumors, whether primary or metastatic, significantly impact the outcomes of patients with cancer. Accurate identification and quantification are crucial for effective patient management, including precise diagnosis, prognosis, and therapy evaluation. We present SALSA (system for automatic liver tumor segmentation and detection), a fully automated tool for liver tumor detection and delineation. Developed on 1,598 computed tomography (CT) scans and 4,908 liver tumors, SALSA demonstrates superior accuracy in tumor identification and volume quantification, outperforming state-of-the-art models and inter-reader agreement among expert radiologists. SALSA achieves a patient-wise detection precision of 99.65%, and 81.72% at lesion level, in the external validation cohorts. Additionally, it exhibits good overlap, achieving a dice similarity coefficient (DSC) of 0.760, outperforming both state-of-the-art and the inter-radiologist assessment. SALSA's automatic quantification of tumor volume proves to have prognostic value across various solid tumors (p = 0.028). SALSA's robust capabilities position it as a potential medical device for automatic cancer detection, staging, and response evaluation.

Authors

  • Maria Balaguer-Montero
    Radiomics Group, Vall d'Hebron Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • Adrià Marcos Morales
    Radiomics Group, Vall d'Hebron Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • Marta Ligero
  • Christina Zatse
    Radiomics Group, Vall d'Hebron Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • David Leiva
    Bellvitge University Hospital, 08907 Barcelona, Spain.
  • Luz M Atlagich
    Radiomics Group, Vall d'Hebron Institute of Oncology (VHIO), 08035 Barcelona, Spain; Oncocentro Apys, Viña Del Mar 2520598, Chile.
  • Nikolaos Staikoglou
    Radiomics Group, Vall d'Hebron Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • Cristina Viaplana
    Oncology Data Science (ODysSey) Group, Vall d'Hebron Institute of Oncology (VHIO), Barcelona, Spain.
  • Camilo Monreal
    Radiomics Group, Vall d'Hebron Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • Joaquin Mateo
    Vall d'Hebron Institute of Oncology (VHIO) and Vall d'Hebron University Hospital, Barcelona, Spain.
  • Jorge Hernando
    Department of Medical Oncology, Vall d'Hebron University Hospital and Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • Alejandro García-Álvarez
    Department of Medical Oncology, Vall d'Hebron University Hospital and Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • Francesc Salvà
    Department of Medical Oncology, Vall d'Hebron University Hospital and Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • Jaume Capdevila
    Medical Oncology Department, Hospital Universitario Vall d'Hebron, Autonomous University of Barcelona, Barcelona, Spain.
  • Elena Elez
    Department of Medical Oncology, Vall d'Hebron University Hospital and Institute of Oncology (VHIO), 08035 Barcelona, Spain.
  • Rodrigo Dienstmann
    Oncology Data Science Group, Vall d'Hebron Institute of Oncology, Barcelona, Spain.
  • Elena Garralda
    Department of Medical Oncology, Vall d'Hebron University Hospital and Institute of Oncology (VHIO), Barcelona, Spain.
  • Raquel Perez-Lopez
    Radiomics Group, Vall d'Hebron Institute of Oncology (VHIO), Barcelona, Spain.