Rewrite SALSA: A new artificial intelligence tool for the automated and precise analysis of liver tumors this news headline for the science magazine post

• Published in Cell Reports Medicine, results of a VHIO-led retrospective study demonstrate the potential of a novel, CT-based deep learning-driven tool to enhance liver cancer diagnosis, treatment planning, and response evaluation.

• Developed by VHIO investigators, the SALSA artificial intelligence (AI)-aided tool (SALSA) works directly on medical images, automatically detecting and delineating tumors.

Investigators of the Vall d’Hebron Institute of Oncology’s (VHIO) Radiomics Group, led by Raquel Perez-Lopez, have developed SALSA (System for Automatic Liver tumor Segmentation And detection), a fully automated deep learning-driven tool for the precise and completely automated detection and monitoring of liver tumors (hepatocellular carcinoma). Results of this work have been published as an open access article in Cell Reports Medicine.

Medical images including computed tomography (CT) scans, provide professionals in oncology with essential information for cancer detection, treatment planning, and response evaluation.

“However, the precise delineation of tumors for volume analysis—tumor contouring—poses practical challenges and often causes a bottleneck in research projects and clinical applications involving volumetric disease assessment. This task is not only time consuming but also prone to variability between and among different observers,” said Raquel Perez-Lopez, corresponding author of the paper.

Primary liver cancers such as hepatocellular carcinoma (HCC) are often diagnosed in advanced stages, with limited treatment options available and an unfavorable prognosis. Furthermore, the liver is a common organ for metastases originating from other primary cancer, significantly impacting treatment outcomes. Aimed at addressing these clinical challenges, VHIO investigators have developed SALSA, an AI-aided tool that works directly on medical images, automatically detecting and delineating liver tumors, both primary and metastatic.

“For the development and training of SALSA, we used the existing nnU-Net segmentation method and introduced data obtained from 1598 computed tomography scans of 4908 primary or metastatic liver tumors,” explained Maria Balaguer-Montero, a PhD Student of VHIO’s Radiomics Group and first author of the study.

Showing superior accuracy for cancer detection and precise quantification of tumor burden, surpassing state-of-the-art models and the radiologists’ inter-agreement, SALSA could potentially enhance cancer detection, treatment planning, and  response evaluation.

“SALSA demonstrated high accuracy tumor detection with precision of over 99% at the patient level, and a lesion-by-lesion detection precision of almost 82% in the external validation cohort,” added Balaguer-Montero.

“This novel deep learning-driven tool has shown precise and  automated identification and delineation of liver cancer on CT images, facilitating a more precise quantification of tumor burden —a crucial factor in cancer prognosis and treatment—with no prior manual prompt requirements. Our validation across several test and external cohorts highlights SALSA’s effectiveness and reliability, matching, and often surpassing, the accuracy of expert radiologists,” observed Perez-Lopez.

Imaging biomarker discovery for treatment response assessment

A key objective of VHIO’s Radiomics Groups is to develop medical imaging tools based on imaging biomarkers including tumor volume, density, and texture. Emerging technologies with artificial intelligence and machine learning techniques are rapidly advancing the field of personalized oncology by providing a more precise evaluation of treatment response in individual patients.

“Current clinical response criteria are limited. Tumor diameter is measured in 2D without considering volume and, in the case of metastases, only two tumors per organ and a maximum of five lesions per patient are evaluated,” said Raquel Pérez-López.

“SALSA could be useful for the management of patients with liver cancer, enabling the measurement of parameters such as total tumor volume, density, or texture to assess response to therapy and support therapeutic decision-making to enhance patient outcomes,” she concluded.

This work has been possible thanks to the support received from the CRIS Cancer Foundation, FERO Foundation, and the “la Caixa” Foundation, that provide funding for various research projects led by VHIO’s Radiomics Group.  

###

About VHIO

The Vall d’Hebron Institute of Oncology (VHIO), established in 2006 and located within the Vall d’Hebron Campus, is a reference comprehensive cancer center for personalized medicine in oncology. Through our purely translational and multidisciplinary research model, we aim to improve the prevention, early diagnosis and treatment of cancer by transforming the latest scientific discoveries made in the laboratory into early phase clinical trials for the development of more effective therapies to improve the quality of life and survival  of cancer patients.

VHIO forms part of the CERCA – Research Centres of Catalonia system and is accredited as a Severo Ochoa Center of Excellence.

Research at VHIO would not be possible without the support received from our patrons –Generalitat de Catalunya, Fundació Privada CELLEX, “La Caixa” Foundation, Fundación FERO, Fundación BBVA and the CRIS Cancer Foundation– and the public funding it receives as well as the generous support from institutional supporters, private institutions, companies, associations, societies, and individual donors. Only with such continued support will VHIO continue to advance personalized and targeted therapies against cancer.