We use cookies to understand how you use our site and to improve your experience. This includes personalizing content and advertising. To learn more, click here. By continuing to use our site, you accept our use of cookies. Cookie Policy.

Features Partner Sites Information LinkXpress hp
Sign In
Advertise with Us
GLOBETECH PUBLISHING LLC

Download Mobile App




CT-Based Deep Learning-Driven Tool to Enhance Liver Cancer Diagnosis

By MedImaging International staff writers
Posted on 08 May 2025
Print article
Image: SALSA is a new artificial intelligence tool for the automated and precise analysis of liver tumors (Photo courtesy of 123RF)
Image: SALSA is a new artificial intelligence tool for the automated and precise analysis of liver tumors (Photo courtesy of 123RF)

Medical imaging, such as computed tomography (CT) scans, plays a crucial role in oncology, offering essential data for cancer detection, treatment planning, and monitoring of response to therapies. However, one of the key challenges in using these images effectively is the accurate delineation of tumors, a task known as tumor contouring. This process, which is vital for volume analysis, is often time-consuming and prone to significant variability between different observers. This variability can create bottlenecks in both clinical settings and research projects, especially when dealing with volumetric disease assessments. Primary liver cancers, including hepatocellular carcinoma (HCC), are frequently diagnosed at later stages when treatment options are limited, and the prognosis is often poor. Additionally, the liver is a common site for metastasis from other primary cancers, which complicates treatment and affects overall outcomes. In response to these challenges, researchers have developed an innovative deep learning-based tool to improve the diagnosis, treatment planning, and monitoring of liver cancer.

The new tool, called SALSA (System for Automatic Liver tumor Segmentation And detection), was developed by researchers at the Vall d'Hebron Institute of Oncology (VHIO, Barcelona, Spain; www.vhio.net) to automate the detection and tracking of liver tumors. This AI-powered system works directly with CT images, automatically detecting and delineating both primary and metastatic liver tumors. To build and train SALSA, the researchers used the nnU-Net segmentation framework, incorporating data from 1598 CT scans that included 4908 liver tumors, both primary and metastatic.

SALSA achieved exceptional results, demonstrating a tumor detection precision of over 99% at the patient level, and nearly 82% lesion-by-lesion detection precision in an external validation cohort. The integration of artificial intelligence and machine learning in this tool signifies a breakthrough in personalized oncology, offering a more accurate method for assessing treatment response on an individual basis. By outperforming existing state-of-the-art models and improving the inter-observer agreement among radiologists, SALSA shows promise in enhancing liver cancer detection, optimizing treatment strategies, and improving response evaluation.

“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,” said Raquel Perez-Lopez, corresponding author of the paper published in Cell Reports Medicine. “Our validation across several test and external cohorts highlights SALSA’s effectiveness and reliability, matching, and often surpassing, the accuracy of expert radiologists.”

Related Links:

VHIO >>> www.vhio.net

Ultrasound Table
Women’s Ultrasound EA Table
New
Mobile X-Ray Machine
MARS 15 / 30
New
Digital Intelligent Ferromagnetic Detector
Digital Ferromagnetic Detector
Portable Color Doppler Ultrasound System
S5000

Print article

Channels

Radiography

view channel
Image: AI can identify “mammographically-visible” types of interval cancers earlier by flagging them at the time of screening (Photo courtesy of ScreenPoint Medical)

AI Improves Early Detection of Interval Breast Cancers

Interval breast cancers, which occur between routine screenings, are easier to treat when detected earlier. Early detection can reduce the need for aggressive treatments and improve the chances of better outcomes.... Read more

MRI

view channel
Image: An MRI scan can reveal a heart’s functional age (Photo courtesy of 123RF)

New MRI Technique Reveals True Heart Age to Prevent Attacks and Strokes

Heart disease remains one of the leading causes of death worldwide. Individuals with conditions such as diabetes or obesity often experience accelerated aging of their hearts, sometimes by decades.... Read more

Nuclear Medicine

view channel
Image: In vivo imaging of U-87 MG xenograft model with varying mass doses of 89Zr-labeled KLG-3 or isotype control (Photo courtesy of L Gajecki et al.; doi.org/10.2967/jnumed.124.268762)

Novel Radiolabeled Antibody Improves Diagnosis and Treatment of Solid Tumors

Interleukin-13 receptor α-2 (IL13Rα2) is a cell surface receptor commonly found in solid tumors such as glioblastoma, melanoma, and breast cancer. It is minimally expressed in normal tissues, making it... Read more

Imaging IT

view channel
Image: The new Medical Imaging Suite makes healthcare imaging data more accessible, interoperable and useful (Photo courtesy of Google Cloud)

New Google Cloud Medical Imaging Suite Makes Imaging Healthcare Data More Accessible

Medical imaging is a critical tool used to diagnose patients, and there are billions of medical images scanned globally each year. Imaging data accounts for about 90% of all healthcare data1 and, until... Read more
Copyright © 2000-2025 Globetech Media. All rights reserved.