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
Sign In
Advertise with Us

Download Mobile App

AI Tool for Automatic Colorectal Cancer Tissue Analysis Outperforms Prior Methods

By LabMedica International staff writers
Posted on 26 Oct 2023
Print article
Image: A new AI tool outperforms prior methods in colorectal cancer analysis (Photo courtesy of 123RF)
Image: A new AI tool outperforms prior methods in colorectal cancer analysis (Photo courtesy of 123RF)

Colorectal Cancer (CRC) ranks as the third most prevalent and the second most lethal cancer. Catching it early and treating it promptly are extremely important. While machine vision technologies have seen remarkable improvements in automatically classifying types of cancer, they largely rely on deep neural networks with millions of parameters fine-tuned for diagnostic and prognostic tasks. Even though deep learning has shown extraordinary capabilities, healthcare professionals still have to inspect biopsied tissue samples to verify the diagnosis and assess the stage of the tumor. To advance this field further, scientists have now introduced an artificial intelligence (AI) solution specifically designed for automated analysis of colorectal cancer tissue that outperforms previous techniques.

The refined neural network developed by researchers from the University of Jyväskylä (Jyväskylä, Finland) has set new performance benchmarks in colorectal cancer tissue analysis. The AI-based system offers a more accurate and quicker way to categorize tissue samples of colorectal cancer from microscope slides. This advancement could significantly ease the work burden on histopathologists, thus enabling faster and more precise prognoses and diagnoses. Despite the promising results, it is important to be cautious while incorporating AI into medical practice.

As AI technologies move closer to becoming a standard part of clinical procedures, it becomes increasingly vital that they go through rigorous clinical validation. This is to ensure that the results they produce are consistently in line with established clinical norms. In a move encouraging collaborative development, the researchers are making this trained neural network publicly available. Their aim is to accelerate progress in the field by allowing scientists, researchers, and developers from around the world to further refine the tool and explore its various potential applications.

“By granting universal access, the aim is to fast-track breakthroughs in colorectal cancer research,” said Fabi Prezja, who was responsible for the design of the method.

Related Links:
University of Jyväskylä 

Platinum Member
COVID-19 Rapid Test
OSOM COVID-19 Antigen Rapid Test
Magnetic Bead Separation Modules
Anti-Cyclic Citrullinated Peptide Test
GPP-100 Anti-CCP Kit
Gold Member
Xylazine Immunoassay Test
Xylazine ELISA

Print article


Clinical Chemistry

view channel
Image: The 3D printed miniature ionizer is a key component of a mass spectrometer (Photo courtesy of MIT)

3D Printed Point-Of-Care Mass Spectrometer Outperforms State-Of-The-Art Models

Mass spectrometry is a precise technique for identifying the chemical components of a sample and has significant potential for monitoring chronic illness health states, such as measuring hormone levels... Read more


view channel
Image: The CAPILLARYS 3 DBS devices have received U.S. FDA 510(k) clearance (Photo courtesy of Sebia)

Next Generation Instrument Screens for Hemoglobin Disorders in Newborns

Hemoglobinopathies, the most widespread inherited conditions globally, affect about 7% of the population as carriers, with 2.7% of newborns being born with these conditions. The spectrum of clinical manifestations... Read more


view channel
Image: A false color scanning election micrograph of lung cancer cells grown in culture (Photo courtesy of Anne Weston)

AI Tool Precisely Matches Cancer Drugs to Patients Using Information from Each Tumor Cell

Current strategies for matching cancer patients with specific treatments often depend on bulk sequencing of tumor DNA and RNA, which provides an average profile from all cells within a tumor sample.... Read more


view channel
Image: Microscope image showing human colorectal cancer tumor with Fusobacterium nucleatum stained in a red-purple color (Photo courtesy of Fred Hutch Cancer Center)

Mouth Bacteria Test Could Predict Colon Cancer Progression

Colon cancer, a relatively common but challenging disease to diagnose, requires confirmation through a colonoscopy or surgery. Recently, there has been a worrying increase in colon cancer rates among younger... Read more
Copyright © 2000-2024 Globetech Media. All rights reserved.