Close this search box.

Study: AI can Detect some Malignancies more Accurately than Biopsies


Research that experts believe could save thousands of lives shows that artificial intelligence is nearly twice as accurate as a biopsy at determining the aggressiveness of some tumors.

Ten million people worldwide lose their lives to cancer each year, according to the World Health Organization. If the illness is identified and treated early on, it can be prevented for millions more sufferers. Identifying and treating patients with high-risk tumors promptly is a major task for medical professionals.

An AI algorithm was found to be far more accurate than a biopsy in accurately grading the aggressiveness of sarcomas, a rare form of cancer that develops in the body’s connective tissues, such as fat, muscle, and nerves. The study was conducted by the Institute of Cancer Research (ICR) and the Royal Marsden NHS Foundation Trust.

Scientists anticipate that AI will help patients by providing physicians with a more precise method of classifying tumors. The new technique may facilitate the faster identification and rapid treatment of high-risk individuals, since high-grade tumors may be an indicator of severe disease.

Patients at low risk may avoid needless medical procedures, follow-up scans, and hospital stays. Thousands more patients might benefit if the algorithm is used in the future to treat various forms of the disease, according to researchers. The Lancet Oncology journal reported their findings.

The group focused on retroperitoneal sarcoma, a condition that arises at the back of the abdomen and is challenging to identify and manage because of its location.

In 82% of cases, the technology correctly classified the tumor’s expected aggressiveness, whereas 44% of biopsies yielded appropriate results. In 84% of the sarcomas examined, AI was also able to distinguish between leiomyosarcoma and liposarcoma, but radiologists were unable to do so in 35% of the cases.

Read More:



Copyright 2023 © Insightscare Magazine ( a Digital Ink brand ) All rights reserved.