New Technology Helps Detect Early Skin Cancer Stage

Skin cancer is the most common cancer in the UK, and over the past three decades, more people have had skin cancer than all other cancers combined. Now, a new technology may be able to detect early stages of skin cancer.

Melanoma Can Be Detected Using Images Of Lesions

Researchers from the Rockefeller University in the US have developed an automated technology which combines digital analysis and machine learning in order to help detect melanoma at an early stage, MedicalXpress said. The technology processes images of lesions using a series of computer programs which extracts information about the number of colors present in a growth and other quantitative data. The overall risk score generated called a Q-score indicates the likelihood of cancer growth.

A study to evaluate the tool's usefulness shows that the Q-score yields 98 percent sensitivity, which means that its ability to correctly identify early melanoma's on skin is very high, Economic Times reported. First author of the study, Daniel Gareau, said: "The success of the Q-score in predicting melanoma is a marked improvement over competing technologies."

The New Technology Could 'Save Lives And Avoid Unnecessary Biopsies'

Previous studies have shown that the number of colors in a lesion is the most significant biomarker to determine if a growth is malignant. Also, some biomarkers were only significant if looked at in a specific color channel. This finding could potentially be exploited to be able to identify additional biomarkers and improve accuracy.

"I think this technology could help detect the disease earlier, which could save lives, and avoid unnecessary biopsies too," Gareau said. "Our next steps are to evaluate this method in larger studies, and take a closer look at how we can use specific color wavelengths to reveal aspects of the lesions that may be invisible to the human eye, but could still be useful in diagnosis."

© 2021 iTech Post All rights reserved. Do not reproduce without permission.

More from iTechPost