AI may be able to predict spread of melanoma, Israeli scientist says
search

AI may be able to predict spread of melanoma, Israeli scientist says

Assaf Zaritsky says video sequence can by analyzed to identify the appearance, behavior of cells; fears use as a functional model for melanoma may be clinically irrelevant

ILLUSTRATIVE -- In this photo taken Aug. 27, 2015, John Jay Hooker shows a melanoma spot that has shrunk since starting treatment at Tennessee Oncology in Nashville, Tennessee (AP Photo/David Goldman)
ILLUSTRATIVE -- In this photo taken Aug. 27, 2015, John Jay Hooker shows a melanoma spot that has shrunk since starting treatment at Tennessee Oncology in Nashville, Tennessee (AP Photo/David Goldman)

A research team headed by a scientist at the Ben-Gurion University of the Negev has developed a new method based on artificial intelligence aimed at identifying melanoma cells that are likely to metastasize.

The method, called “quantitative live cell histology,” was presented at the American Society for Cell Biology/EMBO conference in San Diego last December by Assaf Zaritsky of the Department of Software and Information Systems Engineering at Ben-Gurion University, and Gaudenz Danuser of the University of Texas Southwestern Medical Center (UTSW) at Dallas, who teamed up to work on the project.

According to a press release issued by Ben-Gurion University, the method involves filming live cancer cells with microscopic cameras and using artificial intelligence to analyze the video sequence in order to identify the appearance and behavioral patterns of the cells that associate with metastatic potential, meaning that they could spread to other parts of the body.

The method was first developed as part of Zaritsky’s post-doctoral studies at UTSW.

“Beyond metastasis potential, the computer models also allowed us to distinguish between cancer cells taken from different patients by quantifying factors that are not visible to the naked human eye,” Zaritsky said.

“In addition we found that different melanoma cell lines are much more similar to one another than to tumor cells taken from different patients that have not undergone prolonged culturing outside the human body.”

According to Zaritsky and his partner, the source of the phenomenon is natural selection due to the artificial process of transforming patient-derived tumor cells to cell lines.

However, the researchers fear that the use of the method as a functional model for melanoma may be “clinically irrelevant.”

Skin cancer is an abnormal growth of cells in the skin that mostly develops on areas of the skin that are exposed to sun rays. The disease affects people of all colors and races. If diagnosed and treated early, skin cancer is one of the easiest forms of cancer to cure. When allowed to progress, it can result in disfigurement or death.

read more:
less
comments
more