Google+ Facebook Twitter Twitter

AI system can identify different cancer cells

Researchers in Japan have shown that an artificial intelligence (AI)-based system can identify different types of cancer cells simply by scanning microscopic images.

They claim it is able to achieve higher accuracy than humans and could play a role in the future of oncology.

The system is based on a convolutional neural network, a form of AI modelled on the human visual system. It was applied to distinguish cancer cells from mice and humans, as well as equivalent cells that had also been selected for resistance to radiation.

Hideshi Ishii, lead author of the study published in Cancer Research, said: “We first trained our system on 8,000 images of cells obtained from a phase-contrast microscope.

“We then tested its accuracy on another 2,000 images, to see whether it had learned the features that distinguish mouse cancer cells from human ones, and radioresistant cancer cells from radiosensitive ones.”

On a 2D plot of the findings, the clustering of cell types together showed that, after training, the system could correctly identify cells based on the microscopic images of them alone.

Related Articles


Efficient regulation

A more consistent and transparent regulatory system can only be for the good, writes Sarah May.  


Imaging with LEGO

Researchers have come up with an inexpensive, automated way to image biological samples – using the children’s toy, LEGO.


My Lab: dynamic histopathology service

Trainee Specialist Biomedical Scientist Megan Clarkson gives a guided tour of her laboratory at Salford Royal NHS Foundation Trust.


Equality of opportunity

There is no place for outdated prejudice if there is a serious desire for change.