ROBOTS v CANCER

AI spots disease in women

ARTIFICIAL intelligence able to identify women with high-risk cancers could transform survival rates.

A robot shown thousands of scans spotted aggressive “shape-shifter” tumours too complex for doctors to see.

It could lead to the patients immediately receiving more intensive life-saving treatments, such as chemotherapy.

The Institute of Cancer Research in London created a machine learning tool that looked at 150 million cells from 514 women with ovarian cancer.

It was able to detect clusters of tumour cells with misshapen nuclei, which have hidden weaknesses in their ability to repair DNA. It means they are better at hiding from immune defences.

Only 15% of women with the shapeshifter nuclei were still alive five years later, compared with 53% of other patients with the disease.

Yet such tumours are susceptible to drugs called PARP inhibitors or platinum chemotherapy, which target DNA

OSTEOPOROSIS could be reversed by a technique that grows bone, a study found.

Scientists switched off the brain receptors for brittle bone hormone oestrogen in

mice and found their bone mass increased eightfold.

Prof Holly Ingraham of University of California San Francisco said: “I’m in the clouds about this result.” repair defects. ICR chief Prof Paul Workman said: “By using AI to analyse routinely taken biopsy samples, it is possible to uncover visual clues that reveal how aggressive an ovarian tumour is.

“What makes this test even more exciting is its ability to pick out in a new way those women whose tumours have weaknesses in DNA repair, who might therefore respond to treatments that target these weaknesses.”

Some 7,000 UK women are diagnosed yearly with ovarian cancer. It has one of the worst cancer survival rates: 35% of patients live 10 years beyond diagnosis.