Advertisement
UK markets close in 6 hours 55 minutes
  • FTSE 100

    8,093.74
    +53.36 (+0.66%)
     
  • FTSE 250

    19,689.69
    -29.68 (-0.15%)
     
  • AIM

    754.51
    -0.18 (-0.02%)
     
  • GBP/EUR

    1.1669
    +0.0025 (+0.21%)
     
  • GBP/USD

    1.2519
    +0.0057 (+0.45%)
     
  • Bitcoin GBP

    51,101.86
    -2,225.37 (-4.17%)
     
  • CMC Crypto 200

    1,365.06
    -17.51 (-1.27%)
     
  • S&P 500

    5,071.63
    +1.08 (+0.02%)
     
  • DOW

    38,460.92
    -42.77 (-0.11%)
     
  • CRUDE OIL

    83.23
    +0.42 (+0.51%)
     
  • GOLD FUTURES

    2,337.70
    -0.70 (-0.03%)
     
  • NIKKEI 225

    37,628.48
    -831.60 (-2.16%)
     
  • HANG SENG

    17,284.54
    +83.27 (+0.48%)
     
  • DAX

    18,017.59
    -71.11 (-0.39%)
     
  • CAC 40

    8,065.62
    -26.24 (-0.32%)
     

Google's AI can spot breast cancer better than humans

Left: sample view of a slide tissue with the dark zone on the left being an air bubble, the white streaks are cutting artifacts, the red hue across some regions are blood, the tissue is decaying, and the processing quality was poor. Right: LYNA identifies the tumor region in the center (red), and correctly classifies the surrounding artifact-laden regions as non-tumor (blue). - GOOGLE

Google has developed an AI system that it claims is better than doctors at spotting late-stage breast cancer.

According to the firm, its artificial intelligence can detect secondary cancer cells in medical scans with 99pc accuracy. For comparison, pathologists are on average 81pc accurate at detecting these cells when under time constraints.

Secondary cancer cells are able to break away from the primary cancer and grow in another part of the body. Finding them is a time-intensive and difficult task for pathologists. 

The Google AI, called Lymph Node Assistant (LYNA), was taught to recognise the “characteristics” of tumours by studying scans from cancer patients

ADVERTISEMENT

In a trial of the software, doctors using LYNA were able to carry out inspections of cancer scans in half the time and with better accuracy.

AI has become increasingly common in all parts of medicine and healthcare.

Moorfields Eye Hospital and Google’s DeepMind are implementing an algorithm that enables computers to analyse high-resolution 3D scans of the back of the eye to detect more than 50 eye conditions. 

Great Ormond Street Hospital (GOSH) has also paired up with Cambridge-based technology giant Arm to develop smart “person recognition” cameras that can spot intruders and monitor staff on its premises, in a move that could speed up operations and improve security.   

Google has developed software that is better at spotting late-stage breast cancer than human doctors - Credit: PA
Google has developed software that is better at spotting late-stage breast cancer than human doctors Credit: PA

The technology will be used by the children’s hospital in London to track doctors and nurses on site, while also keeping track of visitors as they pass from the reception area to wards.