Artificial Intelligence is now better at diagnosing cancer than real doctors.
According to research published in The Guardian, AI systems are now faster and more accurate at detecting a variety of cancers than clinicians.
An AI algorithm developed by German, French and US programmers beat dermatologists in accurately identifying cancerous moles over benign skin mutations. The system got 95 per cent of its diagnoses correct, while a team of 58 dermatologists were only 87 per cent correct.
Chinese software is able to diagnose prostate cancer as accurately as a pathologist.
Research leader Hongqian Guo of Nanjing University said: "[This] will help pathologists make better, faster diagnoses, as well as eliminating the day-to-day variation in judgment that can creep into human evaluations."
A system developed by researches at the University of Texas, Houston is able to accurately contour the shape of head and neck tumours, while trained physicians are shown to make widely varying estimates of tumour volume.
This new software means radiation oncologists can target treatment more accurately.
Software developed in Yokohama, Japan, is able to detect colorectal cancer in its earliest stages with 86 per cent accuracy.
This type of cancer is has previously proven difficult to locate before tumours become malignant.
As cancerous cells enter the bloodstream, early detection is critical.
And researchers at Imperial College London are working to develop AI-based techniques to improve the accuracy of breast cancer screening and reduce overdiagnosis.