Get the App
SLTechnology News&Howtos  ›  IT Information  › 

Microsoft uses AI to shorten radiotherapy time for cancer: scan speed is increased by 2.5 times, and accuracy is up to 90%.

Shulou Source: shulou.com Published: 2023-11-24 17:45:19 10月03日 Update

CTOnews.com, June 28 (Xinhua)-- Britain is planning to provide a new artificial intelligence technology to all NHS (National Health Service) trusts at cost, which can help doctors calculate the location of radiation beams faster, thereby effectively reducing the time it takes for patients to receive radiation, BBC reported. The AI training is carried out jointly by Microsoft and Adenbrook Hospital in the UK.

▲ image source Microsoft UK before starting radiation therapy, it usually takes doctors 25 minutes to two hours to scan about 100 body sections of patients and carefully outline the outline of bones and organs in order to guide the radiation beam to the right position. For example, when treating prostate cancer, doctors need to avoid damaging the nearby bladder or rectum, which could lead to lifelong incontinence.

Dr. Raj Jena of Adenbrook Hospital worked with Microsoft to train a program called InnerEye based on data from previous patients. After training, the AI can draw the outline details of the human body with 90% accuracy, while doctors will still check the results manually. In about 2/3 of cases, doctors can use the results for radiotherapy without any modification, which is 2.5 times faster than previous scans and rendering.

The NHS artificial Intelligence Lab has provided 500000 pounds (CTOnews.com Note: currently about 4.6 million yuan) to Adenbrook Hospital to finance the necessary safety checks and assessments. The British government has been investing in NHS's artificial intelligence project, and InnerEye is the first NHS-funded auxiliary AI program in the field of medical imaging.

Tags: Doctor UK patient Microsoft artificial Intelligence Hospital Intelligence Blu Arden Brooke training location Medical Technology Program results profile Inspection funding Apple Docker Huawei Linux macOS MariaDB Microsoft MySQL NVidia OPPO Reno Xiaomi Shulou Tech Info Huawei vpn Docker