Google DeepMind division is to turn its artificial intelligence expertise to the task of analysing CT and MRI scans from cancer patients in a bid to devise an algorithm that can instantly distinguish cancerous from healthy tissue.
An initiative with University College London Hospital NHS Foundation Trust will use DeepMind to study anonymised scans from 700 former cancer patients to see whether it can develop a quick and reliable way to diagnose cancers of the head and neck. If successful, the company could turn its attention to other ailments.
It is also hoped that it will be possible to cut the amount of time it takes to design targeted radiotherapy treatments from four hours to one.
Google acquired London-based AI company DeepMind in January 2014 for $400m.
"At present, it can take clinicians up to four hours to identify and differentiate between cancerous and healthy tissues on CT and MRI scans of head and neck cancer patients," said the UCLH Trust in a news post.
"This process, known as segmentation, is particularly difficult in head and neck cancer patients because the tumours are situated in extremely close proximity to healthy structures such as the eyes and nerves.
"The purpose of the research collaboration between UCLH and DeepMind is to develop artificial intelligence technology to assist clinicians in the segmentation process so that it can be done more rapidly but just as accurately."
Dr Yen-Ching Chang, clinical lead for radiotherapy at UCLH, described it as "very exciting research" that could revolutionise the way in which UCLH plans radiotherapy treatment.
"Developing machine learning that can automatically differentiate between cancerous and healthy tissue on radiotherapy scans will assist clinicians in planning radiotherapy treatment," she said.
"This has the potential to free up clinicians to spend more time on patient care, education and research, all of which would be to the benefit of our patients and the populations we serve."
The project follows an agreement between Google DeepMind and the Royal Free NHS Trust, which gave the company access to some 1.6 million NHS patient records.
The deal was met with circumspection owing to the wide-ranging nature of the access to the records, the promises of anonymisation, which is rarely as anonymous as proponents claim, and the connection with Google.
The Trust subsequently offered an opt-out to patients in a bid to quell the disquiet.
Trump proposes a $1.3bn fine and a round of firings to un-bork ZTE
Findings could mean new optical frequencies to transmit more data along optical cables
Findings made by reconstructing its orbit by numerical simulation
3D printer was specially adapted to build therapeutic biomaterials from multiple materials