Google DeepMind is going to try and help the NHS treat head and neck cancer
Through a new partnership with University College London Hospital - the company's third with an NHS organisation - Google DeepMind will aim to establish whether machine learning techniques could reduce the amount of time it takes to plan radiotherapy treatment for such cancers
Treating cancers that are found in the head and neck with readiotherapy requires a great deal of precision. Clinicians must "segment" the tumour from healthy tissue and feed this information into a radiotherapy machine which then kills off the harmful cells.
Producing these detailed outlines can take clinicians up to four hours but Google DeepMind thinks that this time could be slashed to around an hour.
"Our collaboration will see us carefully analyse anonymised scans from up to seven hundred former patients at UCLH, to determine the potential for machine learning to make radiotherapy planning more efficient," Google DeepMind wrote on its website.
Head and neck cancers affect over 11,000 patients in the UK each year, according to Google DeepMind. The company pledged to treat patient data with the "utmost care and respect." Google DeepMind has also worked with the NHS on an eyecare project and a kidney monitoring project.