The project has the aim of speeding up neck and cancer radiotherapy treatments through using ‘thinking’ computers to examine patient scans. For this Google’s DeepMind Health are working with the U.K. National Health Service. Currently it takes up to four hours to analyze a scan of a patient’s head or neck in order to pinpoint where radiotherapy treatment should be targeted (using a process called ‘segmentation’). Google are hoping to speed up this process and to reduce the typical time-to-result down to one hour. Such analysis is important to ensure that radiotherapy is directed only towards cancerous tissue and that healthy tissues remain unaffected.
In a pilot study, Google’s computers will analyze scans, (from anonymous patients), taken from former cancer patients who were treated at the University College London Hospital. While computers will help with deciding where treatment is to be targeted, clinicians will have the final say.
Speaking with The Guardian, Dr Yen-Ching Chang, clinical lead for radiotherapy at University College London Hospital, said: “Developing machine learning which can automatically differentiate between cancerous and healthy tissue on radiotherapy scans will assist clinicians in planning radiotherapy treatment.”
The medic also added: “This has the potential to free up clinicians to spend even 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 study also has another purpose. According to Engadget, DeepMind will use the data captured from analyzing the samples to develop a radiotherapy segmentation algorithm. This commuter code could potentially be applied to examining for cancer in other parts of the body. As more data is gathered, the more sophisticated the program becomes and the more the computer will be able to ‘decide’ on where to scan.
