Vision intelligence or AI or Artificial Intelligence or Machine Intelligence: Cancer is a very difficult disease to treat. With more than a hundred known species, each of which responds differently to treatment depending on the carrier and the part of the body present (and dozens of other factors), oncologists certainly have their work cut out for them.
Technologically speaking, Artificial Intelligence could soon make their work a little easier.
The IBM Watson supercomputer has a mechanical learning application for personalized cancer treatment for some time.
It examines over 600.000 medical reports and 1.500.000 anonymous patient envelopes and clinical trials, and the data it analyzes are designed to help scientists involved in the treatment. Currently, doctors rely on books, medical journals and clinical research to treat incidents.
However, research from a post-medical school could promote test results in Artificial Intelligence to design a better and more effective treatment.
And IBM is not alone.
Google's DeepMind is learning how to best apply radiation therapy to cancer patients. It examines how the exposure of patients to dangerous doses radiations, it can stop tumors by limiting the damage they can cause to healthy parts of the body. Doctors today use a combination of past experience and current research to try to determine the best way to expose a body to radiation as part of a treatment.
This process, known as partitioning, requires a doctor to know precisely the infected areas, which can not be dealt with only by a three-dimensional scan of the patient's volume. Adding complexity, head, neck, brain or spine, doctors have to make tough decisions on how to apply optimal radiation therapy without destroying critical areas.
Google DeepMind Artificial intelligence
DeepMind works with researchers at University College Hospital in London to develop artificial intelligence systems that could automate major parts of this process.
According to DeepMind:
Clinicians will remain responsible for decision-making on radiotherapy treatment plans, but there is the hope that the fragmentation process could be reduced from four hours to about an hour.
To transform this process, DeepMind will analyze 700 anonymized scans from former patients who had or are suffering from head and neck cancers. The hope is to develop one algorithm which can identify the best target areas from the scans automatically.
Over time, the team hopes that this approach can be applied to the treatment of cancers and other parts of the body.