Artificial Intelligence (AI) in cancer care
What is it and why is it used?
What is AI?
Artificial Intelligence (AI) is a type of computer software (the instructions that tell a computer how to work). Using AI, a computer does not have to be told the steps to solve a problem. Instead, the AI software looks at lots of previous examples of the same problem. It then works out the steps to solve the problem for itself. This allows computers to do lots of clever things they would not normally be able to do. For example, you can unlock lots of phones now with face recognition. When you do, the phone uses AI to work out if it is actually you.
Despite this, AI is not a robot human. It cannot think freely in the way we can. It is very different to how AI is presented in sci-fi films. AI software mostly just solves a narrow problem. Good AI software does this very well and very quickly. However, like humans, AI is not perfect, and a doctor is always required to check its work before doing anything else. An AI and a doctor working together can make a good team.
Why is AI used?
AI helps doctors with cancer care, including for womb cancer. It can help:
- Work out whether you are more likely to be at risk of womb cancer
- Identify different irregularities in the wall of the womb on hysteroscopy images (fibre optic camera test)
- Combine different scans into a more detailed 3D picture
- Highlight areas on your scan or hysteroscopy images that are more likely to be cancer
- Outline where internal organs are on your scan
- Check your medical history to make sure new treatments are safe for you
- Order medicine
- Book your appointments
- Control robots that doctors use as part of your surgery e.g. robotic-assisted hysterectomy
- Check whether your treatment is working
- Make researching new treatments quicker and easier
Photograph obtained from the Centre for Ageing Better
Concerns about using AI
Sometimes people worry about AI in healthcare. They worry it is looking at their information. They worry it might mean the doctor spends less time with them. They worry it could be biased. They worry robotic-assisted surgery might go wrong. It is normal to worry about these things. In response to these concerns:
- The NHS holds data securely and obeys data laws. You can opt out of your information being used for research. The NHS and its partners only use the data for planning health and care and finding new treatments. Your NHS data is never used for marketing or insurance purposes.
- AI can speed up repetitive tasks. Doctors can check what the AI has done rather than doing it all themselves. This means they are free to focus on other parts of your care. This might include spending more time face-to-face with you.
- Like a doctor, if AI is trained badly, it can be biased. We have seen this when AI is used in other parts of life. For example, some phone face recognition was not as good at recognising darker skin tones. In healthcare, AI must be trained very well. This is so it is safe and helpful for everyone. It is still sometimes hard to get a range of past examples to train the AI. However, this is something programmers are aware of and keep a close eye on. AI technologies are rigorously tested before they are introduced into healthcare.
- Robotic-assisted surgery can have better surgical outcomes for some patients. The robot cannot do anything unless the doctor instructs it. It can enhance precision, flexibility, and control during the operation.
Doctors might have used AI in your diagnosis or treatment. The NHS already uses it in many ways, and they are careful to do it safely. They want to use it more because it can help improve your care and advances in this area are happening quickly. Therefore, we hope you find understanding a little more about AI helpful.
Provided by Tom Iszatt, Patient Communications Editor at Mirada Medical.