HEALTH AND WELLBEING
PHOTO : SHUTTERSTOCK
of readmission , and mortality . These predictions are based on 47 variables ( for example , age , weight , fitness , type of surgery ) from Professor Khan ’ s records .
Professor Hopgood says knowing the likely outcomes for a patient , based on clinical factors and the database at the back end of the machine-learning algorithm , helps doctors provide better quality information and counselling to patients . It also facilitates the design of better patient care plans . And from a hospital ’ s perspective , knowing in advance a patient ’ s likely length of stay helps administrators make better use of finite resources .
Most of the treatment cost of bowel cancer is associated with hospital care post-surgery . The AI-generated information will not reduce a patient ’ s length of stay , but having this information allows hospitals to run more efficiently and , by extension , be better healthcare providers .
For those who are seriously ill , knowing a definitive prognosis may help them to make critical decisions about their care as they near the end of their lives . “ They may not actually want to have an extended stay in hospital if their focus is on enjoying whatever period of life they ’ ve got left ,” he says .
AI as an instrument
Bowel cancer is the third most common cancer worldwide , the second biggest killer when men and women are combined , and it is a growing problem . More than 42,000 new cases were diagnosed in the UK in 2020 , and it is estimated that there will be approximately 2,400,000 new cases globally in 2035 . The economic impact of the disease is significant at US $ 17.41 billion in the US alone in 2020 , plus US $ 4.2 billion in lost productivity due to colorectal cancer deaths .
“ For example , if you have a good indicator of what ’ s going to happen to a particular patient immediately after surgery , you can plan for exactly where they should be , which ward they should be in and how long to expect them there ,” says Professor Hopgood .
And with more certainty around their prognosis , patients have more confidence in the critical decisions about their care and their lives .
“ The better informed the patient can be , the better their mental wellbeing tends to be as they enter into surgery .”
And it is often the small day-to-day aspects that worry people when they do not know what is ahead of them , as pointed out by Professor Hopgood : “… even just arranging for someone to feed the cat can be an important consideration for some people .”
For Professor Hopgood – a chartered engineer and fellow of the British Computer Society , the Chartered Institute for IT , whose previous work has focused on AI for engineering and business applications such as rail freight logistics – collaborating with the healthcare sector has been highly rewarding . This , he says , is partly because clinicians themselves are so enthusiastic about the prospects .
“ People can be worried about various aspects of AI – for example , that it will take away their jobs – but it ’ s not seen like that in the health profession ,” he says . “ The clinicians that I have met see it as helping them to do their job better , and to improve patient outcomes . It brings efficiencies so they can concentrate on more challenging medical issues and leave machines to do a lot of the routine work .”
ISSUE 04 / 2022 17