At the moment, most AI research is actually about getting good performance in complicated prediction tasks, with algorithms that have acceptable computational complexity. That doesn’t sound so scary!!
So where exactly is AI at now? One area where AI research is alive and well is in the medical field. With medical misdiagnosis and delayed diagnosis regularly in the news, Medical News(2016) and the disturbing reports coming out of Wexford hospital last week, Cullen (2016), the medical world, patients and doctors alike will greatly benefit from improved diagnostics.
For me, it’s not acceptable in modern medicine that patients are still regularly misdiagnosed, one in six according to an article in the Telegraph, Devlin, K and Smith, R (2009). I for one and I’m sure lots of others, more than welcome the introduction of AI into this field. Intelligent technologies will become invaluable in making health care more efficient. Combining the strengths of doctors and smart software will provide better care than either could alone.
I really like this idea of AI and the medical professionals working together to improve healthcare systems around the world. I hope that professionals in the field embrace the technology and see it as complimentary to their own work rather than a substitute, looking to replace them.
Some interesting and promising work already in progress in this field involves a MIT team which has “developed a computational model that aims to automatically suggest cancer diagnoses by learning from thousands of data points from past pathology reports”, Conner-Simons, A (2015) and a team at Memorial Sloan Kettering Cancer Centre in New York training an artificial intelligence to find similarities between cases that human doctors might miss by combing through doctors notes, Rutkin, A (2016).
Although these projects are still in the research stage it’s clear to see the potential that AI has to offer the medical world, reduced mortality rates, faster and more accurate diagnostics, less pressure on overworked healthcare staff and perhaps a decrease in medical costs!
Of course AI is already being used by doctors with electronic medical assistants available, Hernandez, D.(2014) to advice doctors on diagnosis and develop the right treatments for their patients.
I personally find the future use of AI in medicine very reassuring. Having diagnosis or a 2nd opinion done by someone who isn’t having a bad day, or tired, or overworked etc, will result in increased trust in our healthcare systems.
I recently read an article in the Economist which suggested that people were more likely to trust a diagnosis from a computer rather from doctor!
Medical Negligence Ireland (2016) Available at: http://www.irishmedicalnegligence.ie/news/ [Accessed: 20th February 2016]
Cullen, P. (2016) “‘Avoidable death’ among cases in Wexford cancer review” Irish Times 17th February 2016 Available at: http://www.irishtimes.com/news/health/avoidable-death-among-cases-in-wexford-cancer-review-1.2538607 [Accessed 20th February 2016]
Devlin, K and Smith, R (2009)” One in six NHS patients ‘misdiagnosed’ The Telegraph 21st September 2009 Available at:
http://www.telegraph.co.uk/news/health/news/6216559/One-in-six-NHS-patients-misdiagnosed.html [Accessed 20th February 2016]
Conner-Simons, A (2015) “How a computer can help your doctor better diagnose cancer” MIT News 23rd April 2015 Available at: http://news.mit.edu/2015/how-computer-can-help-your-doctor-better-diagnose-cancer-0423 [Accessed 20th February 2016]
Rutkin, A (2016) “AI reads doctors’ notes to find hidden links in cancer cases” New Scientist 18th February 2016 Available at: https://www.newscientist.com/article/2078159-ai-reads-doctors-notes-to-find-hidden-links-in-cancer-cases/ [Accessed 21st February 2016]
Hernandez, D.(2014) “Artificial Intelligence Is Now Telling Doctors How to Treat You” Wired 6th February 2014 Available at : http://www.wired.com/2014/06/ai-healthcare/ [Accessed 21st February 2016]