Infervision is working on ground-breaking work to diagnose and treat strokes with the help of machine learning algorithms. The AI medical image specialists has already completed successful pilots of its Head CT Augmented Screening platform. It is hoped that the technology will soon go into widespread use and save lives, by allowing doctors to more quickly and accurately diagnose strokes and assess the damage they have caused.
It is the second medical technology based around machine learning which Infervision have reported success with – I previously wrote about their platform which detects early signs of lung cancer in X-ray and CT scans.
Over 100,000 annotated medical image scans were used to train the algorithms, which given more live data will become increasingly efficient at diagnosing the two main types of stroke, hemorrhagic and ischemic.
Infervision founder and CEO Chen Kuan told me “X-ray is a very old type of medical check-up – in China, for example, no one had mentioned chest X-ray in academic conferences for more than 15 years. Until very recently with the arrival of AI. AI has helped radiologists discover problems they previously weren’t able to see. So we are very proud to see radiologists starting to discuss some very interesting and fantastic cases involving AI.”
It’s certainly a fantastic example of the ways new technology can unlock value from data which has been around for a long time.
One of the major problems it solves is how to measure the volume of blood lost in hemorrhagic (bleeding) strokes. When every second is critical following a stroke, doctors generally use a simple mathematical formula to “guesstimate” as best as possible how much blood is lost.
Research shows the more accurately this volume is assessed, the more likelihood a patient has of recovery, due to how it affects treatment.
“Haemorrhage volume is strongly associated with mortality and the best way to intervene”, explains Kuan.
“Volumes over 30ml are strongly associated with mortality and its better to use aggressive surgical methods to intervene. The problem is, during our testing phase we asked radiologists to conduct these calculations and we found that in some cases the margin of error was more than 30ml.”
Not only is it hoped that the algorithms will “learn” to become more accurate than human radiologists at these assessments, they will be able to carry them out far more quickly in reaction to an emergency.
Another advantage is that diagnoses can be made from X-ray and CT scans, rather than MRI scans alone, which are currently the only way to diagnose ischemic (blood clot) strokes. MRI machines are less available, and many hospitals do not have the resources to run them 24-hours a day.
I asked Kuan how radiologists and other clinical staff had reacted when faced with technology which on the face of it seemed aimed at making some of their skills redundant.
“They are very excited”, he told me – “Two or three weeks ago there was a congress of Chinese radiologists and there was a lot of excitement about what we can do. They realise that we are helping them with the diagnosis but also helping with treatment plans for patients too.”
In fact, the results of Infervision’s trial in China will also be announced this week at the Radiological Society of North America annual conference in Chicago where Kuan is hoping for an equally enthusiastic response. He also hopes that far more people will have the opportunity to benefit from the technology soon.
“We’ve expanded it to four hospitals in China at this point and the initial results are promising, so soon we will be expanding to more hospitals and hopefully into the US as well.”
Bernard Marr is a bestselling author, keynote speaker, and advisor to companies and governments. He has worked with and advised many of the world's best-known organisations. LinkedIn has recently ranked Bernard as one of the top 10 Business Influencers in the world (in fact, No 5 - just behind Bill Gates and Richard Branson). He writes on the topics of intelligent business performance for various publications including Forbes, HuffPost, and LinkedIn Pulse. His blogs and SlideShare presentation have millions of readers.