A blockchain-based platform which bills itself as the “future of artificial intelligence (AI) in healthcare” is preparing to launch an alpha version of its infrastructure within weeks.
Skychain believes its diagnostics system – a “distributed open network” of artificial neural networks (or ANNs for short) – can identify conditions in patients and prescribe appropriate treatments in milliseconds, substantially reducing the risk of human error during the crucial diagnosis stage.
By June, the company says early participants including hospitals, medical AI developers and healthcare data providers will have access to a “fully built” ecosystem. It hopes to welcome dozens more participants in the second half of this year – and by the time a beta version is released in December, Skychain believes “more and more neural networks and ready-made datasets” will become available.
Skychain wants all of its plans to be “completely realized” by next summer, when it plans to officially launch – paving the way for it to become “a leader in the medical AI market.” Estimates by IBM indicate this industry could eventually be worth $200 bln, with the company setting the ambitious objective of dominating the sector with a market share of 70 percent.
The financial and human cost of medical errors
According to Skychain, medical errors occur on such a large scale that they have become the third most common cause of death in the US. One example can be found when doctors are analyzing X-ray images – with early cases of lung cancer misdiagnosed in two in three cases. The company claims “there are many hundreds of examples like this” – citing a study which suggests such mistakes cause 250,000 Americans to die prematurely every year.
Of course, this also carries an economic cost – with the Healthcare Financial Management Association estimating that the US economy lost $20.8 bln as a result of misdiagnosis in 2016 alone.
Skychain aspires to reduce the number of premature deaths from medical errors by a staggering 10 mln within a 10-year period – and last year, a prototype which pitted its AI technology against doctors was put to the test.
The results, as yet unverified, indicate that the technology was more effective at diagnosing skin, lung and cardiovascular diseases than some “highly experienced” medical professionals. As reported by Cointelegraph at the time, the company claimed it took just 0.1 seconds on average for its system to identify conditions – considerably faster than the 20 to 30 seconds that doctors needed to come to a conclusion.
Skychain told Cointelegraph: “In real life, medical professionals may get tired or be in low spirits; they may also lack the necessary experience, specialize in a different area, or be biased.
“Each of these factors can have a negative impact on the accuracy of a diagnosis. Moreover, a machine learning system can easily analyze a poorly structured medical history or wade through large amounts of data.”
ICO already concluded
Skychain’s ICO was held from Feb 26 to April 8 – with the company claiming it managed to sell 10.3 mln SKCH tokens despite challenging market conditions brought about by the slump in Bitcoin’s price. Its utility token (SKCH) has now been listed on the LIVECOIN cryptocurrency exchange.
In the months and years to come, the company is confident of solving the problems which is preventing the medical AI market from flourishing. It says Skychain’s infrastructure will make it easier for doctors to access the creations of renowned labs and neural network developers. Likewise, these developers will get the opportunity to securely obtain the data they need to train their networks “inexpensively and fast.”