As neuroscientist V.S. Ramachandran says, “All good science emerges from an imaginative conception of what might be true.”
Health care and the life sciences are currently entering a wave of innovation thanks to disruptive, computer-based technologies. Paradoxically, the evolution of machine learning, which raises the threshold of intelligent analysis beyond that of the human brain, can teach us more about what it means to be human.
Traditionally slow to adopt new technology, medicine often lags behind other disciplines when it comes to systematic change. As an example, the first paper on penicillin was published in 1929, but it took 14 years before production was scaled in 1943, and only then did the U.S. military decide to fund it to coincide with the D-Day invasion of Europe. A more recent example is laparoscopic surgery, which was initially invented in 1901 but only achieved widespread use in the ’90s.
When change does come to medicine, it is often abrupt. There is even an argument that medicine acts like a complex system, resistant to large-scale inputs yet susceptible to the “butterfly effect” — punctate quantum shifts where small-scale inputs can have dramatic effects. The recent history of medicine includes long periods of incremental innovation (i.e., endeavors like mapping the human genome), co-occurring with episodic bursts of radical innovation (i.e., the development of monoclonal antibodies), followed by a return to linear adjustments as these shifts are equilibrated. For example, in 1907, the average human lifespan was roughly 45 years; by 2007, it had risen to about 75, and this is largely due to significant changes in infant mortality through a few key systemic innovations like sanitized water systems and vaccinations.
Today, every industry is affected by disruptive technologies, but it’s medicine that’s ripe for radical change. The 2009 HITECH Act mandated the widespread adoption of electronic medical records in the U.S. By legislating medical treatment into bits and bytes, we are poised for radical innovation facilitated by disruptive software — and particularly software capable of making use of large, complex datasets.
We believe that this shift will occur because of five key technologies: AI, big data, blockchain, robotics and 3-D printing.
Artificial intelligence software used to be simply optimization programs (knowledge engineering) or statistical learning software (machine learning). While some of these programs have achieved routine use as “clinical decision support,” they are still not validated to replace human assessment.
Now we are seeing hypothesis generation AI programs (contextual normalisation). These programs represent a disruption because they have the capacity to increase the intelligence threshold, parsing loose connections to find meaning in previously unrecognized associations.
By Gunjan Bhardwaj | Forbes
Image Credit: Gunjan Bhardwaj/Forbes