Oshiorenoya Agabi, a Silicon Valley-based neurotechnology entrepreneur, has built an Artificial Intelligence (AI) device capable of sniffing out explosives and detecting cancer cells.
Agabi, unveiled the modem-sized device, known as Koniku Kore, at TEDGlobal in Arusha, Tanzania, which brought together 700 researchers, innovators, and academics from all over the world to discuss Africa’s next big leap in technology, science, and politics.
Agabi has established Koniku, a startup through which he hopes to produce and market Koniku Kore on a large scale. At present, devices used to detect explosives at airports and other security zones are made from silicon. Tech experts say if Agabi’s device is produced on a large scale, it could end long security lines at airports and other security checkpoints in future.
Agabi’s inventions is based on live mice neurons. Instead of creating a neural network by simulating the human brain’s interconnections of neurons, he took lab-grown neurons and fused them with electronic circuitry.
“Simulating the processing power of a collection of just 204 neurons would require a supercomputer. [So he thought] Instead of copying a neuron, why not just take the biological cell itself and use it as it is? That thought is radical. The consequence of this is mind-boggling.”
Koniku Kore detects volatile chemicals and explosives, the first device of its kind in the world. Silicon chips are very good at handling complex mathematical equations, but training them to recognise smells would take huge amounts of computational power. Agabi’s lab-grown neurons are able to take instructions to provide an olfactory receptor, while the electronics can then process the incoming data.
The neurons inside the device can currently be kept alive for two months, which means mass producing them is still not a feasible option. However, Agabi expects to mass produce them if the capital base of his startup is increased. The company’s current $8 million revenue expected to increase to $30 million in 2018.
“The world should deploy as many resources to develop devices that would solve global problems like terrorism or cancer. Koniku is already building an assembly line for combining biological machines with silicon devices and creating an entirely new class of devices and a new market. The eventual goal is to build a cognitive system based on living neurons within 5 to 7 years.”