CAN AI IMPROVE ORGAN TRANSPORTATION?
In medical emergencies, every moment counts. That’s why innovations that can shave even fractions of a second off the time it takes someone to get life-saving treatment represents a massive leap forward.
When it comes to organ transplantation, speed really is of the essence. According to the American Transplant Foundation, an average of 16 people die each day because they do not receive organs in time. What’s more, between 2014 and 2019 there were 370 “near misses” where transportation problems caused delays of two hours or more.
The logistical challenges faced in transporting organs are usually down to a combination of geography and biology.
Speaking to Forbes, Patti Niles of the Southwest Transplant Foundation said:
“The distance that organs can travel before they are transplanted varies by organ, based on how long they can stay outside of the body and successfully be transplanted. Either in a temperature-controlled environment or on a perfusion pump that artificially simulates body functioning for the organ, that time can vary from several hours to up to two days.”
When precious organs are couriered by road, rail or air, everything has to be planned down to the finest detail to ensure a timely delivery. But as we all know too well, transport doesn’t always run on time. And the stakes here are a lot higher than the inconvenience of an extra hour or so browsing duty-frees while we wait to embark on our summer holiday.
So, what can be done? Transplant providers have been looking for tech innovations to increase the amount of time organs can spend outside the body while they are being transported. These include devices that better regulate temperature or that artificially pump the organ during transit.
The other part of the solution is to find a way which ensures no organ is ever held up in traffic ever again. While Google Maps has helped me avoid many hours of tailbacks over the years with its reliable live updates and recommended detours, this clearly calls for something much more ambitious.
In a landmark trial, 2019 saw the world’s first successful delivery of an organ by drone. The small aircraft was supervised by remote teams at the University of Maryland every step of the way. It ushered in a new era in medical technology, where it is now possible to ensure the quick and reliable delivery of organs, in a safe, secure and medically-sound way, even in highly congested urban areas.
The next big step is to make sure that unmanned aerial vehicles (UAVs) are able to act as efficiently as possible, with AI being brought on board to calculate optimum routes and ensure not a moment is lost.
IT'S A BEAUTIFUL DAY TO SAVE LIVES
How does it work? Barry Alexander, founder and CEO of Aquiline Drones Corporation, the company responsible for this first successful mission, spoke to Forbes about the integral role AI has to play in this high-stakes sector:
"Automation of a complex, ever-changing logistical footprint is facilitated by the use of a deep learning AI. In 2019, we began creating a life-saving medical delivery service for human organs and tissues using drones. Today, we are augmenting these life-critical drone deliveries with a cognitive digital agent called Spartacus to [..] help solve the transportation logistics conundrum that's responsible for the yearly 25% loss in viable organs."
Essentially, manned aircraft and UAVs will team up with some element of transport on the ground, and all will be directed by Spartacus - a revolutionary AI being trained to consume vast quantities of real-time information to calculate the single best route for the organ to take.
"Spartacus is tied into real-time data sources including the FAA, weather, traffic and other related data sources to leverage its learning capabilities for optimised transportation options," said Alexander.
"From those datasets, it selects the best way forward using customer specifics, such as time of arrival, geo-location and cost. Once properly trained, Spartacus will achieve high rates of mission completion success for each organ transportation and delivery."
The beauty of AI, of course, is that it gets smarter over time. The more journeys Spartacus directs, the better it will become, as data builds up in its historical and temporal layers. And it’s already pretty good.
AI is so critical to this process because of the sheer number of options available.
Ryan Rusnak, CTO of Airspace, explains in the Forbes article that there are more than 180 millions different ways to go just from San Francisco to Boston. He adds: “In addition to choosing the best route, factors like traffic, weather, the driver's knowledge of the route and real-time incidents also need to be considered as events unfold, making it a highly complex and ever-evolving equation.
"With more than 180 million transport options, it’s easy to see how complicated it is without AI - it takes a computer more than four hours just to count to a trillion, which is why traditional computer programmes just aren’t up to the task.
"Before AI and machine learning, about one in three time-critical shipments arrived late.”
Spartacus looks for all of the fastest paths and uses a neural net to score them based on confidence level. Impressively, even with situations it has never encountered before, it can do this in less than a second. The type of organ being transported will also be factored into its decision-making:
"For instance, an organ requires a route that is as fast as possible with minimal risk, while cryogenic tissue requires minimal cost with no later than same-day delivery. The first would prefer something like driving farther to a different airport to make an earlier flight while the second would minimize ground transportation while still making a same-day cutoff."
We already know that AI and logistics make great bedfellows. According to McKinsey, the successful implementation of AI has helped businesses improve logistics costs by 15%, inventory levels by 35%, and service levels by 65%.
Another piece of research by McKinsey estimates that logistics companies will generate $1.3 - $2 trillion per year for the next 20 years in economic value by adopting AI into their processes, including supply planning, demand forecasting, warehouse robots, self-driving vehicles and predictive maintenance.
It seems inevitable that the role of AI in healthcare is only going to grow, with its application going much wider than logistics. Putting life-and-death decision making in the hands of computers will pose some challenging ethical questions for us all to consider. One example of this is where AI is being used to look at matches and decide who does or doesn’t get an organ.
As Harvard Business Review puts it:
“Can AI and analytics be used in a way that improves operational efficiency without jeopardizing our ethical principles? The answer is “yes” - if moral objectives and constraints, now often treated as an afterthought, are considered from the outset when designing models.”
According to the authors, this means taking into account how diverse groups are categorised in terms of urgency and ensuring no patient is discriminated against based on factors like race, sex, age and geographical location.
We are still in the early days of discovering just what this technology can do for healthcare, but it’s certainly heading in the right direction – and avoiding delays and queues to boot!