The LiFE Project
Using artificial intelligence and big data to get ambulances to emergency cases before time runs out.
The Challenge
From the moment they receive the call, ambulance drivers and paramedics are expected to reach Category A callers in no more than 7 minutes. 1 They make every effort to ensure that we get the help we need, precisely when we need it. However, ambulance drivers encounter issues which prevent them from doing just that.
It’s not their fault. There can be many reasons why an ambulance service struggles to reach emergency patients. All too often, it’s the simple issues that have the greatest impact: congestion and road awareness. These are just some of the many obstacles that ambulance drivers have to overcome.
Every second counts. When people encounter these life-threatening conditions, their chances of survival are dependent on the time it takes for paramedics to arrive. In England, no matter where they live, paramedics are expected to reach emergency situations within the aforementioned time limit. Yet, roughly 1 in 3 ambulances do not reach these life-threatening calls in time.
Wanting to learn more about the challenges facing our ambulance services, we got in touch with North West Ambulance Services, alongside Liverpool’s traffic control centre, to try and find a solution. It was time to listen.
Listening
Listening to the staff within the North West Ambulance Service, we realised just how much they have to contend with on a daily basis. Drivers told us time and time again that most of the general public simply do not know how to help emergency vehicles in times of need. They panic, and these actions lead to accidents. Sometimes, these accidents involve the emergency services themselves. More than £300,000 of the NHS’s annual budget is spent repairing ambulances involved in collision.
Ambulance drivers told us that they felt they were “failing the public”. We knew this was far from the case. We realised just how much simple obstructions could have a major impact on the journey towards an emergency situation.
We spent time with the traffic controllers too, researching and learning about all of the existing emergency traffic management systems. After our meeting with Peter Jones, we realised was that all of these systems required manual input, such as green-wave buttons and predefined route green waves.
We knew that advances in artificial intelligence would be able to clear routes for ambulances, getting people the critical care they need in time. So we got to thinking.
Nobody in this world is more important than anyone else. LiFE is about saving lives, so it has to happen.
Thinking
Having listened closely to both sides of the issue, we started by examining the SCOOT system. Making use of sensors to adjust traffic light signals, SCOOT uses algorithms to get people to their destination in the quickest manner possible.
However, the algorithms used by SCOOT fail to take into account the needs of the emergency services. Having listened closely to the ambulance drivers, we understood just how emotionally charged these journeys are. We knew our solution needed to be seamless, clearing the way without needing additional ambulance driver input.
Our experience with GPS data, alongside our work on previous Big Data projects in the past, made us realise that we could create an algorithm to clear congestion ahead of the ambulance With a clear idea of a solution, we were now ready to enter development.
One of the good things that works in applying AI to healthcare is essentially the idea that you can make a difference to the lives of people.
I think it's innovative, it's far-reaching in that it collaborates with a lot of other agencies that need to be involved, it's a great starting point for improving the way that we respond.
Doing
Liverpool City Council and Innovate UK gave us the green light to work and refine this incredible, life-saving project.
Our team of data scientists worked hard to integrate both the GPS data we had collected over the years and historical ambulance data with the coexisting SCOOT system. Testing the LiFE project within a 200m range from Liverpool City Centre, LiFE was set into place and routes were actively calculated for ambulance drivers. Our new algorithm, using principles of artificial intelligence, put ambulance drivers first.
The data collected from these algorithms have given us ample information to predict future behaviours on certain emergency routes with a high degree of accuracy. The data we collected demonstrated that the LiFE project also benefits emergency vehicles in terms of fuel consumption, making them more cost-effective and easing their strain. Furthermore, LiFE was evidently an eco-friendly solution, reducing emissions produced by emergency services across the region.
The LiFE intelligent mobility algorithm can reduce ambulance journey times by up to 40% in urban transportation networks. Every second shaved off the journey time has the potential to save a life.
We are now expanding LiFE to cities across the UK, because traffic should not determine life or death.