Crunch Network

Formula 1 Technology Is Being Used To Make Better Surgeons

Next Story

I/O Spotlights Google’s Machine Learning Smarts

There was plenty of drama at the Monaco Grand Prix last Sunday with driver Nico Rosberg of Mercedes taking first place from teammate Lewis Hamilton late in the race. But there’s always plenty of action off the track too as race engineers analyze and transmit data to teams before and during any race.

McLaren, based in Woking, UK, is one of the biggest names in F1 but is now using their know-how in real-time data to help surgeons. Dr. Caroline Hargrove came to McLaren 18 years ago and is now Technical Director of McLaren Applied Technologies (or MAT), a subsidiary of the McLaren Technology Group.

Hargrove explains that 100s of sensors go into F1 race cars and stream torrents of live data to engineers far away in Woking, helping them make real-time decisions to optimize race strategy. But MAT is now applying this expertise to surgery – a profession where seconds also matter.

McLaren announced just last month an agreement with the University of Oxford to enhance key medical services by jointly developing analysis and decision support tools. The partnership draws on McLaren’s expertise in simulation technology, data management and predictive analytics. They’re starting with 50 surgeons, mostly ones in training but also more experienced surgeons. 

How F1 technology is working in the operating room

 A sensor is placed on a surgeon’s elbow while they operate. The data is sent via Bluetooth technology to computers. Hargrove explains that sensors can produce a stream of data that can be analyzed in real time for immediate feedback on a surgeon. She says, “We know there are certain traits that distinguish a great surgeon, such as speed and dexterity – how jerky or smooth is their movement when they cut. There’s always a subjective element in teaching any surgeon. This adds objectivity to it …in addition to another surgeon’s feedback.”

By comparing the results for different surgeons at all stages of their training – from beginners to highly experienced ones – it’s possible to predict their progression and gauge whether they’re hitting their development targets. She says this kind of analysis will not only save time on an operating table but in training a new surgeon. “Has the surgeon improved the last two to three times they did this procedure? Are they better at one thing more than another? Or, they don’t need to practice something 10 times more because they’re already doing it really well.”

“We know there are certain traits that distinguish a great surgeon, such as speed and dexterity – how jerky or smooth is their movement when they cut. There’s always a subjective element in teaching any surgeon. This adds objectivity to it …in addition to another surgeon’s feedback.”
— Dr. Caroline Hargrove

Hargrove also says by providing more feedback early on in a doctor’s career, the surgeons themselves know if indeed surgery is the best fit for their chosen specialty (or if radiology is because of a really keen eye).  “Surgery is perceived as the rock star of the medical field but it costs a lot of money to train surgeons.”

Hargrove says this partnered effort with Oxford is in beta testing for now, but eventually; McLaren may want to commercialize it. She points out the value of a potential database of quantitative (not qualitative) assessments from surgeries on 1,000s of patients.

A surgeon’s point of view

Dr. Freddie Hamdy is Head of Surgical Sciences at the University of Oxford and Director of Oncology and Surgery at Oxford University Hospitals NHS Trust. “Part of my job is to select the right people to become surgeons and to make sure they’re getting the right training.”

When asked why he chose to work with MAT, Dr. Hamdy said, “They have a lot of experience in expert simulation. In surgery, there’s a need to reproduce that, and a need to see what progress a surgeon is making so we can evaluate them, as well as an opportunity to validate an experienced one.”

How F1 technology helps patients:

In an F1 race, data is immediately analyzed to find the best strategy for the team, and used to run 1,000s of simulations per minute. For example, ‘If I pit now, what will happen?’ Or ‘If Ferrari drivers pit now, how does this affect me?’

As part of the McLaren/Oxford partnership, prescriptive data is also gathered, but from sensors placed on patients …during pre-op and post-op. Hargrove says data can be monitored from a patient at home days before a surgery to see how fit a person is for a procedure, or if they need more time to get into better physical condition.

“With data, we can spot a trend – why has this person’s weight gone up when their diet hasn’t changed. Or, we can monitor a person’s gait (how they walk). When a person’s gait changes, it’s usually due to a medical condition. But you don’t notice this visually.”
— Geoff McGrath

Hargrove points out that no sensor goes on the patient during surgery “as that aspect is already superbly done”. Anesthesiologists take in tons of data already by doing traditional patient monitoring of things like blood oxygenation and blood pressure.

Prescriptive Health Data

Chief Innovation Officer of MAT Geoff McGrath says sometimes the biggest challenge for data in healthcare is “people using prescriptive action based on algorithms across a population”. He adds, “Knowing that you have taken 10,000 steps a day isn’t really meaningful. Will it make a significant impact on your health? It doesn’t’ really tell you how hard you’re working, or what you need to do differently. People can start losing faith in the data.”

McGrath says like in F1 racing, a combination of mathematical data, simulation and modeling can spot a small anomaly in someone’s behavior …whether they’re a patient or a race car driver. “With data, we can spot a trend – why has this person’s weight gone up when their diet hasn’t changed. Or, we can monitor a person’s gait (how they walk). When a person’s gait changes, it’s usually due to a medical condition. But you don’t notice this visually,” McGrath says.

He points out an irony. “In intensive care of any hospital in the world, loads of alarms are going off all the time with loads of complexity there. But go into an F1 racing monitoring room, it’s very calm. But we’re processing way more data there …and presenting precise, detailed information that can then be acted upon out on the track.”

Featured Image: Mary Gorges