Member Spotlight: Andrew Conkie, Founder & Data Scientist, Red Star
This week, we caught up with Andrew Conkie, Founder & Data Scientist at Red Star, a Glasgow based company delivering AI Solutions in Healthcare.
Hey Andrew, thank you so much for joining us! Firstly, can you tell us a bit about your background?
I'm a data scientist and founder at Red Star. I started my career working as a Technical Business Analyst in financial services but found that I was more and more gravitating towards the data analytics side of things.
A few years ago I decided I wanted to change direction and started Red Star as a company working with healthcare data. The goal is to use data analysis and AI approaches to solve real world problems in healthcare. There's a huge amount of data there and much of it is underutilised.
Outside of work, I play squash - badly - but it's the only thing keeping me fit just now!
What does using data analysis and AI approaches to health care look like in practice for you?
Red Star have been applying data analysis and AI to healthcare datasets for a while now.
In terms of specific use cases, we have used Natural Language Processing to read medical records and identify patients with heart failure then linked this up with other data such as prescribing and mortality to compare treatment and outcomes. We can see that patients on 3 different classes of drugs have better outcomes but need to better understand the reasons behind this. Our clinical partners are now contacting people on different combinations of drugs to see if they can optimise the treatment. The hope is to see measurable improvements in quality of life in just 3 months!
We are also working with the Fracture Liaison Service in Glasgow to provide data driven dashboards for people at risk of osteoporosis. This means they don't have to click across multiple systems to get the patient's details. Sometimes getting the right data, in the right format, to the right person is all that's needed.
How do you see AI and machine learning impacting the healthcare more broadly in the coming years?
There's an enormous amount of data in healthcare - 'cradle to grave' in the UK - but it's underutilised and far more than any individual can synthesise. In the future, AI algorithms should be able to understand a patient's entire clinical history and recognise the risks specific to them. This can sit in the background and understand what an abnormal blood test means to that particular patient or detect - at an early stage - when someone's health takes a turn for the worse or they might need further investigations. This should help to get them seen by the right specialist quickly which is often half the battle.
What is your top tip for aspiring data professionals?
Understand the business (or clinical) domain and that will be key to making sure your data science is truly impactful. The focus should be applied data science.
If you could start your career over, would you do anything differently?
I've had a varied career starting with a law degree then a contract Business Analyst in financial services and now data science in healthcare. I don't think I'd do anything differently!
What resource would you recommend to our community?
The Penguin blog by Eric Topol is always a good shout for keeping up to data with the latest uses of AI in healthcare.
The Data Lab values are Support, Respect, Innovate & Grow - which one resonates with you most at the moment and why?
Support resonates most for me. Working at Red Star I've had enormous support from the data community and try to pay it back.
What motivated you to join The Data Lab Community? Have you found it valuable so far?
I've been involved with the Data Lab for a few years now and have found it a great way to bring people together, raise the skills base and share knowledge. The Data Lab Community just puts a formal structure round what's been happening for a while.
How could you help other members of the community?
Always happy to share knowledge and learnings with the community. I've also enjoyed taking on Data Lab Masters students on a few occasions.
Is there any way that members of the community could help you?
Just continue to support the community and spread the word!
Thanks for the chat Andrew - keep up the great work!
If you'd like to be featured in one of our Member Spotlights just get in touch via [email protected] or drop our community manager Paul a DM.