I guide startups to fight cancer and climate change with AI.
After spending two decades bouncing between academia and industry, I’ve come to realize that I’m most effective as a bridge between them. My superpower is distilling and adapting the latest research and getting it into products and services that can make a difference.
I’ve been doing computer vision and machine learning for 20 years. Before it was cool. Before everyone knew what deep learning is.
In the beginning I didn’t know what to do with this skill set. I just found it fascinating to write code to analyze images and extract new insights – automatically.
During my Masters degree at Carnegie Mellon University and an internship at the Mars Space Flight Facility, I explored the power of machine learning in detecting and categorizing rocks and craters on Mars. Soon I saw even greater opportunities here on Earth.
My PhD research at the University of North Carolina applied the recent advancements of deep learning to breast cancer. Predicting molecular properties of tumors that are too complex and abstract for the best trained pathologists. The traditional machine learning approaches we tried in the beginning failed, but deep learning changed the game.
And I’ve used it to transform other applications too. To detect tumor tissue missed during surgery. To predict patient outcomes or whether they are likely to respond to a particular treatment. To estimate emissions from power plants to track and publish the global contributors to climate change. Just to name a few.
But machine learning projects are complex and iterative with uncertain outcomes. Each success I’ve experienced came after a lot of experimentation, teaching me many lessons along the way.
I’ve learned that collaborating with domain experts is essential to the success of the project. Both in training an appropriate model and in understanding how it solves a greater challenge. Machine learning is just one piece of the puzzle. But it is a transformational piece.
I’ve devoted my career to creating that transformation. Accelerating computer vision and machine learning projects. Applying best practices for pathology and remote sensing images. Amplifying results with cutting-edge research. Advising machine learning practitioners from other domains. And reducing the uncertainty of these complex projects.
With my kids now growing up in a world with different challenges than I was raised in, I find it more important than ever to focus on the future. To apply what I’ve learned – and continue to learn – to create a better future.