By Kyle Fogarty…
Revolutionizing Foot Geometry Capture with Machine Learning
My PhD research explores the intersection of geometry processing and machine learning, focusing on making 3D geometry more accessible for capture, editing, and analysis. Recently, I had the opportunity to apply this expertise during a Research Scientist internship at Hike Medical.
Hike Medical is an innovative start-up in the foot healthcare space, combining custom insoles and AI technology to create a more accessible, affordable solution for employers and their employees. By managing everything in-house – from foot scanning to insole printing – their vertical integration enables rapid innovation and precise quality control throughout the production process.
A key challenge we tackled was improving the consumer foot geometry capture pipeline. The existing video-based capture method often resulted in incomplete foot geometry, as customers either struggle to maintain stillness throughout the scanning process or do not have the mobility to capture the entire foot. This partial data posed significant challenges for downstream orthotic creation and foot analysis. To address this, we developed a generative approach for shape completion that could transform partial scans into fully detailed, accurate foot geometry.
The impact of our work was immediately tangible. We saw the first batch of insoles produced under our new pipeline, that would not have been possible otherwise without customers rescanning. The direct translation of research into a product that would enhance the user’s comfort and well-being, highlighted the real-world value of our work.
This experience showed the practical impact ML research can have in industry. I saw first-hand how academic work can solve pressing industrial challenges, with data-driven approaches improving products and services at scale. It also reinforced my belief that generative AI’s true promise lies in enabling personalised experiences and products, across industries – and that realizing this potential depends critically on high-quality data.
The startup environment at Hike also demonstrated how a small, dedicated team can drive significant innovation. The ability to implement ideas quickly and see immediate results was enlightening. For PhD students considering industry internships, I strongly recommend the experience – it offers invaluable opportunities to apply research skills to real-world problems while exploring new domains.
Thank you to Kyle for sharing his experiences of this successful internship.
If you have something you want to share with the AgriFoRwArdS community, please get in touch at agriforwards.cdt@lincoln.ac.uk.