Veluon

Myopia Detection

Software engineeringMechanical engineeringAI & machine learningScientific R&DPrototypingOptical engineeringSimulationData scienceSupply chain

We took a university research concept for detecting myopia progression from eye images through feasibility research, clinical validation and product development into real-world clinical use.

For further detail refer to Phases.

Brief

Myopia is increasing rapidly worldwide and currently affects approximately 1 in 3 people. It commonly develops in children between the ages of 6 and 13, during which refractive error can change by several dioptres over a relatively short period of time. This means that the conventional approach of annual eye examinations may not provide sufficient monitoring for children experiencing rapid progression.

A university-developed technology aimed to address this challenge by detecting changes in refractive error from images of the eye, potentially enabling more frequent monitoring of myopia. However, the team lacked the tools and technical capability required to determine whether the underlying concept was feasible.

Veluon was initially tasked with carrying out the scientific research and development required to assess the technical feasibility of the concept. Following this initial work, the project expanded to explore how the technology could be translated into a viable clinical product, including understanding its place within existing care pathways, validating the technology on patients, and developing a robust and repeatable system for real-world use.

Deliverables

The project progressed from early-stage scientific validation through to the development of a clinically-usable system. We first developed simulation models and a physical model eye to establish whether refractive error produced measurable optical changes. The technology was then translated into an iPhone-based system and evaluated through small-scale clinical studies. Following successful validation, the optical hardware was refined, machine learning models were developed to detect changes in refractive error, and supporting mobile and web applications were created. The resulting system was manufactured in small quantities and progressed into use within clinical environments.

  • Developed a ray-tracing simulation to validate the underlying concept
  • Designed and built a physical model eye for real-world verification
  • Developed an optical system for use with an iPhone
  • Conducted small-scale clinical studies in collaboration with Aston University
  • Developed machine learning models to detect changes in refractive error
  • Developed a mobile application and web portal for clinicians and patients
  • Designed the system for repeatable manufacture and calibration
  • Progressed the technology into use within clinical environments