RJ Nieto
Hi, I’m Rey Joseph “RJ” Nieto.
I am a mathematician and university instructor with a First Class Honours BSc in Mathematics from the University of Bristol. I currently teach mathematics and information technology at Bulacan State University (BulSU) in the Philippines.
Alongside my teaching, I am undertaking structured independent study in dynamical systems, functional analysis, and measure theory and probability. This supports my continuing research in topology-informed computer vision, where I am studying the reliability of measurements derived from retinal images and possible applications in biomedical and industrial imaging.
More than two decades ago, I first studied mathematics formally at UP Diliman, but left before completing the degree because of financial hardship.
Before returning to mathematics, I worked in journalism, broadcasting, public affairs, technical writing, and government communications.
Teaching
Course materials for general education subjects Mathematics in the Modern World and Living in the IT Era, which I teach at BulSU.
Research
My final-year project began as a topology-informed classifier for diabetic retinopathy. During testing, a simulated rotation of roughly three degrees drove the topology-based version close to chance performance. Rather than set that failure aside, I changed direction and studied whether the different ways mathematical measurements respond to blur, rotation, resolution changes, and sensor noise could help identify what had happened to an image.
The work used cubical persistent homology, a computable but conservative stability bound, and a parallelised Python pipeline to examine retinal images under simulated acquisition errors.
I am now returning to a question left open by that project: whether the effects of sampling and interpolation on a pixel grid can be described more precisely than the conservative general bound used in the thesis. I am also examining whether the contrasting responses of topological and geometric measurements can help distinguish different degradation signatures. This is ongoing research; I am not assuming in advance that either question will yield a new theorem.
Professional Background
My earlier career spans journalism, broadcasting, public affairs, government communications, and digital publishing.