Engineering the Unknown in
Biomedical Systems
CBMS is an early-stage research and engineering lab. We design proprietary biomedical devices, embedded hardware, and the computational software that makes them work — from first hypothesis through to hardware that survives a bench.
Three Tracks We Are Actively Working In
Computational Software & Signal Processing
Turning raw physiological signal into defensible, low-latency inference.
Explore TrackEmbedded Biomedical Hardware & Sensor Systems
Custom acquisition hardware where off-the-shelf instrumentation runs out.
Explore TrackClosed-Loop Clinical Engineering
Diagnostic rigs and human-machine interface frameworks where the loop closes.
Explore TrackHow We Build Things
Algorithm Design
Signal processing and predictive physiological modelling built for real-time inference, with evaluation harnesses that make the result reproducible rather than merely plausible.
Hardware–Software Co-Design
Acquisition front-ends, sensor interfaces and firmware designed alongside the algorithms that consume them — because the sensor, not the model, is usually the binding constraint.
Clinical Engineering Systems
Diagnostic rigs and human-machine interface frameworks assembled as closed loops, with the safety case and verification strategy designed in from the first sketch.
Falsify Early, Build Deliberately
Most of what we publish is a negative result. That is the point — a lab that cannot discard an idea quickly never gets to the one worth building.
Frame the Problem
We start from the clinical or measurement problem, not a product category. Most of our work begins by dismantling an assumption nobody had checked.
Prototype Cheaply
A bench rig before a board. We build the smallest instrument that can falsify the hypothesis, because most hypotheses should die early and cheaply.
Design for Evidence
Verification, hazard analysis and reproducibility are part of the design from the first commit — not a phase bolted on before someone asks who signed off.
Partner or Publish
What we build is useful to us only if it transfers. We collaborate with academic and clinical partners, and we publish what does not need to stay proprietary.
Have a Problem Worth Solving?
We partner with academic groups, clinical teams, and companies who need an engineering team that will tell them whether the idea holds up.