A multi-sensor research and analysis platform designed to turn overnight physiological signals into structured sleep, breathing and recovery insights.
Built for research and clinical investigation. Not a medical device.
Flinders University · FHMRI Sleep Health · South Australia
Developed through multidisciplinary sleep research
The platform
MedBug SENSE brings sensing, machine-learning inference and research reporting into one connected workflow—designed to reduce friction without losing signal depth.
Five-stage analysis across Wake, N1, N2, N3 and REM in 30-second epochs.
Structured respiratory-event predictions and overnight pattern summaries.
Reproducible plots, summaries, exports and PDF reporting for each study run.
Heart-rate and respiratory-rate profiles derived from overnight piezo signals.
Audio-feature workflows support snore-event and overnight acoustic investigation.
Separate predictions for respiratory and other arousal events within sleep periods.
In-bed filtering focuses downstream analysis on relevant overnight periods.
Research outputs include ventilation burden and respiratory-drive patterns.
Temperature, environmental and light-sensing inputs support richer study context.
MedBug V3 kit
The research prototype
A mattress-based sensing approach brings physiological, acoustic and contextual signals together for multi-night sleep research in laboratory and home settings.
Prototype shown. Final specifications, availability and regulatory status are not yet published.
The science
MedBug’s modular analysis workflow moves from overnight sensing to quality control, model inference and human-readable reporting with a traceable run history.
Discuss a research study ↗Piezo, acoustic, temperature and environmental signals create a richer research input.
Separate workflows support sleep stage, arousal and respiratory-event analysis.
Configuration, manifests and run-level outputs make analysis easier to trace and reproduce.
Structured CSV, JSON, plots and compiled reports support different study needs.
Research & development team
MedBug brings together sleep science, clinical research, engineering and machine learning through a multidisciplinary Flinders University collaboration.
Sleep and breathing physiology, clinical interpretation and multidisciplinary research leadership.
01Machine-learning workflows, research engineering and integrated sleep-analysis systems.
02Research across sleep-stage analysis, sensor validation and translation into multi-night studies.
03Collaborating investigators
Contributors identified across MedBug research and conference materials.
How it works
A connected research workflow built around repeatability.
Set up the MedBug SENSE unit according to the study protocol and begin the configured period.
Capture continuous overnight signals for configured, research-led analysis.
Process signals through quality checks and models into structured outputs.
Review study-level plots, tables and reports while retaining data lineage.
Built for collaboration
MedBug is being shaped for researchers, clinicians and partners exploring how multi-sensor overnight analysis can support better sleep investigation.
Research developed within Flinders University sleep-health and medical-device programs in South Australia.
Sleep researchers studying sleep architecture, breathing and longitudinal patterns.
Clinical collaborators evaluating new approaches to at-home investigation.
Technology partners interested in sensing, analysis or translation pathways.
Register your interest
Tell us where you see a fit. We’ll use your response to understand interest in future research, collaboration and product updates.
No clinical advice or diagnosis is provided through this form.
Interest registered
Your response has been received. The MedBug research team can now follow up with you.
Registration details are collected to respond to project enquiries and collaboration interest. Before public release, the approved policy will state the data controller, retention period, access rights and contact process in full.
MedBug outputs are generated from sensor data and machine-learning predictions for research and clinical review. They are not a substitute for attended polysomnography or independent clinical judgement. The platform is research-stage and is not marketed as a medical device.