Industrial Experience Project

Asthma Detection Wearable Technology


Fields required

Electrical Engineering, Mechatronics, Medical Engineer (any experience/understanding in Acoustics will be advantageous)

Melt Supervisors
Adam Van Dyck

Project Background

Respia Technology has developed the first asthma smart wearable that monitors and help prevent asthma events in children, called R1. Asthma affects 1 in 8 children in Australia with a total population of 2.7 million living with Asthma symptoms everyday.

Respia aims to help parents monitor their children’s respiratory health through their innovative product including a wearable device paired with proprietary machine learning technology that detects early signs of asthma attacks by alerting parents through notifications in a smart phone app.

Respia’s adaptive machine learning algorithm takes into account the external conditions such as weather and air quality. The device alerts you as soon as the wheezes bypass a defined threshold, before the child gets into a potentially dangerous respiratory status.

Respia can become the first solution to automatically alert about the incoming danger, avoiding the burden of analysing the data collected as required by present devices.

Other companies/research used time domain conversion to extract coughing. Some used computer vision techniques. We use a state estimation layer that constantly checks predictions against actual data (autonomous car stuff) then feeds into a heavy-duty anomaly detection algorithm to determine where the subject is on the asthma severity scale.



Project Aim

Respia will run a “Clinical Study” with 100 control and 100 patients in Westmead Children’s Hospital to test the prototype and collect crucial data to test their algorithm. This data will be used in their next generation models that aim to rival the current gold standard. This project aims to provide support to Respia before, during and after the Clinical Study for which they need electrical engineer/mechatronic engineer to assist with the development and production of the prototype, collecting and analysing the data, and suggest improvements to the product and machine learning and AI technology based on research, Study results and the student recommendations.

Project Objectives

Objective 1: Preparation for Clinical Study.

The first stage of the project is to study the current prototype solution, suggest improvements for the future based on latest research and benchmarks, and assist senior engineering teams to produce the required number of devices for the Study. This includes (but not limited to):

  • Study the current functionality and technical design of the solution to meet its purpose
  • Help the senior team to stress test the prototype before production. Is the recording accurate? What is the level of performance of current device: speed, accuracy, reliability, consistency, resilience?
  • Suggest recommendations for improvement based on research (benchmarks, cross-industry examples, latest technology development/research, etc).

Stage 1 Deliverable:

Once this prototyping and research have been carried out, a report should be produced detailing the outcomes of the investigation and the various trade-offs that have been identified throughout the process (costs, performance, speed, accuracy, reliability, consistency, resilience), engineering effort, maintenance, availability, size, power and so on). The report should also propose a demonstration system. The proposal should detail the required components and be illustrated with a systems integration block diagram.

Objective 2: Clinical Study.

The second stage of the project is to assist the team with the prototype Study in Westmead Children’s Hospital.Assist with conducting the study with patients and control groups

  • Test performance of technical solution, product design, user interaction
  • Support with the analysis of acoustic data/recording device output
  • Review results from Study and suggest recommendation for improvement
  • Use of C++, R, Matlab to show the specificity vs. sensitivity of the adaptive machine learning algorithm
  • Suggest recommendations on Machine Learning and AI techniques for predictive asthma detection

Stage 2 Deliverable: The students will support the team during the Study on-site and involved in the processing of the data and measuring results.

Objective 3: Results Final Report.

The final stage will be the final report to include all the learnings and outcomes of the demonstration and field Study. The student should also include any updated recommendations, options to cost down, scope for further development and an overall conclusion of the feasibility of the aim of the project.

1/323 Hillsborough Rd, Warners Bay, with limited travel to Sydney
Available from
Application process

Apply here by completing a form and uploading your CV and Uni transcripts.


About Melt Accelerator

Melt Accelerator was created to satisfy a gap in the market of accelerating ideas with a hardware component in a commercially viable and risk-free environment. With a broad range of partners and supporters Melt Accelerator is the first Australian hardware accelerator and industry prototyping lab offering a rigorous stage-gated process combined with ‘state of the art’ industrial engineering equipment and team offering product proof-of-concept experimentation, prototyping, pilot builds and pre-commercialisation activities, along with scientific consultancy and technical engineering and a proven stage gated process to go from idea to large scale production.