IoT telemedicine platform for cough sound recognition
Carepath Technologies GmbH is a telemedicine platform to ambiently monitor symptom-specific data in a home environment, aiding physicians with diagnoses and treatment of the patient’s chronic disease.
The challenge
Our client aimed to develop a hardware device for patient monitoring in Germany, which would be installed within their home. In the first release the device would use an AI algorithm for smart sound processing to filter only medical-useful audio. Doctors would then be able to remotely visualize collected symptom data, then take necessary actions quickly.
Our client appreciated Lemberg Solutions' full-stack IoT expertise and we engaged on this project to help design the solution architecture, build out back-end/cloud features as well as the embedded software for the IoT device, ensuring the whole system is scalable, secure, and complies with software and hardware device requirements. Moving forward, we scoped out a user interface - a web-based portal for doctors, where they could track the state of their patients and receive notifications and alerts.
Delivered value
Lemberg Solutions developed an SBC-based IoT gateway device with secure embedded software which actively monitors patients and helps to quickly deliver health-related data to caregivers, allowing them to look for signs and deviations from normal patient conditions. The platform enables an accurate chronological overview of the symptom progression, features AI-based smart filtering of audio, and sufficiently protects patient-generated data.
The process
The client partnered with Lemberg Solutions' IoT team at the stage of product ideation. We started the project with brainstorming and discovery of core features, user stories and user flows, then designing the system architecture, and then moved to other embedded Linux development services by implementing the major components of our platform on the edge as well as in the cloud.
Together with product owners our team realized that development of such a platform will require frequent changes to the scope. This is why we settled on Scrum methodology and took a sprint-based approach to ensure flexibility and fast development pace to keep up with the deadlines. We also set up a Continuous Delivery pipeline to ensure we can ship new valuable code into designated environments as quickly as possible.
Special attention was given to quality assurance aspects related to the IoT device and together with the client we compiled thorough test cases to run rigorous validation testing in QA and Stage environments and make sure our device can withstand peak loads even after being deployed to production.
At later stages we helped the client to optimize the platform infrastructure in the cloud. And also analysed and suggested how many virtual machines were necessary to sufficiently run AI algorithms.
Like in most healthcare projects, security is king. To make sure our solution was compliant with data protection regulations we came up with specific security measures and solutions. These measures were implemented with the help of our client and its partners, being validated by an independent security consulting firm.
Currently, we are developing a web portal for doctors and other care professionals, to better visualize and understand the underlying data behind the patient’s condition.
How it works
