Project / product name: DiApp
Link to the project: https://github.com/fejtcz/ehh2022
Team leader: Martin Fejt
Challenge: 5. Continuous glycemic monitoring in intensive care units
Problem: Many hospitals around the world are addressing the underlying problem. There is a lot of activities, that are still manual and require significant effort from hospital staff. These activities include, for example, regular blood sugar readings. This time could certainly be used more efficiently and allow staff to work in a more pleasant environment.
Solution: We offer a system that is ready for communication through the api of individual suppliers of closed solutions as well as our own HW that can communicate with Dexcom sensors directly via BlueTooth. A big benefit is the ability to read data in parallel alongside the patient's client application. Our application can easily pair users, visualize waveforms, raise alarms and analyze the data retrospectively.
Impact: Simplification of repetitive tasks. Working in a less stressful environment. Data collection for subsequent analysis and the ability to look for anomalies in the patient's condition. It's cheap and easy to implement into existing systems. Better overwiev of patients health and automatic alerting based on real time readings increasing safety.
Feasibility & financials: At the hackathon we tested the possibility of independent reading of glucose meters with the help of third-party api, as well as direct wireless reading with our own hardware from the sensors. Therefore, we believe that we are able to design and deliver simple and cheap hardware in the tens of euros, which will ultimately save the customer considerable costs.
What is new about your solution?: We have used standard formats and protocols such as RestAPI, MQTT, which makes it easy for interacting with existing systems. You can visualize data in your intern systems or use connections for training neural networks for possible future predictions of severe or critical conditions. Our solution is independent of Dexcom cloud services, which can be possible "bottleneck" as unknown third party SW. The hospital can decide, where the patients data will be securely stored.
What you have built at the hackathon - text explanation + code (e.g. GitHub link): HW prototype for transmitting data between sensors and servers. Backned server and demo frontend app for visualizing data Source codes: https://github.com/fejtcz/ehh2022
What you had before the hackathon, please mention open source as well: We had nothing prepared before the hackhaton. We used open source python library pydexcom for communicating with Dexcom API.
What comes next and what you wish to achieve: We would like to continue in our work and build usable application for nurses from our prototype.