Project / product name: Tereza, the virtual nurse
Link to the project: http://
Team leader: Martin Fejt
Challenge: 4. Efficient Glucometer Data Integration
Problem: Before entering the doctor’s office, patients need to provide the nurse data from their glucometers or CGM . Also blood analysis reports may be required for further treatment. Nurses need to validate presence of these documents, ensure completeness, review & select relevant data and rewrite them into hospital IS. These tasks do not require professional medical skills. There are other repeated tasks such as education and providing them information relevant to their diabetes compensation results.
Solution: Our solution is a self service front office kiosk that combines various tasks that can be automated and digitized. Patient will identify himself by scanning insurance card, provide required reports from CGM or blood analysis via download or scan and these documents will be automatically processed and transferred into hospital IS. Patient can be added to waiting queue and provided pickup number. Kiosk will also evaluate the data and provide the patient further information and education materials.
Impact: Most visible effect of this solution is simplification and streamlining of the patient inbound process. While the administrative tasks are automated, a possibility of human error is eliminated. Automatic data processing and evaluation will provide patient relevant education documents immediately. Digitized process and dynamic user interface allow further extension of the functionality, such as implementation of fee payments, doctor office date picker, and more.
Feasibility & financials: During the hackathon we built a proof of concept of the app. We learned we can automate the documents processing using OCR, work with FHIR and the core part of the technical solution is confirmed feasible. Further development will require implementation of the UI steps and integrations with CGM providers and hospital information system. Our solution does not require any specific hardware or software as it runs in conventional web browser and the only specialty is need of physical kiosk stand.
What is new about your solution?: The most interesting part is the automated processing of CGM/glucose or other reports that is done using OCR. Currently these processes are mostly manual and automating this process will eliminate the risk of human error and makes the data readily available in the hospital IS.
What you have built at the hackathon - text explanation + code (e.g. GitHub link): Sofware for OCR, app for terminal/kiosk, FHIR integration server and hardware accessorie for terminal wall mount GitHub repo: https://github.com/fejtcz/EHH-2023
What you had before the hackathon, please mention open source as well: Some experience with OCR, and greate idea
What comes next and what you wish to achieve: Sleep well, and we are prepared for pilot implementation in the hospital..