The goal of this project is to provide tools to enable patients with diabetes to communicate their blood sugar level and blood pressure test results. These test results data must be visualised for doctors so that they can track the health and wellbeing of their patients and take actions if necessary.
In countries like Somalia, in which poor health literacy is prevalent, diabetic patients are unable to attend frequent doctor appointments. Diabetes mellitus is a chronic condition, predominantly affecting the pancreas, that fails to adequately produce insulin. This results in dangerously high blood sugar levels. These spikes often cause comorbidities such as diabetic ulcers, blindness and hypertension. As these can rapidly develop, regularly scheduled doctor appointments are highly recommended to maximise patient care. However, poor online infrastructure in Somalia has resulted in poor patient compliance for these appointments as there is currently no digital platform for doctors to track and monitor their patients.
To combat this problem we have developed an app entitled "Korta", that aims to resolve the disconnect between patients and doctors. This results in an improvement in both patient compliance and overall care. Korta is a mobile and web-based app designed for both diabetic patients and their doctors. Using the app, doctors can create individual patient profiles to allow for the ability to automatically and securely upload patient data with minimal literary requirements. The biometric data is displayed graphically for the clinician with any abnormalities highlighted. Notifications alert the doctor to contact a patient should they need to be called in for an appointment.
Overall, a platform has been created to track and monitor diabetic patients in Somalia, thereby improving patient care. The app has the potential to be translated to other regions or to monitor additional biometrics like cholesterol.
Finally, the app is particularly relevant to COVID-19 (of which diabetes is a large risk factor), as allowing doctors to remotely identify and monitor these high-risk patients without the need for an in-person assessment would reduce a diabetic patient's risk of contracting the virus.
Allows patients to input readings. Designed for illiterate users. Readings can be inputted manually or through optical character recognition.
Allows doctors to sign-up patients and monitor them. Includes a visualisation and alert system, patient follow-up reminders and patient profiles.
Allows signing-up doctors, clinics and admins. This dashboard serves both administrators and doctors. Includes an email verification procedure.
-
Android Studio for the doctor and patient apps.
-
Google Vision API and OpenCV for OCR.
-
MPAndroidChart for visualisations.
-
Firebase for authentication.
-
Node.js for the server.
-
Express to handle requests.
-
MySQL as the database.
-
Developed using TypeScript.
-
Firebase for authentication.
-
Azure Blob Storage to store images.
Patient App
Doctor App
Web App
Developed by a team comprising of Senior Sophister, BA (Mod.) Computer Science students and MSc Bioengineering students at Trinity College, Dublin.
Please get in touch with one of the contributors to discuss installation.