DEVELOPMENT OF AN ANDROID-BASED APPLICATION INTEGRATED WITH A BLOOD GLUCOSE MONITORING DEVICE FOR DIABETES MELLITUS PATIENTS

• Main Author: Larasati, Rachmaniar
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/78124
• Institution: Institut Teknologi Bandung

Diabetesmellitus does not yet have a cure or treatment method that can completely cure this disease. Maintaining a healthy lifestyle and regular monitoring become the only way to manage this disease. Traditional monitoring methods using invasive or minimally invasive measurement tools are less effective due to the risk of infection. These methods are also inefficient for patients because they have to manually record the measurement results, often resulting in non compliance of periodic measurement over time. Additionally, diabetic patients cannot determine the categorization of their blood glucose levels during each monitoring session, when the normal standard for blood glucose levels may vary depending on the patient's condition, specifically time to prior last meal. This research focuses on the development of an Android-based application to facilitate patient monitoring. The application is integrated with a specific non-invasive blood glucose monitoring device through Bluetooth and a Cloud Firestore-based database management system. This application consists of several features, including automatic and manual measurements, recording the severity of common complications in diabetic patients, and viewing the history of blood glucose level measurements and complication severity. The data sent to Cloud Firestore is stored and generates categorizations of blood glucose levels based on the time of measurement after eating. Users can also view measurement histories in the form of lists and scatter charts. All of the actual results of functionality testing are align with the expected outcomes for each functionality test case. Qualitative usability testing involving ten samples was also conducted which resulted in 16,83 out of 20 in scoring when the result is being quantified. Based on the questionnaires provided, the majority of samples stated that the features were not too difficult to understand and navigate.