Syahzada, Mochammad Naufal Ihza Syahzada (2026) Aplikasi Mobile Flutter-Flask Untuk Monitoring Dan Prediksi Risiko Serangan Jantung Berbasis Smartwatch. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5025211260-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5025211260-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (6MB) | Request a copy |
Abstract
Penyakit kardiovaskular merupakan penyebab kematian utama global dengan 20,5 juta kematian pada tahun 2021. Indonesia memiliki prevalensi tertinggi kedua di ASEAN dengan peningkatan 148,1% sejak 1990. Keterbatasan akses fasilitas kesehatan dan teknologi smartwatch yang belum optimal mendorong penelitian ini. Penelitian ini bertujuan mengembangkan aplikasi mobile berbasis arsitektur client server untuk monitoring elektrokardiogram (EKG) dan prediksi risiko serangan jantung menggunakan data dari Apple Watch. Sistem mengintegrasikan Flutter sebagai frontend, Flask sebagai backend, dan dua model machine learning: CNN Autoencoder untuk deteksi anomali EKG dan Decision Forest untuk prediksi risiko. Data EKG diambil melalui Apple HealthKit API, diproses untuk deteksi anomali, kemudian dikombinasikan dengan data gaya hidup pengguna untuk prediksi risiko. Evaluasi usability menggunakan heuristic Nielsen menunjukkan aplikasi memenuhi 9 dari 10 prinsip. Model Decision Forest mencapai akurasi 94% dengan precision 93-95% dan recall 93-95%. CNN Autoencoder mencapai akurasi training 93,89%, testing 90,61%, namun anomaly detection 72,38%. Response time sistem di bawah 3 detik memenuhi target responsiveness. Sistem berhasil mengintegrasikan deteksi anomali EKG otomatis dengan prediksi risiko berbasis faktor gaya hidup dalam aplikasi mobile yang user-friendly untuk konteks Indonesia.
========================================================================================================================================================
Cardiovascular diseases are the leading cause of death globally, with 20.5 million deaths recorded in 2021. Indonesia has the second-highest prevalence in ASEAN, with a 148.1% increase since 1990. Limited access to healthcare facilities and suboptimal smartwatch technology utilization motivated this research. This study aims to develop a client-server-based mobile application for electrocardiogram (ECG) monitoring and heart attack risk prediction using Apple Watch data. The system integrates Flutter as frontend, Flask as backend, and two machine learning models: CNN Autoencoder for ECG anomaly detection and Decision Forest for risk prediction. ECG data is acquired through Apple HealthKit API, processed for anomaly detection, then combined with user lifestyle data for risk prediction. Usability evaluation using Nielsen's heuristics showed the application met 9 out of 10 principles. The Decision Forest model achieved 94% accuracy with 93-95% precision and recall. CNN Autoencoder achieved 93.89% training accuracy, 90.61% testing accuracy, but 72.38% anomaly detection accuracy. System response time below 3 seconds met the responsiveness target. The system successfully integrated automatic ECG anomaly detection with lifestyle-based risk prediction in a user-friendly mobile application designed for the Indonesian context.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | cnn autoencoder, decision forest, ekg, flutter, prediksi serangan jantung, smartwatch, cnn autoencoder, decision forest, ecg, flutter, heart attack prediction, smartwatch |
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing T Technology > T Technology (General) > T58.8 Productivity. Efficiency |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Mochammad Naufal Ihza Syahzada |
| Date Deposited: | 30 Jan 2026 09:35 |
| Last Modified: | 30 Jan 2026 09:39 |
| URI: | http://repository.its.ac.id/id/eprint/130824 |
Actions (login required)
 |
View Item |