Pengembangan Lanjutan Frontend Pulse Wise – Aplikasi Mobile untuk Memonitor Pasien Gagal Jantung Berbasis Machine Learning dan Smartwatch

Gusyanto, Frans Nicklaus (2026) Pengembangan Lanjutan Frontend Pulse Wise – Aplikasi Mobile untuk Memonitor Pasien Gagal Jantung Berbasis Machine Learning dan Smartwatch. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5026221089-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Gagal jantung memerlukan pemantauan jangka panjang dan manajemen mandiri yang konsisten untuk membantu pasien mengenali perubahan kondisi serta menjalankan perawatan harian. Pulse Wise versi awal telah menyediakan fungsi pencatatan kesehatan, tetapi masih memiliki keterbatasan pada antarmuka, integrasi data perangkat, dukungan prediksi, dan pemantauan oleh dokter. Tugas akhir ini bertujuan mengembangkan lebih lanjut frontend Pulse Wise sebagai aplikasi seluler berbasis Android yang menerapkan prinsip desain ramah lansia, mengintegrasikan data kesehatan melalui Health Connect, menampilkan hasil prediksi machine learning, serta mendukung pemantauan pasien oleh dokter. Pengembangan dilakukan melalui analisis sistem terdahulu, perancangan antarmuka, implementasi menggunakan Flutter, integrasi layanan backend, serta pengujian unit, pengujian fungsional, validasi pakar, dan User Acceptance Testing. Hasil tugas akhir menghasilkan aplikasi dengan tiga peran, yaitu pasien, dokter, dan admin. Fitur pasien mencakup diari kesehatan, pengingat obat, kontak darurat, edukasi, dasbor metrik, sinkronisasi Health Connect, serta prediksi dan rekomendasi. Dokter dapat mengakses data pasien yang telah ditautkan, sedangkan admin mengelola pengguna dan verifikasi dokter. Sebanyak 342 kasus uji unit dan 24 skenario pengujian fungsional memperoleh status lulus pada pengujian akhir. Validasi oleh dua dokter spesialis jantung menghasilkan nilai rata-rata 3,93 dari skala 5, sedangkan evaluasi terbatas yang melibatkan dua calon pengguna dan satu ahli UI/UX menghasilkan nilai rata-rata 4,33 dari skala 5. Hasil tersebut menunjukkan bahwa Pulse Wise telah memenuhi kebutuhan fungsional dasar dan memperoleh tanggapan awal yang positif. Meskipun demikian, pengujian dengan pasien gagal jantung lanjut usia dalam jumlah lebih besar dan validasi klinis terhadap model prediksi masih diperlukan.
=====================================================================================================================================
Heart failure requires long-term monitoring and consistent self-management to help patients recognize changes in their condition and perform daily care activities. The initial version of Pulse Wise provided basic health-recording functions but remained limited in terms of interface design, device-data integration, predictive support, and physician monitoring. This study aims to further develop the Pulse Wise frontend as an Android-based mobile application that applies age-friendly design principles, integrates health data through Health Connect, presents machine learning prediction results, and supports patient monitoring by physicians. The development process involved analyzing the previous system, designing the user interface, implementing the application using Flutter, integrating backend services, and conducting unit testing, functional testing, expert validation, and User Acceptance Testing. The resulting application supports three user roles: patients, physicians, and administrators. Patient features include a health diary, medication reminders, emergency contacts, educational content, a health-metrics dashboard, Health Connect synchronization, and prediction and recommendation functions. Physicians can access data from linked patients, while administrators manage users and verify physician accounts. A total of 342 unit test cases and 24 functional testing scenarios passed in the final testing stage. Validation by two cardiologists produced an average score of 3.93 out of 5, while a limited evaluation involving two prospective users and one UI/UX expert produced an average score of 4.33 out of 5. These results indicate that Pulse Wise has fulfilled its basic functional requirements and received positive initial feedback. However, further testing involving a larger number of older adults with heart failure and clinical validation of the prediction models are still required.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Gagal Jantung, Health Connect, Machine Learning, Mobile Health, Pulse Wise
Subjects:	Q Science > QA Mathematics > QA76.758 Software engineering
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Frans Nicklaus Gusyanto
Date Deposited:	29 Jul 2026 06:02
Last Modified:	29 Jul 2026 06:02
URI:	http://repository.its.ac.id/id/eprint/139448

Actions (login required)

View Item