Waloni, Rigel Ramadhani (2026) Pengembangan Sistem Teleradiologi IBrain2U Dengan Arsitektur Pull-Based Routing Untuk Koordinasi Diagnosis Antar Rumah Sakit. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5024221058-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5024221058-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (18MB) | Request a copy |
Abstract
Ketimpangan akses layanan diagnostik radiologi di Indonesia mendorong kebutuhan sistem teleradiologi yang aman dan adaptif terhadap kapasitas fasilitas kesehatan. Penelitian ini mengembangkan sistem berbasis web yang terintegrasi dengan platform iBrain2U untuk memperluas aksesibilitas diagnosis gangguan otak menggunakan deep learning. Sistem mengimplementasikan arsitektur pull-based routing yang memberikan fleksibilitas kepada rumah sakit untuk mengambil tugas diagnosis sesuai kapasitas sumber daya mereka, berbeda dari pendekatan push konvensional. Keamanan data medis dijamin melalui mekanisme berlapis meliputi validasi format DICOM, pemindaian malware, enkripsi AES-256, autentikasi OAuth2/JWT, dan audit log yang sesuai dengan UU No. 27 Tahun 2022. Interface web dirancang untuk mendukung upload file DICOM oleh pasien dengan verifikasi otomatis, sementara connector berbasis pull melakukan polling periodik untuk mengambil tugas baru. Arsitektur sistem memastikan integrasi seamless dengan pipeline AI iBrain2U tanpa modifikasi pada sistem existing, sambil mempertahankan alur verifikasi dokter. Penelitian ini berkontribusi pada pemerataan akses diagnosis berbasis AI dengan keamanan dan fleksibilitas operasional yang lebih baik.
======================================================================================================================================
The disparity in access to radiological diagnostic services in Indonesia necessitates a secure and adaptive teleradiology system tailored to healthcare facility capacities. This research develops a web-based system integrated with the iBrain2U platform to expand accessibility for brain disorder diagnosis using deep learning. The system implements a pull-based routing architecture that provides flexibility for hospitals to retrieve diagnostic tasks according to their resource capacity, diverging from conventional push approaches. Medical data security is ensured through multi-layer mechanisms including DICOM format validation, malware scanning, AES-256 encryption, OAuth2/JWT authentication, and audit logging compliant with Law No. 27/2022. The web interface is designed to support DICOM file uploads by patients with automatic verification, while a pull-based connector performs periodic polling to retrieve new tasks. The system architecture ensures seamless integration with the existing iBrain2U AI pipeline without modifications, while maintaining doctor verification workflows. This research contributes to democratizing AI-based diagnosis access with improved security and operational flexibility.
Actions (login required)
 |
View Item |