Pengembangan Platform Anotasi Gambar Medis Berbasis Web dengan Fitur Edge Detection dan Multi-Layer Segmentation

Lathifa, Arsya Dewi and Kamaski, Adelia Putri (2026) Pengembangan Platform Anotasi Gambar Medis Berbasis Web dengan Fitur Edge Detection dan Multi-Layer Segmentation. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5025221015_5025221320-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Artificial Intelligence di bidang medis, khususnya model untuk melakukan segmentasi citra medis membutuhkan data latih yang teranotasi dengan akurat. Namun, proses segmentasi dan anotasi citra medis secara manual memakan waktu lama dan membutuhkan ketelitian tinggi. Kerja Praktik ini bertujuan untuk mengembangkan sebuah platform anotasi gambar medis berbasis web yang dapat mempercepat pra-proses data latih. Platform ini dibangun menggunakan antarmuka berbasis web dengan teknologi React.js yang menerapkan arsitektur Single Page Application (SPA) untuk responsivitas tinggi dan Progressive Web App (PWA) untuk memungkinkan aksesibilitas luring (luar jaringan). Platform anotasi gambar medis ini memiliki beberapa fitur unggulan seperti integrasi algoritma Canny Edge Detection yang membantu pengguna mendeteksi tepi objek secara otomatis, fitur Multi-Layer Segmentation yang memungkinkan pengelolaan anotasi kompleks, dan ekspor data segmentasi ke format biner (.bin) yang dirancang agar kompatibel dengan kebutuhan training model machine learning. Pengujian fungsional menunjukkan bahwa sistem berhasil mengintegrasikan tools anual seperti brush, eraser, dan flood fill yang dapat memudahkan proses annotasi maotasi data medis.
===================================================================================================================================
Artificial Intelligence in the medical field, particularly models for medical image segmentation, requires accurately annotated training data. However, the process of manually segmenting and annotating medical images is time-consuming and requires a high degree of precision. This practical project aims to develop a web-based medical image annotation platform that can accelerate the pre-processing of training data. This platform is built using a web-based interface with React.js technology that implements a Single Page Application (SPA) architecture for high responsiveness and a Progressive Web App (PWA) to enable offline accessibility. This medical image annotation platform has several outstanding features, such as the integration of the Canny Edge Detection algorithm, which helps users automatically detect object edges, the Multi-Layer Segmentation feature, which enables complex annotation management, and the export of segmentation data to a binary format (.bin) designed to be compatible with machine learning model training requirements. Functional testing shows that the system successfully integrates annual tools such as brush, eraser, and flood fill, which can facilitate the medical data annotation process.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	Medical image annotation, Edge detection, Binary format, Multi-Layer Segmentation, PWA, React.js, Anotasi gambar medis, Edge detection, Format Biner, Multi-Layer Segmentation, PWA, React.js.
Subjects:	T Technology > T Technology (General) > T58.6 Management information systems
Divisions:	Faculty of Industrial Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Arsya Dewi Lathifa
Date Deposited:	09 Jan 2026 03:08
Last Modified:	09 Jan 2026 03:08
URI:	http://repository.its.ac.id/id/eprint/129358

Actions (login required)

View Item