Nástroj pro anotaci medicínských snímků

Abstract

This bachelor’s thesis focuses on the design and implementation of a cross-platform application for annotating medical images, intended for use on both desktop operating systems (Windows, Linux, macOS) and in modern web browsers. The main goal was to develop an intuitive and powerful tool that enables doctors to easily and precisely annotate regions of interest in medical images in the DICOM format.

The application was developed using the modern Tauri framework, which combines the performance of native desktop applications with the flexibility of web technologies. The user interface is built in HTML with Svelte framework and TypeScript.

The resulting application supports loading DICOM images, working with annotations including exporting them to JSON format, responsive interface for different screen sizes, and functioning without an internet connection. Emphasis was placed not only on user-friendliness and high annotation accuracy, but also on reliability, stability, and ease of code maintainability.