NEURASCAN: Sistem Deteksi Tumor Berbasis Web pada Citra MRI

NEURASCAN: Brain Tumor Detection System

Deep Learning • Medical Imaging • Mask R-CNN • FastAPI

Overview

NEURASCAN is a web-based brain tumor detection system designed to assist in the analysis of MRI scans using deep learning. The platform leverages Mask R-CNN for automated tumor detection and localization, providing visual segmentation results through an interactive web interface.

Problem to Solve

Manual examination of MRI images can be time-consuming and requires specialized expertise. Early and accurate tumor identification is crucial for supporting clinical decision-making and reducing diagnostic workload. The challenge was to develop an accessible web-based solution capable of automatically detecting tumor regions from MRI scans.

Solution Approach

Developed an automated tumor detection pipeline using Mask R-CNN to perform object detection and pixel-level segmentation on brain MRI images. The model identifies tumor regions and generates segmentation masks that highlight affected areas for visual analysis.

Built a FastAPI-based backend to serve the deep learning model and process MRI image uploads in real time. The web application provides an intuitive interface for image submission, prediction visualization, and result interpretation, enabling efficient interaction between users and the AI model.

Key Technologies

Python
Mask R-CNN
Deep Learning
Computer Vision
MRI Image Processing
FastAPI
REST API
Web Application

Impact

        Result:
        Delivered a functional AI-powered web application capable of detecting and localizing
        brain tumors from MRI scans. The system demonstrates the practical application of deep
        learning in medical image analysis while providing a scalable architecture for deployment
        in research, educational, and healthcare-support environments.
    

Live Demo

▶

NEURASCAN Demonstration

Brain Tumor Detection and Segmentation on MRI Images using Mask R-CNN

Watch how NEURASCAN analyzes MRI images, detects tumor regions, and generates visual segmentation results through a web-based interface powered by Mask R-CNN and FastAPI.