BraNet: A mobil Application for Breast image classification based on Deep Learning algorithms.

Mobile health apps are widely used for breast cancer detection and diagnosis. Artificial intelligence plays a crucial role in developing medical tools, providing radiologists with second opinions, and reducing false diagnoses. Aim: This study aims to develop an open-source mobile app, named "BraNet," for 2D breast imaging segmentation and classification using deep learning algorithms. Methods: The BraNet app was developed using the React Native framework, offering a modular deep learning pipeline for mammography and ultrasound breast imaging classification. This application operates on a client-server architecture and was implemented in Python for iOS and Android devices. It performs image analysis, extracts masks, and classifies them into benign or malignant classes. The components include data loading, Region of Interest (RoI) extraction, segmentation, classification, and statistical evaluation.

Description
The development of a mobile application for Android and iOS devices is proposed, in which breast ultrasound or mammography images are loaded, then a segmenter is used to extract the regions of interest (ROI) (optional), finally a classifier algorithm is applied to these to determine if the image is benign or malignant.

Languages used
TypeScript
JavaScript
Python 3.11
Tools/Frameworks used
Figma
VS Code
Jupyter Notebooks
Flask
PyTorch 2.0.1
React Native 0.71
Database and cloud plataforms
Google Firebase
Google Cloud