Mexican dataset of digital mammograms (MEXBreast) with suspicious clusters of microcalcifications

Published: 6 January 2025| Version 2 | DOI: 10.17632/ym4mm9xvhm.2
Contributors:
,
,
,
,
, Karina Núñez Barragán

Description

MEXBreast is a curated and annotated database of Mexican digital mammograms featuring microcalcification clusters. It includes images with resolutions of 50, 70, and 100 microns. The primary purpose of MEXBreast is to support the training, validation, and testing of deep learning models, particularly Convolutional Neural Networks (CNNs).

Files

Steps to reproduce

Here is the content of the database: MARKED MAMMOGRAMS Folder: This folder contains 620 mammograms in PNG format, with clusters of microcalcifications highlighted using bounding boxes. It serves as a visual reference to easily locate these lesions. UNMARKED MAMMOGRAMS Folder: This folder contains the same 620 mammograms in PNG format as the previous folder, but without any markings or annotations. These mammograms are suitable for conducting experiments. DATA File: This comma-separated file includes detailed information about each of the 620 mammograms, such as image name, scanner specifications, mammogram dimensions and resolutions, patient age, laterality, view type, BI-RADS code, as well as the coordinates and dimensions of the microcalcification clusters.

Institutions

Universidad Autonoma de Ciudad Juarez

Categories

Artificial Intelligence, Machine Learning, Breast Imaging, Calcification, Image Classification, Features Detection, Digital Mammography, Convolutional Neural Network, Deep Learning, Image Analysis

Funding

Consejo Nacional de Humanidades, Ciencias y Tecnologías

Licence