Breast Thermography Image Database

This repository contains breast thermographic images, segmentation results, extracted features, and classification data developed for automated breast cancer identification using thermography image analysis and deep learning techniques. Repository Contents The repository contains: 1. Original Thermographic Images Breast thermography images in RGB colour format. JPEG image format. Image size: 680 × 480 × 3. 2. Segmented Breast Images Contains segmented breast regions obtained using the proposed curvature and Gradient Vector Flow methodology. Dataset Characteristics Imaging modality: Infrared thermography Disease: Breast cancer Number of images: 63 Normal images: 35 Abnormal images: 28 Image format: JPEG Image dimensions: 680 × 480 × 3 Total extracted features: 155 features per image Dataset Source The thermographic images were kindly provided by: Silva, L.F.; Saade, D.C.M.; Sequeiros, G.O.; et al.A New Database for Breast Research with Infrared Image.Journal of Medical Imaging and Health Informatics, 2014, 4(1), 92–100. The images provided in this dataset were generated in: TELLO-MIJARES, Santiago; WOO, Fomuy; FLORES, Francisco.Breast Cancer Identification via Thermography Image Segmentation with a Gradient Vector Flow and a Convolutional Neural Network.Journal of Healthcare Engineering, 2019, Vol. 2019, Article ID 9807619.