CNN for medical imaging
Project: Deep Learning applied to the analysis of mammography images.
Mammography is a particular form of radiography, which works with levels of voltages and currents at specific intervals, intended to record images of the breast in order to diagnose the presence or absence of structures that may indicate diseases.
Images extracted from: https://wiki.cancerimagingarchive.net/display/Public/CBIS-DDSM
Digital Database for Screening Mammography (DDSM): http://www.eng.usf.edu/cvprg/Mammography/Database.html containing images of normal cases, benign and malignant.
Lee, R.S.; Gimenez, F.; Hoogi, A.; Miyaki, K.K.; Gorovoy, M.; Rubin, D.L. 2017. Data Descriptor: A curated mammography data set for use in computer-aided detection and diagnosis research. Scientific Data.4: 1-9.
Meaning of some research-related terms:
MLO = Oblique Medium-Lateral Incidence; CC = Craniocaudal Incidence; DICOM = Digital Imaging and Communications in Medicine - Standard image format for medical images; ROI = Region of interest. Abnormalities are represented as binary mask images (ROI mask images); BI-RADS = Breast Density Descriptions (1-almost entirely fat; 2-containing scattered areas of fibrogranular density; 3-heterogeneously dense; 4-extremely dense).