Mammography Datasets for Neural Networks-Survey

Abstract

Deep neural networks have gained popularity in the field of mammography. Data play an integral role in training these models, as training algorithms requires a large amount of data to capture the general relationship between the model's input and output. Open-access databases are the most accessible source of mammography data for training neural networks. Our work focuses on conducting a comprehensive survey of mammography databases that contain images with defined abnormal areas of interest. The survey includes databases such as INbreast, the Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM), the OPTIMAM Medical Image Database (OMI-DB), and The Mammographic Image Analysis Society Digital Mammogram Database (MIAS). Additionally, we surveyed recent studies that have utilized these databases in conjunction with neural networks and the results they have achieved. From these databases, it is possible to obtain at least 3801 unique images with 4125 described findings from approximately 1842 patients. The number of patients with important findings can be increased to approximately 14,474, depending on the type of agreement with the OPTIMAM team. Furthermore, we provide a description of the annotation process for mammography images to enhance the understanding of the information gained from these datasets.

Keywords: artificial intelligence; deep neural networks; machine learning; mammograms; mammography; open-access databases.

Figure 1

Four standard views: (a) right CC, (b) left CC, (c) right MLO, (d) left MLO. Source: [11].

Figure 2

Standard reporting system.

Figure 3

ACR standardized breast density. Source: [12].

Figure 4

Different types and margins of masses. Source: [12].

Figure 5

Different types of calcifications. Source: [11].

Figure 6

Distribution of calcifications. Source: [17].

Figure 7

Architectural distortion examples. Source: [11].

Figure 8

Asymmetry seen on LMLO and LCC views. Source: [17].

Figure 9

Binary mask of mass finding. Source: [11].

Figure 10

Overview of the MIAS, CBIS-DDSM, and INbreast datasets.

Figure 11

The MIAS database statistics.

Figure 12

The CBIS-DDSM database statistics.

Figure 13

The INbreast database statistics.

Figure 14

Artifacts pointed to by arrows, left and middle images without pectoral muscle, and middle and right images do not contain whole breasts. Source: [19].

Figure 15

Redundant elements marked with red squares and arrows. Source: [19].

Figure 16

Lined contour points of the pectoral muscle. Source: [11].