Multiview multimodal network for breast cancer diagnosis in contrast-enhanced spectral mammography images

Deep learning-enabled fully automated pipeline system for segmentation and classification of single-mass breast lesions using contrast-enhanced mammography: a prospective, multicentre study

“…Studies have attempted to adopt the method of feature fusion in CC and MLO views using deep learning. 19 , 20 After performing the global average pooling on the feature maps, the obtained one-dimensional feature vectors were concatenated and used for prediction. Differently, channel fusion that first concatenated these feature maps in a channel-level was adopted in our study.…”

“… 21 and Song et al. 20 separately constructed a CNN network based on 54 and 95 patients, respectively, using CEM images to discriminate benign and malignant breast lesions. However, these small-sample and single-centre studies cannot learn the diverse features of images and lack a test of generalisation performance.…”

“…Recently, Song et al. 20 attempted to solve the above challenges by constructing a deep learning model trained on multimodal breast ultrasound images of 634 patients across two hospitals and prospectively tested 141 patients. However, false positives may exist in ultrasound images owing to the strong subjectivity of manual operation.…”

“… 14 , 15 , 16 , 17 , 18 Studies have attempted to predict benign and malignant breast lesions by combining CEM images and convolutional neural networks. 19 , 20 , 21 However, small-sample and single-centre datasets cannot adequately learn diverse image features and meet the requirements for generalised performance testing. In addition, the form based on manual segmentation is time-consuming and labor-intensive, which is not conducive to clinical translation.…”

Predicting Malignancy of Breast Imaging Findings Using Quantitative Analysis of Contrast-Enhanced Mammography (CEM)

“…The few early quantitative studies that have been performed using CEM have focused primarily on investigating the associations between automatically extracted CEM features and lesion malignancy or benignity [ 24 , 25 , 26 , 27 , 28 , 29 ], and between CEM features and the molecular subtype of invasive disease [ 30 ]. Most of these studies have reported promising results utilizing machine learning [ 24 , 26 , 27 , 28 ] or deep learning [ 25 , 29 ] techniques to predict tumor pathology features from CEM imaging features; however, these techniques are often difficult to interpret biologically or morphologically, and thus the underlying causes of imaging features often remain unclear.…”