Synthesis of Mammogram From Digital Breast Tomosynthesis Using Deep Convolutional Neural Network With Gradient Guided cGANs

Jiang, Gongfa; Wei, Jun; Xu, Yuesheng; He, Zhiwei; Zeng, Hui; Wu, Jiefang; Qin, Genggeng; Chen, Weiguo; Lu, Yao

doi:10.1109/tmi.2021.3071544

Cited by 28 publications

(20 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Software options include modeling source blurring in the reconstruction, which has the potential to improve image quality for continuous motion acquisition DBT systems 63,64 and could be applied to the reconstruction steps used in SM algorithms. Improved SM algorithms 65 and methods utilizing artificial intelligence 66 are likely to improve lesion detectability in SM images in the future. DBT reconstruction and SM algorithms may make assumptions on the image content expected to be present in the breast.…”

Section: Discussionmentioning

confidence: 99%

The impact on lesion detection via a multi‐vendor study: A phantom‐based comparison of digital mammography, digital breast tomosynthesis, and synthetic mammography

et al. 2021

View full text Add to dashboard Cite

The aim of this study is to perform a test object-based comparison of the imaging performance of digital mammography (DM), digital breast tomosynthesis (DBT), and synthetic mammography (SM). Methods: Two test objects were used, the CDMAM and the L1-structured phantom. Small-detail detectability was assessed using CDMAM and the microcalcification simulating specks in the L1-structured background. Detection of spiculated and non-spiculated mass-like objects was assessed using the L1 phantom. Six different systems were included: Amulet Innovality (Fujifilm), Senographe Pristina (GEHC), 3Dimensions (Hologic), Giotto Class (IMS), Clarity 2D/3D (Planmed), and Mammomat Revelation (Siemens). Images were acquired under automatic exposure control (AEC) and at adjusted levels of AEC/2 and 2 × AEC level. Threshold gold thickness (T tr ) was established for the 0.13-mm-diameter CDMAM discs. Threshold diameters for the calcifications (d tr_c ), the spiculated masses (d tr_sm ), and for the non-spiculated masses (d tr_nsm ) were established. The threshold condition was defined as the thickness or diameter for a 62.5% correct score. Results: T tr for DM was generally superior to DBT, which in turn was superior to SM, but for most systems, these differences between modes were not significant. For L1, no significant differences in d tr_c were found between DM and DBT. The increase in d tr_c from DM to SM at AEC dose was 1%, 19%, 11%, 14%, 46%, and 27% for the Fujifilm, GEHC, Hologic, IMS, Planmed, and Siemens, respectively, indicating significantly poorer performance for all vendors except for Fujifilm, Hologic, and IMS. For both mass types, DBT performed better than SM, while SM showed no significant difference with DM (except for Fujifilm spiculated masses). The dose had an impact on small-detail detectability for both phantoms but did not influence the detection of either mass type. Conclusions: Both phantoms indicated potentially reduced small-detail detectability for SM versus DM and DBT and should therefore not be used in stand-alone mode. The L1 phantom demonstrated no significant difference in microcalcification detection between DM and DBT and also demonstrated the superiority of DBT, compared to DM for mass detection, for all six systems.

show abstract

Section: Discussionmentioning

confidence: 99%

The impact on lesion detection via a multi‐vendor study: A phantom‐based comparison of digital mammography, digital breast tomosynthesis, and synthetic mammography

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The latest burst began in 2019, including the following keywords: "convolutional neural network", "magnetic resonance image" and "histopathological image". With the popularization of artificial intelligence and the renewal of deep learning algorithm, convolutional neural network has become the most important algorithm for processing medical images, especially in radiology and histopathology [58][59][60]. However, deep learning-based AI has been queried by both clinician and pathologists for the lack of good interpretability, hindering the clinical application of AI model [61][62][63].…”

Section: Abstract Co-occurrence Analysismentioning

confidence: 99%

Global research trends and foci of artificial intelligence-based tumor pathology: a scientometric study

Shen

et al. 2022

J Transl Med

View full text Add to dashboard Cite

Background With the development of digital pathology and the renewal of deep learning algorithm, artificial intelligence (AI) is widely applied in tumor pathology. Previous researches have demonstrated that AI-based tumor pathology may help to solve the challenges faced by traditional pathology. This technology has attracted the attention of scholars in many fields and a large amount of articles have been published. This study mainly summarizes the knowledge structure of AI-based tumor pathology through bibliometric analysis, and discusses the potential research trends and foci. Methods Publications related to AI-based tumor pathology from 1999 to 2021 were selected from Web of Science Core Collection. VOSviewer and Citespace were mainly used to perform and visualize co-authorship, co-citation, and co-occurrence analysis of countries, institutions, authors, references and keywords in this field. Results A total of 2753 papers were included. The papers on AI-based tumor pathology research had been continuously increased since 1999. The United States made the largest contribution in this field, in terms of publications (1138, 41.34%), H-index (85) and total citations (35,539 times). We identified the most productive institution and author were Harvard Medical School and Madabhushi Anant, while Jemal Ahmedin was the most co-cited author. Scientific Reports was the most prominent journal and after analysis, Lecture Notes in Computer Science was the journal with highest total link strength. According to the result of references and keywords analysis, “breast cancer histopathology” “convolutional neural network” and “histopathological image” were identified as the major future research foci. Conclusions AI-based tumor pathology is in the stage of vigorous development and has a bright prospect. International transboundary cooperation among countries and institutions should be strengthened in the future. It is foreseeable that more research foci will be lied in the interpretability of deep learning-based model and the development of multi-modal fusion model.

show abstract

“…In the accelerating evolution of deep learning, the transition from a convolutional neural network [ 24 ] to a generative adversarial network (GAN) [ 25 , 26 ] has contributed to digital tomosynthesis imaging [ 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 ]. Prior studies reported that GANs are particularly useful for reducing metal artifacts [ 29 , 30 ] and noise [ 27 ] and are expected to contribute to improvements in image quality processes to reduce the exposure dose.…”

Section: Introductionmentioning

confidence: 99%

“…Thus, application of deep learning can be used to improve image quality further and reduce the radiation-dose. In the DBT imaging field, recent reports have detailed the detection of masses and the image quality improvement process that introduces deep learning [ 28 , 31 , 32 , 33 , 34 , 35 ]. With regard to image quality improvement processing that uses conditional GAN (cGAN, or pix2pix) [ 25 ], “pix2pix,” which approximates the object image to the referenced image using the concept of an adversarial network using “generator” and “discriminator,” has been shown to be useful for noise reduction.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of a Generative Adversarial Network to Improve Image Quality and Reduce Radiation-Dose during Digital Breast Tomosynthesis

et al. 2022

View full text Add to dashboard Cite

In this study, we evaluated the improvement of image quality in digital breast tomosynthesis under low-radiation dose conditions of pre-reconstruction processing using conditional generative adversarial networks [cGAN (pix2pix)]. Pix2pix pre-reconstruction processing with filtered back projection (FBP) was compared with and without multiscale bilateral filtering (MSBF) during pre-reconstruction processing. Noise reduction and preserve contrast rates were compared using full width at half-maximum (FWHM), contrast-to-noise ratio (CNR), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) in the in-focus plane using a BR3D phantom at various radiation doses [reference-dose (automatic exposure control reference dose: AECrd), 50% and 75% reduction of AECrd] and phantom thicknesses (40 mm, 50 mm, and 60 mm). The overall performance of pix2pix pre-reconstruction processing was effective in terms of FWHM, PSNR, and SSIM. At ~50% radiation-dose reduction, FWHM yielded good results independently of the microcalcification size used in the BR3D phantom, and good noise reduction and preserved contrast. PSNR results showed that pix2pix pre-reconstruction processing represented the minimum in the error with reference FBP images at an approximately 50% reduction in radiation-dose. SSIM analysis indicated that pix2pix pre-reconstruction processing yielded superior similarity when compared with and without MSBF pre-reconstruction processing at ~50% radiation-dose reduction, with features most similar to the reference FBP images. Thus, pix2pix pre-reconstruction processing is promising for reducing noise with preserve contrast and radiation-dose reduction in clinical practice.

show abstract

Synthesis of Mammogram From Digital Breast Tomosynthesis Using Deep Convolutional Neural Network With Gradient Guided cGANs

Cited by 28 publications

References 33 publications

The impact on lesion detection via a multi‐vendor study: A phantom‐based comparison of digital mammography, digital breast tomosynthesis, and synthetic mammography

The impact on lesion detection via a multi‐vendor study: A phantom‐based comparison of digital mammography, digital breast tomosynthesis, and synthetic mammography

Global research trends and foci of artificial intelligence-based tumor pathology: a scientometric study

Evaluation of a Generative Adversarial Network to Improve Image Quality and Reduce Radiation-Dose during Digital Breast Tomosynthesis

Contact Info

Product

Resources

About