1
|
Kim G, Baek J. Power-law spectrum-based objective function to train a generative adversarial network with transfer learning for the synthetic breast CT image. Phys Med Biol 2023; 68:205007. [PMID: 37722388 DOI: 10.1088/1361-6560/acfadf] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/02/2023] [Accepted: 09/18/2023] [Indexed: 09/20/2023]
Abstract
Objective.This paper proposes a new objective function to improve the quality of synthesized breast CT images generated by the GAN and compares the GAN performances on transfer learning datasets from different image domains.Approach.The proposed objective function, named beta loss function, is based on the fact that x-ray-based breast images follow the power-law spectrum. Accordingly, the exponent of the power-law spectrum (beta value) for breast CT images is approximately two. The beta loss function is defined in terms of L1 distance between the beta value of synthetic images and validation samples. To compare the GAN performances for transfer learning datasets from different image domains, ImageNet and anatomical noise images are used in the transfer learning dataset. We employ styleGAN2 as the backbone network and add the proposed beta loss function. The patient-derived breast CT dataset is used as the training and validation dataset; 7355 and 212 images are used for network training and validation, respectively. We use the beta value evaluation and Fréchet inception distance (FID) score for quantitative evaluation.Main results.For qualitative assessment, we attempt to replicate the images from the validation dataset using the trained GAN. Our results show that the proposed beta loss function achieves a more similar beta value to real images and a lower FID score. Moreover, we observe that the GAN pretrained with anatomical noise images achieves better equality than ImageNet for beta value evaluation and FID score. Finally, the beta loss function with anatomical noise as the transfer learning dataset achieves the lowest FID score.Significance.Overall, the GAN using the proposed beta loss function with anatomical noise images as the transfer learning dataset provides the lowest FID score among all tested cases. Hence, this work has implications for developing GAN-based breast image synthesis methods for medical imaging applications.
Collapse
Affiliation(s)
- Gihun Kim
- School of Integrated Technology, Yonsei University, Republic of Korea
| | - Jongduk Baek
- Department of Artificial Intelligence, Yonsei University, Republic of Korea
- Baruenex Imaging, Republic of Korea
| |
Collapse
|
2
|
Liu Z, Wolfe S, Yu Z, Laforest R, Mhlanga JC, Fraum TJ, Itani M, Dehdashti F, Siegel BA, Jha AK. Observer-study-based approaches to quantitatively evaluate the realism of synthetic medical images. Phys Med Biol 2023; 68:10.1088/1361-6560/acc0ce. [PMID: 36863028 PMCID: PMC10411234 DOI: 10.1088/1361-6560/acc0ce] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/17/2022] [Accepted: 03/02/2023] [Indexed: 03/04/2023]
Abstract
Objective.Synthetic images generated by simulation studies have a well-recognized role in developing and evaluating imaging systems and methods. However, for clinically relevant development and evaluation, the synthetic images must be clinically realistic and, ideally, have the same distribution as that of clinical images. Thus, mechanisms that can quantitatively evaluate this clinical realism and, ideally, the similarity in distributions of the real and synthetic images, are much needed.Approach.We investigated two observer-study-based approaches to quantitatively evaluate the clinical realism of synthetic images. In the first approach, we presented a theoretical formalism for the use of an ideal-observer study to quantitatively evaluate the similarity in distributions between the real and synthetic images. This theoretical formalism provides a direct relationship between the area under the receiver operating characteristic curve, AUC, for an ideal observer and the distributions of real and synthetic images. The second approach is based on the use of expert-human-observer studies to quantitatively evaluate the realism of synthetic images. In this approach, we developed a web-based software to conduct two-alternative forced-choice (2-AFC) experiments with expert human observers. The usability of this software was evaluated by conducting a system usability scale (SUS) survey with seven expert human readers and five observer-study designers. Further, we demonstrated the application of this software to evaluate a stochastic and physics-based image-synthesis technique for oncologic positron emission tomography (PET). In this evaluation, the 2-AFC study with our software was performed by six expert human readers, who were highly experienced in reading PET scans, with years of expertise ranging from 7 to 40 years (median: 12 years, average: 20.4 years).Main results.In the ideal-observer-study-based approach, we theoretically demonstrated that the AUC for an ideal observer can be expressed, to an excellent approximation, by the Bhattacharyya distance between the distributions of the real and synthetic images. This relationship shows that a decrease in the ideal-observer AUC indicates a decrease in the distance between the two image distributions. Moreover, a lower bound of ideal-observer AUC = 0.5 implies that the distributions of synthetic and real images exactly match. For the expert-human-observer-study-based approach, our software for performing the 2-AFC experiments is available athttps://apps.mir.wustl.edu/twoafc. Results from the SUS survey demonstrate that the web application is very user friendly and accessible. As a secondary finding, evaluation of a stochastic and physics-based PET image-synthesis technique using our software showed that expert human readers had limited ability to distinguish the real images from the synthetic images.Significance.This work addresses the important need for mechanisms to quantitatively evaluate the clinical realism of synthetic images. The mathematical treatment in this paper shows that quantifying the similarity in the distribution of real and synthetic images is theoretically possible by using an ideal-observer-study-based approach. Our developed software provides a platform for designing and performing 2-AFC experiments with human observers in a highly accessible, efficient, and secure manner. Additionally, our results on the evaluation of the stochastic and physics-based image-synthesis technique motivate the application of this technique to develop and evaluate a wide array of PET imaging methods.
Collapse
Affiliation(s)
- Ziping Liu
- Department of Biomedical Engineering, Washington University, St. Louis, MO 63130, United States of America
| | - Scott Wolfe
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO 63110, United States of America
| | - Zitong Yu
- Department of Biomedical Engineering, Washington University, St. Louis, MO 63130, United States of America
| | - Richard Laforest
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO 63110, United States of America
- Alvin J. Siteman Cancer Center, Washington University School of Medicine, St. Louis, MO 63110, United States of America
| | - Joyce C Mhlanga
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO 63110, United States of America
| | - Tyler J Fraum
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO 63110, United States of America
- Alvin J. Siteman Cancer Center, Washington University School of Medicine, St. Louis, MO 63110, United States of America
| | - Malak Itani
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO 63110, United States of America
| | - Farrokh Dehdashti
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO 63110, United States of America
- Alvin J. Siteman Cancer Center, Washington University School of Medicine, St. Louis, MO 63110, United States of America
| | - Barry A Siegel
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO 63110, United States of America
- Alvin J. Siteman Cancer Center, Washington University School of Medicine, St. Louis, MO 63110, United States of America
| | - Abhinav K Jha
- Department of Biomedical Engineering, Washington University, St. Louis, MO 63130, United States of America
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO 63110, United States of America
- Alvin J. Siteman Cancer Center, Washington University School of Medicine, St. Louis, MO 63110, United States of America
| |
Collapse
|
3
|
Physical and digital phantoms for 2D and 3D x-ray breast imaging: Review on the state-of-the-art and future prospects. Radiat Phys Chem Oxf Engl 1993 2022. [DOI: 10.1016/j.radphyschem.2022.110715] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/14/2022]
|
4
|
Sarno A, Mettivier G, di Franco F, Varallo A, Bliznakova K, Hernandez AM, Boone JM, Russo P. Dataset of patient-derived digital breast phantoms for in silico studies in breast computed tomography, digital breast tomosynthesis, and digital mammography. Med Phys 2021; 48:2682-2693. [PMID: 33683711 DOI: 10.1002/mp.14826] [Citation(s) in RCA: 16] [Impact Index Per Article: 5.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/05/2020] [Revised: 02/22/2021] [Accepted: 02/28/2021] [Indexed: 01/10/2023] Open
Abstract
PURPOSE To present a dataset of computational digital breast phantoms derived from high-resolution three-dimensional (3D) clinical breast images for the use in virtual clinical trials in two-dimensional (2D) and 3D x-ray breast imaging. ACQUISITION AND VALIDATION METHODS Uncompressed computational breast phantoms for investigations in dedicated breast CT (BCT) were derived from 150 clinical 3D breast images acquired via a BCT scanner at UC Davis (California, USA). Each image voxel was classified in one out of the four main materials presented in the field of view: fibroglandular tissue, adipose tissue, skin tissue, and air. For the image classification, a semi-automatic software was developed. The semi-automatic classification was compared via manual glandular classification performed by two researchers. A total of 60 compressed computational phantoms for virtual clinical trials in digital mammography (DM) and digital breast tomosynthesis (DBT) were obtained from the corresponding uncompressed phantoms via a software algorithm simulating the compression and the elastic deformation of the breast, using the tissue's elastic coefficient. This process was evaluated in terms of glandular fraction modification introduced by the compression procedure. The generated cohort of 150 uncompressed computational breast phantoms presented a mean value of the glandular fraction by mass of 12.3%; the average diameter of the breast evaluated at the center of mass was 105 mm. Despite the slight differences between the two manual segmentations, the resulting glandular tissue segmentation did not consistently differ from that obtained via the semi-automatic classification. The difference between the glandular fraction by mass before and after the compression was 2.1% on average. The 60 compressed phantoms presented an average glandular fraction by mass of 12.1% and an average compressed thickness of 61 mm. DATA FORMAT AND ACCESS The generated digital breast phantoms are stored in DICOM files. Image voxels can present one out of four values representing the different classified materials: 0 for the air, 1 for the adipose tissue, 2 for the glandular tissue, and 3 for the skin tissue. The generated computational phantoms datasets were stored in the Zenodo public repository for research purposes (http://doi.org/10.5281/zenodo.4529852, http://doi.org/10.5281/zenodo.4515360). POTENTIAL APPLICATIONS The dataset developed within the INFN AGATA project will be used for developing a platform for virtual clinical trials in x-ray breast imaging and dosimetry. In addition, they will represent a valid support for introducing new breast models for dose estimates in 2D and 3D x-ray breast imaging and as models for manufacturing anthropomorphic physical phantoms.
Collapse
Affiliation(s)
| | - Giovanni Mettivier
- INFN Sezione di Napoli, Naples, Italy.,Dipartimento di Fisica "Ettore Pancini", Università di Napoli Federico II, Naples, Italy
| | - Francesca di Franco
- INFN Sezione di Napoli, Naples, Italy.,Dipartimento di Fisica "Ettore Pancini", Università di Napoli Federico II, Naples, Italy.,Léon Bérard Cancer Center, University of Lyon & CREATiS, University of Lyon, CNRS, Lyon, France
| | - Antonio Varallo
- Dipartimento di Fisica "Ettore Pancini", Università di Napoli Federico II, Naples, Italy
| | - Kristina Bliznakova
- Department of Medical Equipment, Electronic and Information Technologies in Healthcare, Medical University of Varna, Varna, Bulgaria
| | - Andrew M Hernandez
- Department of Radiology, University of California Davis, Sacramento, CA, USA
| | - John M Boone
- Department of Radiology, University of California Davis, Sacramento, CA, USA
| | - Paolo Russo
- INFN Sezione di Napoli, Naples, Italy.,Dipartimento di Fisica "Ettore Pancini", Università di Napoli Federico II, Naples, Italy
| |
Collapse
|
5
|
Chang TY, Lai KJ, Tu CY, Wu J. Three-layer heterogeneous mammographic phantoms for Monte Carlo simulation of normalized glandular dose coefficients in mammography. Sci Rep 2020; 10:2234. [PMID: 32042071 PMCID: PMC7010737 DOI: 10.1038/s41598-020-59317-4] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/09/2019] [Accepted: 01/27/2020] [Indexed: 12/15/2022] Open
Abstract
Normalized glandular dose (DgN) coefficients obtained using homogeneous breast phantoms are commonly used in breast dosimetry for mammography. However, glandular tissue is heterogeneously distributed in the breast. This study aimed to construct three-layer heterogeneous mammographic phantoms (THEPs) to examine the effect of glandular distribution on DgN coefficient. Each layer of THEPs was set to 25%, 50%, or 75% glandular fraction to emulate heterogeneous glandular distribution. Monte Carlo simulation was performed to attain mean glandular dose (MGD) and air kerma at 22-36 kVp and W/Al, W/Rh, and W/Ag target-filter combinations. The heterogeneous DgN coefficient was calculated as functions of the mean glandular fraction (MGF), breast thickness, tube voltage, and half-value layer. At 50% MGF, the heterogeneous DgN coefficients for W/Al, W/Rh, and W/Ag differed by 40.3%, 36.7%, and 31.2%. At 9-cm breast thickness, the DgN values of superior and inferior glandular distributions were 25.4% higher and 29.2% lower than those of uniform distribution. The proposed THEPs can be integrated with conventional breast dosimetry to consider the heterogeneous glandular distribution in clinical practice.
Collapse
Affiliation(s)
- Tien-Yu Chang
- Department of Radiology, Cheng Hsin General Hospital, Taipei, Taiwan
| | - Kuan-Jen Lai
- Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, Taipei, Taiwan
| | - Chun-Yuan Tu
- Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, Taipei, Taiwan
- Department of Radiology, Mackay Memorial Hospital, Taipei, Taiwan
| | - Jay Wu
- Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, Taipei, Taiwan.
| |
Collapse
|
6
|
Kainz W, Neufeld E, Bolch WE, Graff CG, Kim CH, Kuster N, Lloyd B, Morrison T, Segars P, Yeom YS, Zankl M, Xu XG, Tsui BMW. Advances in Computational Human Phantoms and Their Applications in Biomedical Engineering - A Topical Review. IEEE TRANSACTIONS ON RADIATION AND PLASMA MEDICAL SCIENCES 2019; 3:1-23. [PMID: 30740582 PMCID: PMC6362464 DOI: 10.1109/trpms.2018.2883437] [Citation(s) in RCA: 42] [Impact Index Per Article: 8.4] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/05/2023]
Abstract
Over the past decades, significant improvements have been made in the field of computational human phantoms (CHPs) and their applications in biomedical engineering. Their sophistication has dramatically increased. The very first CHPs were composed of simple geometric volumes, e.g., cylinders and spheres, while current CHPs have a high resolution, cover a substantial range of the patient population, have high anatomical accuracy, are poseable, morphable, and are augmented with various details to perform functionalized computations. Advances in imaging techniques and semi-automated segmentation tools allow fast and personalized development of CHPs. These advances open the door to quickly develop personalized CHPs, inherently including the disease of the patient. Because many of these CHPs are increasingly providing data for regulatory submissions of various medical devices, the validity, anatomical accuracy, and availability to cover the entire patient population is of utmost importance. The article is organized into two main sections: the first section reviews the different modeling techniques used to create CHPs, whereas the second section discusses various applications of CHPs in biomedical engineering. Each topic gives an overview, a brief history, recent developments, and an outlook into the future.
Collapse
Affiliation(s)
- Wolfgang Kainz
- Food and Drug Administration (FDA), Center for Devices and Radiological Health (CDRH), Silver Spring, MD 20993 USA
| | - Esra Neufeld
- Foundation for Research on Information Technologies in Society (IT'IS), Zurich, Switzerland
| | | | - Christian G Graff
- Food and Drug Administration (FDA), Center for Devices and Radiological Health (CDRH), Silver Spring, MD 20993 USA
| | | | - Niels Kuster
- Swiss Federal Institute of Technology, ETH Zürich, and the Foundation for Research on Information Technologies in Society (IT'IS), Zürich, Switzerland
| | - Bryn Lloyd
- Foundation for Research on Information Technologies in Society (IT'IS), Zurich, Switzerland
| | - Tina Morrison
- Food and Drug Administration (FDA), Center for Devices and Radiological Health (CDRH), Silver Spring, MD 20993 USA
| | | | | | - Maria Zankl
- Helmholtz Zentrum München German Research Center for Environmental Health, Munich, Germany
| | - X George Xu
- Rensselaer Polytechnic Institute, Troy, NY, USA
| | | |
Collapse
|
7
|
Badano A, Graff CG, Badal A, Sharma D, Zeng R, Samuelson FW, Glick SJ, Myers KJ. Evaluation of Digital Breast Tomosynthesis as Replacement of Full-Field Digital Mammography Using an In Silico Imaging Trial. JAMA Netw Open 2018; 1:e185474. [PMID: 30646401 PMCID: PMC6324392 DOI: 10.1001/jamanetworkopen.2018.5474] [Citation(s) in RCA: 102] [Impact Index Per Article: 17.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/14/2022] Open
Abstract
IMPORTANCE Expensive and lengthy clinical trials can delay regulatory evaluation of innovative technologies, affecting patient access to high-quality medical products. Simulation is increasingly being used in product development but rarely in regulatory applications. OBJECTIVES To conduct a computer-simulated imaging trial evaluating digital breast tomosynthesis (DBT) as a replacement for digital mammography (DM) and to compare the results with a comparative clinical trial. DESIGN, SETTING, AND PARTICIPANTS The simulated Virtual Imaging Clinical Trial for Regulatory Evaluation (VICTRE) trial was designed to replicate a clinical trial that used human patients and radiologists. Images obtained with in silico versions of DM and DBT systems via fast Monte Carlo x-ray transport were interpreted by a computational reader detecting the presence of lesions. A total of 2986 synthetic image-based virtual patients with breast sizes and radiographic densities representative of a screening population and compressed thicknesses from 3.5 to 6 cm were generated using an analytic approach in which anatomical structures are randomly created within a predefined breast volume and compressed in the craniocaudal orientation. A positive cohort contained a digitally inserted microcalcification cluster or spiculated mass. MAIN OUTCOMES AND MEASURES The trial end point was the difference in area under the receiver operating characteristic curve between modalities for lesion detection. The trial was sized for an SE of 0.01 in the change in area under the curve (AUC), half the uncertainty in the comparative clinical trial. RESULTS In this trial, computational readers analyzed 31 055 DM and 27 960 DBT cases from 2986 virtual patients with the following Breast Imaging Reporting and Data System densities: 286 (9.6%) extremely dense, 1200 (40.2%) heterogeneously dense, 1200 (40.2%) scattered fibroglandular densities, and 300 (10.0%) almost entirely fat. The mean (SE) change in AUC was 0.0587 (0.0062) (P < .001) in favor of DBT. The change in AUC was larger for masses (mean [SE], 0.0903 [0.008]) than for calcifications (mean [SE], 0.0268 [0.004]), which was consistent with the findings of the comparative trial (mean [SE], 0.065 [0.017] for masses and -0.047 [0.032] for calcifications). CONCLUSIONS AND RELEVANCE The results of the simulated VICTRE trial are consistent with the performance seen in the comparative trial. While further research is needed to assess the generalizability of these findings, in silico imaging trials represent a viable source of regulatory evidence for imaging devices.
Collapse
Affiliation(s)
- Aldo Badano
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland
| | - Christian G Graff
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland
| | - Andreu Badal
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland
| | - Diksha Sharma
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland
| | - Rongping Zeng
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland
| | - Frank W Samuelson
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland
| | - Stephen J Glick
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland
| | - Kyle J Myers
- Division of Imaging, Diagnostics, and Software Reliability, Office of Science and Engineering Laboratories, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland
| |
Collapse
|
8
|
Glick SJ, Ikejimba LC. Advances in digital and physical anthropomorphic breast phantoms for x-ray imaging. Med Phys 2018; 45:e870-e885. [DOI: 10.1002/mp.13110] [Citation(s) in RCA: 22] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/12/2017] [Revised: 06/05/2018] [Accepted: 06/10/2018] [Indexed: 01/27/2023] Open
Affiliation(s)
- Stephen J. Glick
- Division of Imaging, Diagnostics, and Software Reliability; Office of Science and Engineering Laboratories; Center for Devices and Radiological Health, Food and Drug Administration; Silver Spring MD 20993 USA
| | - Lynda C. Ikejimba
- Division of Imaging, Diagnostics, and Software Reliability; Office of Science and Engineering Laboratories; Center for Devices and Radiological Health, Food and Drug Administration; Silver Spring MD 20993 USA
| |
Collapse
|