Assessment of video frame interpolation network to generate digital breast tomosynthesis projections

Costa, Arthur C.; Vimieiro, Rodrigo de Barros; Borges, Lucas R.; Barufaldi, Bruno; Maidment, Andrew D. A.; Vieira, Marcelo A. C.

doi:10.1117/12.2625748

Cited by 2 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our preliminary study, we have used multiple tests to overcome the increase in radiation dose with interpolated images, but there is still work to do to improve the quality of interpolation and its limitations regarding alignment of structures. In previous work, 12 the approach was tested in DBT projections with different acquisition characteristics from the images used in this current work, and in that case, the results showed greater visual quality with the original reconstruction. Further investigation is needed to verify the implications of the image characteristics in the interpolation process and assess different acquisition geometries.…”

Section: Discussionmentioning

confidence: 71%

“…10,11 In our previous work, a pilot study evaluated how the addition and replacement of some DBT projections by VFI network interpolated images would interfere in reconstructed slices. 12 In the present work, we adapt and trained a residual refinement interpolation neural network (RRIN), 13 commonly used in computer vision applications of video frame interpolation, to generate synthetic DBTMI projections from pairs of original projections. The objective is to compensate for the increase in exposure due to the use of the MI sensor, replacing some of the original projections with synthetic images.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Assessment of projection interpolation to compensate for the increased radiation dose in DBTMI

Costa

Bjerkén

Dustler

et al. 2023

Medical Imaging 2023: Physics of Medical Imaging

View full text Add to dashboard Cite

The combination of digital breast tomosynthesis (DBT) with other imaging modalities has been investigated in order to improve the detection and diagnosis of breast cancer. Mechanical Imaging (MI) measures the stress over the surface of the compressed breast, using a pressure sensor, during radiographic examination and its response has shown a correlation with the presence of malignant lesions. Thus, the combination of DBT and MI (DBTMI) has shown potential to reduce false positive results in breast cancer screening. However, compared to the conventional DBT exam, the presence of the MI sensor during mammographic image acquisition may cause a slight increase in the radiation dose. This work presents a proposal to reduce the radiation dose in DBTMI exams by removing some projections from the original set and replacing them with synthetic projections generated by a video frame interpolation (VFI) neural network. We compared several DBTMI acquisition arrangements, considering the removal of 16% of the original projections, using a deformable physical breast phantom, and evaluated the quality of the reconstructed images based on the Normalized Root Mean Squared Error (NRMSE). The results showed that, for some arrangements, the slices reconstructed with the addition of synthetic DBTMI projections presented better quality than when they were reconstructed with the reduced set of projections. Further studies must be carried out to optimize the interpolation approach.

show abstract

Section: Discussionmentioning

confidence: 71%

Section: Introductionmentioning

confidence: 99%

Assessment of projection interpolation to compensate for the increased radiation dose in DBTMI

Costa

Bjerkén

Dustler

et al. 2023

Medical Imaging 2023: Physics of Medical Imaging

View full text Add to dashboard Cite

show abstract

“…In addition to this reassuring observation, our lab has been investigating options to reduce the radiation dose and compensate for a potential increase in AGD. In our ongoing project, in collaboration with the University of São Paulo, São Carlos Campus, Brazil, Costa et al 17 has trained a convolutional neural network to synthesize DBT projections by interpolation from neighboring projection, which enables to reduce the radiation dose without compromising image quality. An alternative option would be to explore MI sensors without metallic elements, which would not additionally attenuate x-rays during simultaneous acquisition.…”

Section: Discussionmentioning

confidence: 99%

Dose evaluation of simultaneous breast radiography and mechanical imaging

Bjerkén

Tomic

Bernhardsson

et al. 2023

Medical Imaging 2023: Physics of Medical Imaging

View full text Add to dashboard Cite

This study investigates the impact in terms of radiation dose when performing simultaneous digital breast tomosynthesis (DBT) and mechanical imaging (MI) -DBTMI. DBTMI has demonstrated the potential to increase specificity of cancer detection, and reduce unnecessary biopsies, as compared to digital mammography (DM) screening. The presence of the MI sensor during simultaneous image acquisition may increase the radiation dose when automatic exposure control is used. In this project, a radiation dose study was conducted on clinically available breast imaging systems with and without the MI sensor. We have investigated three approaches to analyse the dose increase in DBTMI, using (i) the estimates of average glandular dose (AGD) reported in DICOM headers of radiography images; (ii) AGD measured by a conventional dosemeter; and (iii) AGD measured by optically stimulated luminescence using NaCl pellets. The relative increase in AGD estimated from DICOM headers when using the MI sensor was on average 10.7% and 12.4%, for DM and DBT measurements, respectively. The relative increase in AGD using the conventional dosemeter was 11.2% in DM mode and 12.2% in DBT mode. The relative increase in AGD using NaCl pellets was 14.6% in DM mode. Our measurements suggest that the use of simultaneous breast radiography and MI increases the AGD by 13% on average. The increase in dose is still below the acceptable values in mammography screening recommended by the European Guidelines.

show abstract

Assessment of video frame interpolation network to generate digital breast tomosynthesis projections

Cited by 2 publications

References 13 publications

Assessment of projection interpolation to compensate for the increased radiation dose in DBTMI

Assessment of projection interpolation to compensate for the increased radiation dose in DBTMI

Dose evaluation of simultaneous breast radiography and mechanical imaging

Contact Info

Product

Resources

About