Dr. Sarah E. Divel Profile

Dr. Sarah E. Divel

Engineer at RapidAI

SPIE Involvement:

Author

Publications (7)

Proceedings Article | 15 February 2021 Poster + Presentation + Paper

CNN-based CT denoising with an accurate image domain noise insertion technique

Byeongjoon Kim, Sarah Divel, Norbert Pelc, Jongduk Baek

Proceedings Volume 11595, 1159545 (2021) https://doi.org/10.1117/12.2580997

KEYWORDS: Denoising, Computed tomography, Modulation transfer functions, Image quality, Statistical analysis, Optical filters, Image filtering, Gaussian filters, Error analysis, Convolutional neural networks

Read Abstract +

Proceedings Article | 16 March 2020 Paper

Simulation of contrast agent dynamics in digital brain phantom for CT perfusion optimization

Sarah Divel, Søren Christensen, Maarten Lansberg, Norbert Pelc

Proceedings Volume 11312, 1131217 (2020) https://doi.org/10.1117/12.2549741

KEYWORDS: Brain, Tissues, Arteries, Veins, Iodine, Blood, Neuroimaging, Cerebral blood flow, Computed tomography

Read Abstract +

Proceedings Article | 7 March 2019 Presentation + Paper

Virtual clinical trial for task-based evaluation of a deep learning synthetic mammography algorithm

Andreu Badal, Kenny Cha, Sarah Divel, Christian Graff, Rongping Zeng, Aldo Badano

Proceedings Volume 10948, 109480O (2019) https://doi.org/10.1117/12.2513062

KEYWORDS: Digital breast tomosynthesis, Mammography, Image processing, Denoising, Clinical trials, Computer simulations, Breast, Gaussian filters, Medical imaging, Monte Carlo methods

Read Abstract +

Image processing algorithms based on deep learning techniques are being developed for a wide range of medical applications. Processed medical images are typically evaluated with the same kind of image similarity metrics used for natural scenes, disregarding the medical task for which the images are intended. We propose a com- putational framework to estimate the clinical performance of image processing algorithms using virtual clinical trials. The proposed framework may provide an alternative method for regulatory evaluation of non-linear image processing algorithms. To illustrate this application of virtual clinical trials, we evaluated three algorithms to compute synthetic mammograms from digital breast tomosynthesis (DBT) scans based on convolutional neural networks previously used for denoising low dose computed tomography scans. The inputs to the networks were one or more noisy DBT projections, and the networks were trained to minimize the difference between the output and the corresponding high dose mammogram. DBT and mammography images simulated with the Monte Carlo code MC-GPU using realistic breast phantoms were used for network training and validation. The denoising algorithms were tested in a virtual clinical trial by generating 3000 synthetic mammograms from the public VICTRE dataset of simulated DBT scans. The detectability of a calcification cluster and a spiculated mass present in the images was calculated using an ensemble of 30 computational channelized Hotelling observers. The signal detectability results, which took into account anatomic and image reader variability, showed that the visibility of the mass was not affected by the post-processing algorithm, but that the resulting slight blurring of the images severely impacted the visibility of the calcification cluster. The evaluation of the algorithms using the pixel-based metrics peak signal to noise ratio and structural similarity in image patches was not able to predict the reduction in performance in the detectability of calcifications. These two metrics are computed over the whole image and do not consider any particular task, and might not be adequate to estimate the diagnostic performance of the post-processed images.

Proceedings Article | 1 March 2019 Presentation + Paper

Image-domain insertion of spatially correlated, locally varying noise in CT images

Sarah Divel, Norbert Pelc

Proceedings Volume 10948, 1094821 (2019) https://doi.org/10.1117/12.2512453

KEYWORDS: Fluctuations and noise, Brain, Computed tomography, Neuroimaging, Image filtering, Image analysis, Computer simulations, Image quality, Fourier transforms, CT reconstruction

Read Abstract +

Proceedings Article | 9 March 2018 Presentation + Paper

Can image-domain filtering of FBP CT reconstructions match low-contrast performance of iterative reconstructions?

Sarah Divel, Scott Hsieh, Jia Wang, Norbert Pelc

Proceedings Volume 10573, 1057314 (2018) https://doi.org/10.1117/12.2292599

KEYWORDS: Image filtering, Gaussian filters, CT reconstruction, 3D image reconstruction, Image quality, Linear filtering, Scanners, 3D image processing

Read Abstract +

Proceedings Article | 9 March 2017 Paper

Method for decreasing CT simulation time of complex phantoms and systems through separation of material specific projection data

Sarah Divel, Soren Christensen, Max Wintermark, Maarten Lansberg, Norbert Pelc

Proceedings Volume 10132, 1013259 (2017) https://doi.org/10.1117/12.2254076

KEYWORDS: Computer simulations, Brain, Complex systems, Systems modeling, Neuroimaging, Computing systems, Signal attenuation, Computed tomography, Algorithm development, Arteries, Veins, Tissues

Read Abstract +

Proceedings Article | 22 March 2016 Paper

Development of a realistic, dynamic digital brain phantom for CT perfusion validation

Sarah Divel, W. Paul Segars, Soren Christensen, Max Wintermark, Maarten Lansberg, Norbert Pelc

Proceedings Volume 9783, 97830Y (2016) https://doi.org/10.1117/12.2214997

KEYWORDS: Brain, Computer simulations, Computed tomography, Blood, Neuroimaging, Data modeling, Cerebral blood flow, Diagnostics, Signal attenuation, Tissues, 3D modeling, Iodine, Arteries

Read Abstract +

Showing 5 of 7 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years