Mr. Jason L. Granstedt Profile

Jason L. Granstedt

at Univ of Illinois

SPIE Involvement:

Author

Publications (9)

SPIE Journal Paper | 26 September 2023 Open Access

Approximating the Hotelling observer with autoencoder-learned efficient channels for binary signal detection tasks

Jason Granstedt, Weimin Zhou, Mark Anastasio

JMI, Vol. 10, Issue 05, 055501, (September 2023) https://doi.org/10.1117/12.10.1117/1.JMI.10.5.055501

KEYWORDS: Education and training, Signal detection, Imaging systems, Signal attenuation, Breast, Binary data, Information operations, Image restoration, Image processing, Signal processing

Read Abstract +

PurposeThe objective assessment of image quality (IQ) has been advocated for the analysis and optimization of medical imaging systems. One method of computing such IQ metrics is through a numerical observer. The Hotelling observer (HO) is the optimal linear observer, but conventional methods for obtaining the HO can become intractable due to large image sizes or insufficient data. Channelized methods are sometimes employed in such circumstances to approximate the HO. The performance of such channelized methods varies, with different methods obtaining superior performance to others depending on the imaging conditions and detection task. A channelized HO method using an AE is presented and implemented across several tasks to characterize its performance.ApproachThe process for training an AE is demonstrated to be equivalent to developing a set of channels for approximating the HO. The efficiency of the learned AE-channels is increased by modifying the conventional AE loss function to incorporate task-relevant information. Multiple binary detection tasks involving lumpy and breast phantom backgrounds across varying dataset sizes are considered to evaluate the performance of the proposed method and compare to current state-of-the-art channelized methods. Additionally, the ability of the channelized methods to generalize to images outside of the training dataset is investigated.ResultsAE-learned channels are demonstrated to have comparable performance with other state-of-the-art channel methods in the detection studies and superior performance in the generalization studies. Incorporating a cleaner estimate of the signal for the detection task is also demonstrated to significantly improve the performance of the proposed method, particularly in datasets with fewer images.ConclusionsAEs are demonstrated to be capable of learning efficient channels for the HO. The resulting significant increase in detection performance for small dataset sizes when incorporating a signal prior holds promising implications for future assessments of imaging technologies.

DOWNLOAD PAPER

Proceedings Article | 4 April 2022 Poster + Presentation + Paper

Learned Hotelling observers for use with multi-modal data

Jason Granstedt, Fu Li, Umberto Villa, Mark Anastasio

Proceedings Volume 12035, 1203512 (2022) https://doi.org/10.1117/12.2613138

KEYWORDS: Imaging systems, Signal detection, Signal to noise ratio, Data modeling, Stochastic processes, Information operations, Image quality, Image processing, Computed tomography, Targeting Task Performance metric, Ultrasound tomography, Machine learning, Artificial neural networks

Read Abstract +

SPIE Journal Paper | 16 November 2021 Open Access

Impact of deep learning-based image super-resolution on binary signal detection

Xiaohui Zhang, Varun Kelkar, Jason Granstedt, Hua Li, Mark Anastasio

JMI, Vol. 8, Issue 06, 065501, (November 2021) https://doi.org/10.1117/12.10.1117/1.JMI.8.6.065501

KEYWORDS: Signal detection, Lawrencium, Super resolution, Binary data, Information operations, Network architectures, Medical imaging, Surgery, Signal attenuation, Image quality

Read Abstract +

DOWNLOAD PAPER

Proceedings Article | 15 February 2021 Poster + Presentation + Paper

Task-based evaluation of deep image super-resolution in medical imaging

Varun Kelkar, Xiaohui Zhang, Jason Granstedt, Hua Li, Mark Anastasio

Proceedings Volume 11599, 115990X (2021) https://doi.org/10.1117/12.2582011

KEYWORDS: Medical imaging, Super resolution, Image analysis, Image quality, Signal to noise ratio, Quality measurement, Neural networks, Diagnostics

Read Abstract +

Proceedings Article | 15 February 2021 Poster + Presentation + Paper

Learning a projection operator onto the null space of a linear imaging operator

Joseph Kuo, Jason Granstedt, Umberto Villa, Mark Anastasio

Proceedings Volume 11595, 115953X (2021) https://doi.org/10.1117/12.2582263

KEYWORDS: Space operations, Reconstruction algorithms, Imaging systems, Magnetic resonance imaging, Linear algebra, Image restoration, Image resolution, Artificial neural networks

Read Abstract +

Proceedings Article | 15 February 2021 Presentation + Paper

Advancing the AmbientGAN for learning stochastic object models

Weimin Zhou, Sayantan Bhadra, Frank Brooks, Jason Granstedt, Hua Li, Mark Anastasio

Proceedings Volume 11599, 115990A (2021) https://doi.org/10.1117/12.2581687

KEYWORDS: Systems modeling, Stochastic processes, Imaging systems, 3D image processing, Gallium nitride, 3D metrology, Stereoscopy, Network architectures, Medical imaging, Magnetic resonance imaging

Read Abstract +

Proceedings Article | 15 February 2021 Presentation + Paper

SlabGAN: a method for generating efficient 3D anisotropic medical volumes using generative adversarial networks

Jason Granstedt, Varun Kelkar, Weimin Zhou, Mark Anastasio

Proceedings Volume 11596, 1159617 (2021) https://doi.org/10.1117/12.2581380

KEYWORDS: Gallium nitride, 3D image processing, Medical imaging, Magnetic resonance imaging, Super resolution, Stochastic processes, Neuroimaging, Medical image reconstruction, Imaging systems, Image restoration

Read Abstract +

Proceedings Article | 16 March 2020 Presentation + Paper

Learning efficient channels with a dual loss autoencoder

Jason Granstedt, Weimin Zhou, Mark Anastasio

Proceedings Volume 11316, 113160C (2020) https://doi.org/10.1117/12.2549363

KEYWORDS: Neural networks, Signal detection, Imaging systems, Signal to noise ratio, Breast, Performance modeling, Image quality

Read Abstract +

Proceedings Article | 4 March 2019 Presentation + Paper

Autoencoder embedding of task-specific information

Jason Granstedt, Weimin Zhou, Mark Anastasio

Proceedings Volume 10952, 1095207 (2019) https://doi.org/10.1117/12.2513120

KEYWORDS: Computer programming, Signal detection, Statistical analysis, Imaging systems, Targeting Task Performance metric, Machine learning, Image processing algorithms and systems

Read Abstract +

Showing 5 of 9 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