Deep neural networks-based denoising models for CT imaging and their efficacy

Kc, Prabhat; Zeng, Rongping; Farhangi, M. Mehdi; Myers, Kyle J.

doi:10.1117/12.2581418

Cited by 9 publications

(12 citation statements)

References 20 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, the performance only worsened after applying the denoising operation. These results are similar to results in previous studies 9,10,[23][24][25] and again directly contradict the observer-study-based findings. Given these results, we used the proposed observer-study-based characterization to investigate the limited performance of the DL-based denoising approach.…”

Section: Resultssupporting

confidence: 87%

“…We observed that performing the DL-based denoising operation does not yield superior performance on the defect-detection task for any of the signal types. This shows that the observations of limited performance of the CNN seen in previous studies 9,10,[23][24][25] are valid not just for a specific population, but across a range of signal sizes and contrasts. Further, we observed that for an SBR of 2:1, as the size of the signal increased, the difference in the AUC values before and after applying reduced.…”

Section: Resultssupporting

confidence: 66%

See 1 more Smart Citation

Investigating the limited performance of a deep-learning-based SPECT denoising approach: an observer-study-based characterization

Rahman

Jha

2022

Medical Imaging 2022: Image Perception, Observer Performance, and Technology Assessment

View full text Add to dashboard Cite

Multiple objective assessment of image-quality (OAIQ)-based studies have reported that several deep-learning (DL)-based denoising methods show limited performance on signal-detection tasks. Our goal was to investigate the reasons for this limited performance. To achieve this goal, we conducted a task-based characterization of a DL-based denoising approach for individual signal properties. We conducted this study in the context of evaluating a DL-based approach for denoising single photon-emission computed tomography (SPECT) images.The training data consisted of signals of different sizes and shapes within a clustered-lumpy background, imaged with a 2D parallel-hole-collimator SPECT system. The projections were generated at normal and 20% low-count level, both of which were reconstructed using an ordered-subset-expectation-maximization (OSEM) algorithm. A convolutional neural network (CNN)-based denoiser was trained to process the low-count images. The performance of this CNN was characterized for five different signal sizes and four different signal-tobackground ratio (SBRs) by designing each evaluation as a signal-known-exactly/background-known-statistically (SKE/BKS) signal-detection task. Performance on this task was evaluated using an anthropomorphic channelized Hotelling observer (CHO). As in previous studies, we observed that the DL-based denoising method did not improve performance on signal-detection tasks. Evaluation using the idea of observer-study-based characterization demonstrated that the DL-based denoising approach did not improve performance on the signal-detection task for any of the signal types. Overall, these results provide new insights on the performance of the DL-based denoising approach as a function of signal size and contrast. More generally, the observer study-based characterization provides a mechanism to evaluate the sensitivity of the method to specific object properties, and may be explored as analogous to characterizations such as modulation transfer function for linear systems. Finally, this work underscores the need for objective task-based evaluation of DL-based denoising approaches.

show abstract

Section: Resultssupporting

confidence: 87%

Section: Resultssupporting

confidence: 66%

Investigating the limited performance of a deep-learning-based SPECT denoising approach: an observer-study-based characterization

Rahman

Jha

2022

Medical Imaging 2022: Image Perception, Observer Performance, and Technology Assessment

View full text Add to dashboard Cite

show abstract

“…Similarly, AI-based segmentation algorithms are evaluated using FoMs such as Dice scores. However, studies, including recent ones, show that these evaluation strategies may not correlate with clinical-task performance and task-based evaluations may be needed (2,3,(11)(12)(13)(14)(15). One study observed that evaluation of a reconstruction algorithm for whole-body FDG PET using fidelitybased FoMs indicated excellent performance, but on the lesiondetection task, the algorithm was yielding both false-negatives and -positives due to blurring and pseudo-low uptake patterns, respectively (2).…”

Section: Introductionmentioning

confidence: 99%

Nuclear Medicine and Artificial Intelligence: Best Practices for Evaluation (the RELAINCE Guidelines)

et al. 2022

Self Cite

View full text Add to dashboard Cite

An important need exists for strategies to perform rigorous objective clinical-task-based evaluation of artificial intelligence (AI) algorithms for nuclear medicine. To address this need, we propose a 4-class framework to evaluate AI algorithms for promise, technical task-specific efficacy, clinical decision making, and postdeployment efficacy. We provide best practices to evaluate AI algorithms for each of these classes. Each class of evaluation yields a claim that provides a descriptive performance of the AI algorithm. Key best practices are tabulated as the RELAINCE (Recommendations for EvaLuation of AI for NuClear medicinE) guidelines. The report was prepared by the Society of Nuclear Medicine and Molecular Imaging AI Task Force Evaluation team, which consisted of nuclear-medicine physicians, physicists, computational imaging scientists, and representatives from industry and regulatory agencies.

show abstract

“…In ref. , 3 we showed that the performance of a DNN depends on several choices. These choices include patch size (e.g.…”

Section: Computational Optimizationmentioning

confidence: 96%

On use of augmentation for the DNN-based CT image denoising

Myers²,

Farhangi³

et al. 2022

7th International Conference on Image Formation in X-Ray Computed Tomography

View full text Add to dashboard Cite

Augmentation strategies have been suggested to overcome issues related to limited training data for the Deep Learning (DL)-based CT image denoising problem. Although augmentation is, indeed, a good machine learning practice, the extent of improvements achieved by DL-based CT denoising solvers through augmentation pipelines remains to be quantified. Accordingly, in this work, we make use of two different deep neural networks (the REDCNN, the DnCNN) to quantify gains in CT image quality through augmentation. The augmentation strategies considered include computer vision inspired strategies (like scaling, rotating, flipping, image-blending) and the CT forward model-based noise (radiation dose) insertion. Likewise, image qualities considered in this work include visual perception-and data fidelity-based global metrics (like the PSNR, the SSIM, the RSE) which are common in the computer vision literature, and CT bench tests (like the NPS, the MTF) and a task-based detectability assessment (the LCD) from the CT imaging literature. Our preliminary results indicate that the DL solvers trained on augmented data show gains in the global metrics, in low-frequency noise texture components and in the MTF values as compared to the ones trained on non-augmented data. However, when the augmented DL-solvers were compared against their low-dose counterparts, their performance -with respect to the noise frequency components, resolution, and detection task -was not all improved, and in some cases even worse.

show abstract

Deep neural networks-based denoising models for CT imaging and their efficacy

Cited by 9 publications

References 20 publications

Investigating the limited performance of a deep-learning-based SPECT denoising approach: an observer-study-based characterization

Investigating the limited performance of a deep-learning-based SPECT denoising approach: an observer-study-based characterization

Nuclear Medicine and Artificial Intelligence: Best Practices for Evaluation (the RELAINCE Guidelines)

On use of augmentation for the DNN-based CT image denoising

Contact Info

Product

Resources

About