Trainable joint bilateral filters for enhanced prediction stability in low-dose CT

Abstract

Low-dose computed tomography (CT) denoising algorithms aim to enable reduced patient dose in routine CT acquisitions while maintaining high image quality. Recently, deep learning (DL)-based methods were introduced, outperforming conventional denoising algorithms on this task due to their high model capacity. However, for the transition of DL-based denoising to clinical practice, these data-driven approaches must generalize robustly beyond the seen training data. We, therefore, propose a hybrid denoising approach consisting of a set of trainable joint bilateral filters (JBFs) combined with a convolutional DL-based denoising network to predict the guidance image. Our proposed denoising pipeline combines the high model capacity enabled by DL-based feature extraction with the reliability of the conventional JBF. The pipeline’s ability to generalize is demonstrated by training on abdomen CT scans without metal implants and testing on abdomen scans with metal implants as well as on head CT data. When embedding RED-CNN/QAE, two well-established DL-based denoisers in our pipeline, the denoising performance is improved by 10%/82% (RMSE) and 3%/81% (PSNR) in regions containing metal and by 6%/78% (RMSE) and 2%/4% (PSNR) on head CT data, compared to the respective vanilla model. Concluding, the proposed trainable JBFs limit the error bound of deep neural networks to facilitate the applicability of DL-based denoisers in low-dose CT pipelines.

Deep learning for terahertz image denoising in nondestructive historical document analysis

“…In this work, the CycleGAN and Pix2pixGAN models are purely data driven. Data driven deep learning models may not generalize well to out-of-distribution test data and are sensitive to noise or perturbations 49,60 . Therefore, in our CycleGAN results, some synthetic images have different appearance characteristics (e.g., Fig.…”

“…The bilateral filter 47 and its trainable version 48 are applied to compare with well-known adaptive www.nature.com/scientificreports/ filters. In particular trainable bilateral filter versions have been shown to provide robust denoising performance in the context of medical imaging 49 . The iterative reweighted total variation (wTV) algorithm 50 is selected as a compressed sensing representative.…”

CT image denoising methods for image quality improvement and radiation dose reduction

“…Evaluating a DL denoising model involves assessing its ability to effectively reduce noise while preserving important image details. Generated denoised images are usually compared against the ground truth images such as FBP, 164–166 IR, 167–170 or other DL methods 158,159,163,171–182 . This comparison can provide insights into the model's relative strengths and weaknesses in terms of denoising performance.…”

“…Generated denoised images are usually compared against the ground truth images such as FBP, 164 , 165 , 166 IR, 167 , 168 , 169 , 170 or other DL methods. 158 , 159 , 163 , 171 , 172 , 173 , 174 , 175 , 176 , 177 , 178 , 179 , 180 , 181 , 182 This comparison can provide insights into the model's relative strengths and weaknesses in terms of denoising performance. If the model's performance is not satisfactory, consider iterative improvements such as architecture modifications, hyperparameter tuning, or dataset augmentation.…”

Abstract: Trainable Joint Bilateral Filters for Enhanced Prediction Stability in Low-dose CT

“…Concluding, the proposed trainable JBFs limit the error bound of deep neural networks to facilitate the applicability of DL-based denoisers in low-dose CT pipelines. We made our trainable bilateral filter layer package (PyTorch, GPU accelerated) publicly available [1,2].…”