16 May 2022 in Physics in Medicine & Biology
Objective. There are several x-ray computed tomography (CT) scanning strategies used to reduce radiation dose, such as (1) sparse-view CT, (2) low-dose CT and (3) region-of-interest (ROI) CT (called interior tomography). To further reduce the dose, sparse-view and/or low-dose CT settings can be applied together with interior tomography. Interior tomography has various advantages in terms of reducing the number of detectors and decreasing the x-ray radiation dose. However, a large patient or a small field-of-view (FOV) detector can cause truncated projections, and then the reconstructed images suffer from severe cupping artifacts. In addition, although low-dose CT can reduce the radiation exposure dose, analytic reconstruction algorithms produce image noise. Recently, many researchers have utilized image-domain deep learning (DL) approaches to remove each artifact and demonstrated impressive performances, and the theory of deep convolutional framelets supports the reason for the performance improvement. Approach. In this paper, we found that it is difficult to solve coupled artifacts using an image-domain convolutional neural network (CNN) based on deep convolutional framelets. Significance. To address the coupled problem, we decouple it into two sub-problems: (i) image-domain noise reduction inside the truncated projection to solve low-dose CT problem and (ii) extrapolation of the projection outside the truncated projection to solve the ROI CT problem. The decoupled sub-problems are solved directly with a novel proposed end-to-end learning method using dual-domain CNNs. Main results. We demonstrate that the proposed method outperforms the conventional image-domain DL methods, and a projection-domain CNN shows better performance than the image-domain CNNs commonly used by many researchers.
22 January 2025 in IEEE Access
Deepfake content-including audio, video, images, and text-synthesized or modified using artificial intelligence is designed to convincingly mimic real content. As deepfake generation technology advances, detecting deepfake content presents significant challenges. While recent progress has been made in detection techniques, identifying deepfake audio remains particularly challenging. Previous approaches have attempted to capture deepfake features by combining video and audio content; however, these methods are ineffective when video and audio are mismatched due to occlusion. To address this, we proposes a novel dual-channel deepfake audio detection model that leverages the direct and reverberant components extracted from raw audio signals, focusing exclusively on audio-based detection without reliance on video content. Across various datasets, including ASVspoof2019, FakeAVCeleb, and sport press conference datasets collected by our group, the proposed dual-channel model demonstrates significant improvements in quantitative metrics such as equal error rate and area under the curve. The implementation is available at https://github.com/gunwoo5034/Dual-Channel-Audio-Deepfake-Detection.
3 April 2024 in Physics in Medicine & Biology
Objective . X-ray computed tomography employing sparse projection views has emerged as a contemporary technique to mitigate radiation dose. However, due to the inadequate number of projection views, an analytic reconstruction method utilizing filtered backprojection results in severe streaking artifacts. Recently, deep learning (DL) strategies employing image-domain networks have demonstrated remarkable performance in eliminating the streaking artifact caused by analytic reconstruction methods with sparse projection views. Nevertheless, it is difficult to clarify the theoretical justification for applying DL to sparse view computed tomography (CT) reconstruction, and it has been understood as restoration by removing image artifacts, not reconstruction. Approach . By leveraging the theory of deep convolutional framelets (DCF) and the hierarchical decomposition of measurement, this research reveals the constraints of conventional image and projection-domain DL methodologies, subsequently, the research proposes a novel dual-domain DL framework utilizing hierarchical decomposed measurements. Specifically, the research elucidates how the performance of the projection-domain network can be enhanced through a low-rank property of DCF and a bowtie support of hierarchical decomposed measurement in the Fourier domain. Main results . This study demonstrated performance improvement of the proposed framework based on the low-rank property, resulting in superior reconstruction performance compared to conventional analytic and DL methods. Significance . By providing a theoretically justified DL approach for sparse-view CT reconstruction, this study not only offers a superior alternative to existing methods but also opens new avenues for research in medical imaging. It highlights the potential of dual-domain DL frameworks to achieve high-quality reconstructions with lower radiation doses, thereby advancing the field towards safer and more efficient diagnostic techniques. The code is available at https://github.com/hanyoseob/HDD-DL-for-SVCT .
1 February 2025 in Computerized Medical Imaging and Graphics
Deep learning-based image synthesis for medical imaging is currently an active research topic with various clinically relevant applications. Recently, methods allowing training with misaligned data have started to emerge, yet current solution lack robustness and cannot handle other corruptions in the dataset. In this work, we propose a solution to this problem for training synthesis network for datasets affected by mis-registration, artifacts, and deformations. Our proposed method consists of three key innovations: meta-learning inspired re-weighting scheme to directly decrease the influence of corrupted instances in a mini-batch by assigning lower weights in the loss function, non-local feature-based loss function, and joint training of image synthesis network together with spatial transformer (STN)-based registration networks with specially designed regularization. Efficacy of our method is validated in a controlled synthetic scenario, as well as public dataset with such corruptions. This work introduces a new framework that may be applicable to challenging scenarios and other more difficult datasets.
27 October 2025 in IEEE Access
Medical image segmentation is a crucial component of disease diagnosis and treatment planning. The Segment Anything Model (SAM), which has recently gained prominence in natural image processing, exhibits remarkable zero-shot generalization performance. However, the SAM architecture is fundamentally limited to a single-modality input, which restricts its ability to leverage multi-modality medical images such as multi-contrast MRI. To address this limitation, in this study we introduce Collaborative Medical SAM (CoMed-SAM), an enhanced segmentation model designed to integrate multiple medical imaging modalities. CoMed-SAM incorporates two novel contributions for robust performance, even with a variable number of inputs: (1) an embedding fusion module that effectively merges features from multiple encoders, and (2) a dropout learning strategy that ensures generalization despite missing modalities. Experimental results on the IVDM3Seg dataset for lumbar intervertebral disc segmentation and the CHAOS dataset for abdominal organ segmentation demonstrate that CoMed-SAM significantly outperforms conventional SAM-based models. Notably, it also achieves superior segmentation accuracy in single-modality scenarios, highlighting its enhanced feature extraction capabilities. Furthermore, ablation studies confirm that the dropout learning strategy is critical, as models trained with this strategy consistently outperform those trained without it. The source code and our pretrained model are available at https://github.com/hunzo300/CoMed-SAM.git.