Deep learning-based deformable registration methods have been widely investigated in diverse medical applications. Learning-based deformable registration relies on weighted objective functions trading off registration accuracy and smoothness of the deformation field. Therefore, they inevitably require tuning the hyperparameter for optimal registration performance. Tuning the hyperparameters is highly computationally expensive and introduces undesired dependencies on domain knowledge. In this study, we construct a registration model based on the gradient surgery mechanism, named GSMorph, to achieve a hyperparameter-free balance on multiple losses. In GSMorph, we reformulate the optimization procedure by projecting the gradient of similarity loss orthogonally to the plane associated with the smoothness constraint, rather than additionally introducing a hyperparameter to balance these two competing terms. Furthermore, our method is model-agnostic and can be merged into any deep registration network without introducing extra parameters or slowing down inference. In this study, We compared our method with state-of-the-art (SOTA) deformable registration approaches over two publicly available cardiac MRI datasets. GSMorph proves superior to five SOTA learning-based registration models and two conventional registration techniques, SyN and Demons, on both registration accuracy and smoothness.
Localization of the narrowest position of the vessel and corresponding vessel and remnant vessel delineation in carotid ultrasound (US) are essential for carotid stenosis grading (CSG) in clinical practice. However, the pipeline is time-consuming and tough due to the ambiguous boundaries of plaque and temporal variation. To automatize this procedure, a large number of manual delineations are usually required, which is not only laborious but also not reliable given the annotation difficulty. In this study, we present the first video classification framework for automatic CSG. Our contribution is three-fold. First, to avoid the requirement of laborious and unreliable annotation, we propose a novel and effective video classification network for weakly-supervised CSG. Second, to ease the model training, we adopt an inflation strategy for the network, where pre-trained 2D convolution weights can be adapted into the 3D counterpart in our network for an effective warm start. Third, to enhance the feature discrimination of the video, we propose a novel attention-guided multi-dimension fusion (AMDF) transformer encoder to model and integrate global dependencies within and across spatial and temporal dimensions, where two lightweight cross-dimensional attention mechanisms are designed. Our approach is extensively validated on a large clinically collected carotid US video dataset, demonstrating state-of-the-art performance compared with strong competitors.
Deep classifiers may encounter significant performance degradation when processing unseen testing data from varying centers, vendors, and protocols. Ensuring the robustness of deep models against these domain shifts is crucial for their widespread clinical application. In this study, we propose a novel approach called Fourier Test-time Adaptation (FTTA), which employs a dual-adaptation design to integrate input and model tuning, thereby jointly improving the model robustness. The main idea of FTTA is to build a reliable multi-level consistency measurement of paired inputs for achieving self-correction of prediction. Our contribution is two-fold. First, we encourage consistency in global features and local attention maps between the two transformed images of the same input. Here, the transformation refers to Fourier-based input adaptation, which can transfer one unseen image into source style to reduce the domain gap. Furthermore, we leverage style-interpolated images to enhance the global and local features with learnable parameters, which can smooth the consistency measurement and accelerate convergence. Second, we introduce a regularization technique that utilizes style interpolation consistency in the frequency space to encourage self-consistency in the logit space of the model output. This regularization provides strong self-supervised signals for robustness enhancement. FTTA was extensively validated on three large classification datasets with different modalities and organs. Experimental results show that FTTA is general and outperforms other strong state-of-the-art methods.
The Segment Anything Model (SAM) is the first foundation model for general image segmentation. It designed a novel promotable segmentation task, ensuring zero-shot image segmentation using the pre-trained model via two main modes including automatic everything and manual prompt. SAM has achieved impressive results on various natural image segmentation tasks. However, medical image segmentation (MIS) is more challenging due to the complex modalities, fine anatomical structures, uncertain and complex object boundaries, and wide-range object scales. Meanwhile, zero-shot and efficient MIS can well reduce the annotation time and boost the development of medical image analysis. Hence, SAM seems to be a potential tool and its performance on large medical datasets should be further validated. We collected and sorted 52 open-source datasets, and built a large medical segmentation dataset with 16 modalities, 68 objects, and 553K slices. We conducted a comprehensive analysis of different SAM testing strategies on the so-called COSMOS 553K dataset. Extensive experiments validate that SAM performs better with manual hints like points and boxes for object perception in medical images, leading to better performance in prompt mode compared to everything mode. Additionally, SAM shows remarkable performance in some specific objects and modalities, but is imperfect or even totally fails in other situations. Finally, we analyze the influence of different factors (e.g., the Fourier-based boundary complexity and size of the segmented objects) on SAM's segmentation performance. Extensive experiments validate that SAM's zero-shot segmentation capability is not sufficient to ensure its direct application to the MIS.
Efficient point cloud representation is a fundamental element of Lidar-based 3D object detection. Recent grid-based detectors usually divide point clouds into voxels or pillars and construct single-stream networks in Bird's Eye View. However, these point cloud encoding paradigms underestimate the point representation in the vertical direction, which cause the loss of semantic or fine-grained information, especially for vertical sensitive objects like pedestrian and cyclists. In this paper, we propose an explicit vertical multi-scale representation learning framework, VPFusion, to combine the complementary information from both voxel and pillar streams. Specifically, VPFusion first builds upon a sparse voxel-pillar-based backbone. The backbone divides point clouds into voxels and pillars, then encodes features with 3D and 2D sparse convolution simultaneously. Next, we introduce the Sparse Fusion Layer (SFL), which establishes a bidirectional pathway for sparse voxel and pillar features to enable the interaction between them. Additionally, we present the Dense Fusion Neck (DFN) to effectively combine the dense feature maps from voxel and pillar branches with multi-scale. Extensive experiments on the large-scale Waymo Open Dataset and nuScenes Dataset demonstrate that VPFusion surpasses the single-stream baselines by a large margin and achieves state-of-the-art performance with real-time inference speed.
The success of existing video super-resolution (VSR) algorithms stems mainly exploiting the temporal information from the neighboring frames. However, none of these methods have discussed the influence of the temporal redundancy in the patches with stationary objects and background and usually use all the information in the adjacent frames without any discrimination. In this paper, we observe that the temporal redundancy will bring adverse effect to the information propagation,which limits the performance of the most existing VSR methods. Motivated by this observation, we aim to improve existing VSR algorithms by handling the temporal redundancy patches in an optimized manner. We develop two simple yet effective plug and play methods to improve the performance of existing local and non-local propagation-based VSR algorithms on widely-used public videos. For more comprehensive evaluating the robustness and performance of existing VSR algorithms, we also collect a new dataset which contains a variety of public videos as testing set. Extensive evaluations show that the proposed methods can significantly improve the performance of existing VSR methods on the collected videos from wild scenarios while maintain their performance on existing commonly used datasets. The code is available at https://github.com/HYHsimon/Boosted-VSR.
Deep segmentation models often face the failure risks when the testing image presents unseen distributions. Improving model robustness against these risks is crucial for the large-scale clinical application of deep models. In this study, inspired by human learning cycle, we propose a novel online reflective learning framework (RefSeg) to improve segmentation robustness. Based on the reflection-on-action conception, our RefSeg firstly drives the deep model to take action to obtain semantic segmentation. Then, RefSeg triggers the model to reflect itself. Because making deep models realize their segmentation failures during testing is challenging, RefSeg synthesizes a realistic proxy image from the semantic mask to help deep models build intuitive and effective reflections. This proxy translates and emphasizes the segmentation flaws. By maximizing the structural similarity between the raw input and the proxy, the reflection-on-action loop is closed with segmentation robustness improved. RefSeg runs in the testing phase and is general for segmentation models. Extensive validation on three medical image segmentation tasks with a public cardiac MR dataset and two in-house large ultrasound datasets show that our RefSeg remarkably improves model robustness and reports state-of-the-art performance over strong competitors.
Standard plane (SP) localization is essential in routine clinical ultrasound (US) diagnosis. Compared to 2D US, 3D US can acquire multiple view planes in one scan and provide complete anatomy with the addition of coronal plane. However, manually navigating SPs in 3D US is laborious and biased due to the orientation variability and huge search space. In this study, we introduce a novel reinforcement learning (RL) framework for automatic SP localization in 3D US. Our contribution is three-fold. First, we formulate SP localization in 3D US as a tangent-point-based problem in RL to restructure the action space and significantly reduce the search space. Second, we design an auxiliary task learning strategy to enhance the model's ability to recognize subtle differences crossing Non-SPs and SPs in plane search. Finally, we propose a spatial-anatomical reward to effectively guide learning trajectories by exploiting spatial and anatomical information simultaneously. We explore the efficacy of our approach on localizing four SPs on uterus and fetal brain datasets. The experiments indicate that our approach achieves a high localization accuracy as well as robust performance.
Ultrasound (US) is widely used for its advantages of real-time imaging, radiation-free and portability. In clinical practice, analysis and diagnosis often rely on US sequences rather than a single image to obtain dynamic anatomical information. This is challenging for novices to learn because practicing with adequate videos from patients is clinically unpractical. In this paper, we propose a novel framework to synthesize high-fidelity US videos. Specifically, the synthesis videos are generated by animating source content images based on the motion of given driving videos. Our highlights are three-fold. First, leveraging the advantages of self- and fully-supervised learning, our proposed system is trained in weakly-supervised manner for keypoint detection. These keypoints then provide vital information for handling complex high dynamic motions in US videos. Second, we decouple content and texture learning using the dual decoders to effectively reduce the model learning difficulty. Last, we adopt the adversarial training strategy with GAN losses for further improving the sharpness of the generated videos, narrowing the gap between real and synthesis videos. We validate our method on a large in-house pelvic dataset with high dynamic motion. Extensive evaluation metrics and user study prove the effectiveness of our proposed method.
This paper reviews the NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video. In this challenge, we proposed the LDV 2.0 dataset, which includes the LDV dataset (240 videos) and 95 additional videos. This challenge includes three tracks. Track 1 aims at enhancing the videos compressed by HEVC at a fixed QP. Track 2 and Track 3 target both the super-resolution and quality enhancement of HEVC compressed video. They require x2 and x4 super-resolution, respectively. The three tracks totally attract more than 600 registrations. In the test phase, 8 teams, 8 teams and 12 teams submitted the final results to Tracks 1, 2 and 3, respectively. The proposed methods and solutions gauge the state-of-the-art of super-resolution and quality enhancement of compressed video. The proposed LDV 2.0 dataset is available at https://github.com/RenYang-home/LDV_dataset. The homepage of this challenge (including open-sourced codes) is at https://github.com/RenYang-home/NTIRE22_VEnh_SR.