To reduce the radiation dose, sparse-view computed tomography (CT) reconstruction has been proposed, aiming to recover high-quality CT images from sparsely sampled sinogram. To eliminate the artifacts present in sparse-view CT images, a new dual-domain diffusion model (DDDM) is proposed, which is composed of a sinogram upgrading module (SUM) and an image refining module (IRM) connected in series. In the sinogram domain, a novel degrading and upgrading framework is defined, in which SUM is trained to upgrade sparse-view sinograms step by step to reverse the degradation process of CT images caused by successive down-sampling of scanning views. In the image domain, IRM adopts an improved denoising diffusion framework to further reduce remaining artifacts and restore image details, where a skip connection from the original sparseview sinogram is introduced to constrain the generation of details. Our DDDM shows significant improvement over deep-learning baseline models in both classical similarity metrics and perceptual loss, and has good generalization to untrained organs. We release our code at https://github.com/YC-Markus/code-for-DDDM.

In robot-assisted cardiac surgery, predicting heart motion can help improve the operation accuracy and safety of surgical robots. Different from the conventional prediction schemes which model the point of interest (POI) with only temporal correlation of past observations, this paper proposes an LSTM-based method by exploiting the spatio-temporal correlation of the 3D movements of POI and auxiliary points (APs) on the same surface of the heart. Three different LSTM models are investigated. The first two models define the POI prediction as a pure time-series forecasting problem based on past POI trajectory, and the third model combines the past observations of POI and new observations of APs to take into consideration the extra spatial correlations for prediction. Experimental comparison studies based on 3D coordinates obtained from real stereo-endoscopic videos demonstrate the superior performance of the proposed spatio-temporal LSTM model.

In dynamic minimally invasive surgery environments, 3D reconstruction of deformable soft-tissue surfaces with stereo endoscopic images is very challenging. A simple self-supervised stereo reconstruction framework is proposed to address this issue, which bridges the traditional geometric deformable models and the newly revived neural networks. The equivalence between the classical thin plate spline (TPS) model and a single-layer fully-connected or convolutional network is studied. By alternating training of two TPS equivalent networks within the self-supervised framework, disparity priors are learnt from the past stereo frames of target tissues to form an optimized disparity basis, on which disparity maps of subsequent frames can be estimated more accurately without sacrificing computational efficiency and robustness. The proposed method was verified on stereo-endoscopic videos recorded by the da Vinci®surgical robots.

A dynamic image fusion scheme for infrared and visible sequence based on region target detection is proposed in this paper. Target detection technique is employed to segment the source images into target and background regions. Different fusion rules are adopted respectively in target and background regions. A limitedly redundant discrete wavelet transform (LR DWT) method is introduced to achieve shift invariant multi-resolution representation of each source images. Fusion experiments on real world image sequences indicate that the proposed method is effective and efficient, which achieves better performance than the generic fusion method.

This paper presents conditions under which the sampling lattice for a filter bank can be replaced without loss of perfect reconstruction. This is the generalization of common knowledge that removing up/downsampling will not lose perfect reconstruction. The results provide a simple way of building oversampled filter banks. If the original filter banks are orthogonal, these oversampled banks construct tight frames of l2 (Z(n)) when iterated. As an example, a quincunx lattice is used to replace the rectangular one of the standard wavelet transform. This replacement leads to a tight frame that has a higher sampling in both time and frequency. The frame transform is nearly shift invariant and has intermediate scales. An application of the transform to image fusion is also presented.