The success of deep learning on image classification and recognition tasks has led to new applications in diverse contexts, including the field of medical imaging. However, two properties of deep neural networks (DNNs) may limit their future use in medical applications. The first is that DNNs require a large amount of labeled training data, and the second is that the deep learning-based models lack interpretability. In this paper, we propose and investigate a data-efficient framework for the task of general medical image segmentation. We address the two aforementioned challenges by introducing domain knowledge in the form of a strong prior into a deep learning framework. This prior is expressed by a customized dynamical system. We performed experiments on two different datasets, namely JSRT and ISIC2016 (heart and lungs segmentation on chest X-ray images and skin lesion segmentation on dermoscopy images). We have achieved competitive results using the same amount of training data compared to the state-of-the-art methods. More importantly, we demonstrate that our framework is extremely data-efficient, and it can achieve reliable results using extremely limited training data. Furthermore, the proposed method is rotationally invariant and insensitive to initialization.

Deformable image registration, the estimation of the spatial transformation between different images, is an important task in medical imaging. Deep learning techniques have been shown to perform 3D image registration efficiently. However, current registration strategies often only focus on the deformation smoothness, which leads to the ignorance of complicated motion patterns (e.g., separate or sliding motions), especially for the intersection of organs. Thus, the performance when dealing with the discontinuous motions of multiple nearby objects is limited, causing undesired predictive outcomes in clinical usage, such as misidentification and mislocalization of lesions or other abnormalities. Consequently, we proposed a novel registration method to address this issue: a new Motion Separable backbone is exploited to capture the separate motion, with a theoretical analysis of the upper bound of the motions' discontinuity provided. In addition, a novel Residual Aligner module was used to disentangle and refine the predicted motions across the multiple neighboring objects/organs. We evaluate our method, Residual Aligner-based Network (RAN), on abdominal Computed Tomography (CT) scans and it has shown to achieve one of the most accurate unsupervised inter-subject registration for the 9 organs, with the highest-ranked registration of the veins (Dice Similarity Coefficient (%)/Average surface distance (mm): 62%/4.9mm for the vena cava and 34%/7.9mm for the portal and splenic vein), with a smaller model structure and less computation compared to state-of-the-art methods. Furthermore, when applied to lung CT, the RAN achieves comparable results to the best-ranked networks (94%/3.0mm), also with fewer parameters and less computation.

Image semantic segmentation is a dense prediction task in computer vision that is dominated by deep learning techniques in recent years. UNet, which is a symmetric encoder-decoder end-to-end Convolutional Neural Network (CNN) with skip connections, has shown promising performance. Aiming to process the multiscale feature information efficiently, we propose a new Densely Connected Swin-UNet (DCS-UNet) with multiscale information aggregation for medical image segmentation. Firstly, inspired by Swin-Transformer to model long-range dependencies via shift-window-based self-attention, this work proposes the use of fully ViT-based network blocks with a shift-window approach, resulting in a purely self-attention-based U-shape segmentation network. The relevant layers including feature sampling and image tokenization are re-designed to align with the ViT fashion. Secondly, a full-scale deep supervision scheme is developed to process the aggregated feature map with various resolutions generated by different levels of decoders. Thirdly, dense skip connections are proposed that allow the semantic feature information to be thoroughly transferred from different levels of encoders to lower level decoders. Our proposed method is validated on a public benchmark MRI Cardiac segmentation data set with comprehensive validation metrics showing competitive performance against other variant encoder-decoder networks. The code is available at https://github.com/ziyangwang007/VIT4UNet.

Deep Convolutional Neural Networks (DCNNs) are used extensively in medical image segmentation and hence 3D navigation for robot-assisted Minimally Invasive Surgeries (MISs). However, current DCNNs usually use down sampling layers for increasing the receptive field and gaining abstract semantic information. These down sampling layers decrease the spatial dimension of feature maps, which can be detrimental to image segmentation. Atrous convolution is an alternative for the down sampling layer. It increases the receptive field whilst maintains the spatial dimension of feature maps. In this paper, a method for effective atrous rate setting is proposed to achieve the largest and fully-covered receptive field with a minimum number of atrous convolutional layers. Furthermore, a new and full resolution DCNN - Atrous Convolutional Neural Network (ACNN), which incorporates cascaded atrous II-blocks, residual learning and Instance Normalization (IN) is proposed. Application results of the proposed ACNN to Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) image segmentation demonstrate that the proposed ACNN can achieve higher segmentation Intersection over Unions (IoUs) than U-Net and Deeplabv3+, but with reduced trainable parameters.

The current standard of intra-operative navigation during Fenestrated Endovascular Aortic Repair (FEVAR) calls for the need of 3D alignments between inserted devices and aortic branches. The navigation commonly via 2D fluoroscopic images, lacks anatomical information, resulting in longer operation hours and radiation exposure. In this paper, a skeleton instantiation framework of Abdominal Aortic Aneurysm (AAA) from a single 2D fluoroscopic image is introduced for real-time 3D robotic path planning. A graph matching method is proposed to establish the correspondences between the 3D preoperative and 2D intra-operative AAA skeletons, and then the two skeletons are registered by skeleton deformation and regularization in respect to skeleton length and smoothness. Furthermore, deep learning was used to segment 3D preoperative AAA from Computed Tomography (CT) scans to facilitate the framework automation. Simulation, phantom and patient AAA data sets have been used to validate the proposed framework. 3D distance error of 2mm was achieved in the phantom setup. Performance advantages were also achieved in terms of accuracy, robustness and time-efficiency.