Semantic segmentation of polyps and depth estimation are two important research problems in endoscopic image analysis. One of the main obstacles to conduct research on these research problems is lack of annotated data. Endoscopic annotations necessitate the specialist knowledge of expert endoscopists and due to this, it can be difficult to organise, expensive and time consuming. To address this problem, we investigate an active learning paradigm to reduce the number of training examples by selecting the most discriminative and diverse unlabelled examples for the task taken into consideration. Most of the existing active learning pipelines are task-agnostic in nature and are often sub-optimal to the end task. In this paper, we propose a novel task-aware active learning pipeline and applied for two important tasks in endoscopic image analysis: semantic segmentation and depth estimation. We compared our method with the competitive baselines. From the experimental results, we observe a substantial improvement over the compared baselines. Codes are available at https://github.com/thetna/endo-active-learn.
We present our novel deep multi-task learning method for medical image segmentation. Existing multi-task methods demand ground truth annotations for both the primary and auxiliary tasks. Contrary to it, we propose to generate the pseudo-labels of an auxiliary task in an unsupervised manner. To generate the pseudo-labels, we leverage Histogram of Oriented Gradients (HOGs), one of the most widely used and powerful hand-crafted features for detection. Together with the ground truth semantic segmentation masks for the primary task and pseudo-labels for the auxiliary task, we learn the parameters of the deep network to minimise the loss of both the primary task and the auxiliary task jointly. We employed our method on two powerful and widely used semantic segmentation networks: UNet and U2Net to train in a multi-task setup. To validate our hypothesis, we performed experiments on two different medical image segmentation data sets. From the extensive quantitative and qualitative results, we observe that our method consistently improves the performance compared to the counter-part method. Moreover, our method is the winner of FetReg Endovis Sub-challenge on Semantic Segmentation organised in conjunction with MICCAI 2021.
Video streams are utilised to guide minimally-invasive surgery and diagnostic procedures in a wide range of procedures, and many computer assisted techniques have been developed to automatically analyse them. These approaches can provide additional information to the surgeon such as lesion detection, instrument navigation, or anatomy 3D shape modeling. However, the necessary image features to recognise these patterns are not always reliably detected due to the presence of irregular light patterns such as specular highlight reflections. In this paper, we aim at removing specular highlights from endoscopic videos using machine learning. We propose using a temporal generative adversarial network (GAN) to inpaint the hidden anatomy under specularities, inferring its appearance spatially and from neighbouring frames where they are not present in the same location. This is achieved using in-vivo data of gastric endoscopy (Hyper-Kvasir) in a fully unsupervised manner that relies on automatic detection of specular highlights. System evaluations show significant improvements to traditional methods through direct comparison as well as other machine learning techniques through an ablation study that depicts the importance of the network's temporal and transfer learning components. The generalizability of our system to different surgical setups and procedures was also evaluated qualitatively on in-vivo data of gastric endoscopy and ex-vivo porcine data (SERV-CT, SCARED). We also assess the effect of our method in computer vision tasks that underpin 3D reconstruction and camera motion estimation, namely stereo disparity, optical flow, and sparse point feature matching. These are evaluated quantitatively and qualitatively and results show a positive effect of specular highlight inpainting on these tasks in a novel comprehensive analysis.
Automatic surgical scene segmentation is fundamental for facilitating cognitive intelligence in the modern operating theatre. Previous works rely on conventional aggregation modules (e.g., dilated convolution, convolutional LSTM), which only make use of the local context. In this paper, we propose a novel framework STswinCL that explores the complementary intra- and inter-video relations to boost segmentation performance, by progressively capturing the global context. We firstly develop a hierarchy Transformer to capture intra-video relation that includes richer spatial and temporal cues from neighbor pixels and previous frames. A joint space-time window shift scheme is proposed to efficiently aggregate these two cues into each pixel embedding. Then, we explore inter-video relation via pixel-to-pixel contrastive learning, which well structures the global embedding space. A multi-source contrast training objective is developed to group the pixel embeddings across videos with the ground-truth guidance, which is crucial for learning the global property of the whole data. We extensively validate our approach on two public surgical video benchmarks, including EndoVis18 Challenge and CaDIS dataset. Experimental results demonstrate the promising performance of our method, which consistently exceeds previous state-of-the-art approaches. Code will be available at https://github.com/YuemingJin/STswinCL.
In robotic-assisted partial nephrectomy, surgeons remove a part of a kidney often due to the presence of a mass. A drop-in ultrasound probe paired to a surgical robot is deployed to execute multiple swipes over the kidney surface to localise the mass and define the margins of resection. This sub-task is challenging and must be performed by a highly skilled surgeon. Automating this sub-task may reduce cognitive load for the surgeon and improve patient outcomes. The overall goal of this work is to autonomously move the ultrasound probe on the surface of the kidney taking advantage of the use of the Pneumatically Attachable Flexible (PAF) rail system, a soft robotic device used for organ scanning and repositioning. First, we integrate a shape-sensing optical fibre into the PAF rail system to evaluate the curvature of target organs in robotic-assisted laparoscopic surgery. Then, we investigate the impact of the stiffness of the material of the PAF rail on the curvature sensing accuracy, considering that soft targets are present in the surgical field. Finally, we use shape sensing to plan the trajectory of the da Vinci surgical robot paired with a drop-in ultrasound probe and autonomously generate an Ultrasound scan of a kidney phantom.
With the advent of soft robotics the research community has been exploring how to exploit the inherent adaptability of soft elastic materials to overcome the limitations of systems based on rigid materials. A proof-of-concept design of a crawling robot for pipe inspection and preliminary analysis of its locomotion capabilities are presented in this work. The novelty of the proposed design is the combination of silicone based elements of different stiffness to enable global shape reconfiguration and whole-body track-based locomotion. The proposed system makes use of a plurality of multi-materials elastic tracks actuated by an on-board motor through a worm gear that pairs with all of them, causing their constant eversion from the inside to the outside of the chassis of the robot. Two toroidal inflatable chambers surround the chassis of the robot while passing through the loops created by the tracks. Upon inflation of the chambers the tracks are deformed, changing the overall diameter of the system. This feature allows the system to adjust to the local diameter of the pipe navigated, enabling also active contact force control between the tracks and the pipe walls. We demonstrate how the proposed system efficiently moves through rigid pipes of different diameters, both straight and curved, incrementing its outer diameter up to 100% of his original size. Maximum navigation speed and stall force applied are evaluated. With two front cameras embedded, the proposed robotic system can represent a cost-effective and easy-to-control solution for inspection applications, when adaptability and compliance are critical requirements.
Learning a disentangled, interpretable, and structured latent representation in 3D generative models of faces and bodies is still an open problem. The problem is particularly acute when control over identity features is required. In this paper, we propose an intuitive yet effective self-supervised approach to train a 3D shape variational autoencoder (VAE) which encourages a disentangled latent representation of identity features. Curating the mini-batch generation by swapping arbitrary features across different shapes allows to define a loss function leveraging known differences and similarities in the latent representations. Experimental results conducted on 3D meshes show that state-of-the-art methods for latent disentanglement are not able to disentangle identity features of faces and bodies. Our proposed method properly decouples the generation of such features while maintaining good representation and reconstruction capabilities.
With soft robotics being increasingly employed in settings demanding high and controlled contact forces, recent research has demonstrated the use of soft robots to estimate or intrinsically sense forces without requiring external sensing mechanisms. Whilst this has mainly been shown in tendon-based continuum manipulators or deformable robots comprising of push-pull rod actuation, fluid drives still pose great challenges due to high actuation variability and nonlinear mechanical system responses. In this work we investigate the capabilities of a hydraulic, parallel soft robot to intrinsically sense and subsequently control contact forces. A comprehensive algorithm is derived for static, quasi-static and dynamic force sensing which relies on fluid volume and pressure information of the system. The algorithm is validated for a single degree-of-freedom soft fluidic actuator. Results indicate that axial forces acting on a single actuator can be estimated with an accuracy of 0.56 +- 0.66N within the validated range of 0 to 6N in a quasi-static configuration. The force sensing methodology is applied to force control in a single actuator as well as the coupled parallel robot. It can be seen that forces are accurately controllable for both systems, with the capability of controlling directional contact forces in case of the multi degree-of-freedom parallel soft robot.
Surgical scene segmentation is essential for anatomy and instrument localization which can be further used to assess tissue-instrument interactions during a surgical procedure. In 2017, the Challenge on Automatic Tool Annotation for cataRACT Surgery (CATARACTS) released 50 cataract surgery videos accompanied by instrument usage annotations. These annotations included frame-level instrument presence information. In 2020, we released pixel-wise semantic annotations for anatomy and instruments for 4670 images sampled from 25 videos of the CATARACTS training set. The 2020 CATARACTS Semantic Segmentation Challenge, which was a sub-challenge of the 2020 MICCAI Endoscopic Vision (EndoVis) Challenge, presented three sub-tasks to assess participating solutions on anatomical structure and instrument segmentation. Their performance was assessed on a hidden test set of 531 images from 10 videos of the CATARACTS test set.