Search

This is a preprint. The latest version has been published here: https://pubs.rsna.org/doi/10.1148/ryai.230275 Purpose: Sparse-view computed tomography (CT) is an effective way to reduce dose by lowering the total number of views acquired, albeit at the expense of image quality, which, in turn, can impact the ability to detect diseases. We explore deep learning-based artifact reduction in sparse-view cranial CT scans and its impact on automated hemorrhage detection. Methods: We trained a U-Net for artefact reduction on simulated sparse-view cranial CT scans from 3000 patients obtained from a public dataset and reconstructed with varying levels of sub-sampling. Additionally, we trained a convolutional neural network on fully sampled CT data from 17,545 patients for automated hemorrhage detection. We evaluated the classification performance using the area under the receiver operator characteristic curves (AUC-ROCs) with corresponding 95% confidence intervals (CIs) and the DeLong test, along with confusion matrices. The performance of the U-Net was compared to an analytical approach based on total variation (TV). Results: The U-Net performed superior compared to unprocessed and TV-processed images with respect to image quality and automated hemorrhage diagnosis. With U-Net post-processing, the number of views can be reduced from 4096 (AUC-ROC: 0.974; 95% CI: 0.972-0.976) views to 512 views (0.973; 0.971-0.975) with minimal decrease in hemorrhage detection (P<.001) and to 256 views (0.967; 0.964-0.969) with a slight performance decrease (P<.001). Conclusion: The results suggest that U-Net based artifact reduction substantially enhances automated hemorrhage detection in sparse-view cranial CTs. Our findings highlight that appropriate post-processing is crucial for optimal image quality and diagnostic accuracy while minimizing radiation dose., Comment: 11 pages, 6 figures, 1 table

We aim to optimize the binary detection of Chronic Obstructive Pulmonary Disease (COPD) based on emphysema presence in the lung with convolutional neural networks (CNN) by exploring manually adjusted versus automated window-setting optimization (WSO) on computed tomography (CT) images. 7,194 CT images (3,597 with COPD; 3,597 healthy controls) from 78 subjects were selected retrospectively (10.2018-12.2021) and preprocessed. For each image, intensity values were manually clipped to the emphysema window setting and a baseline 'full-range' window setting. Class-balanced train, validation, and test sets contained 3,392, 1,114, and 2,688 images. The network backbone was optimized by comparing various CNN architectures. Furthermore, automated WSO was implemented by adding a customized layer to the model. The image-level area under the Receiver Operating Characteristics curve (AUC) [lower, upper limit 95% confidence] was utilized to compare model variations. Repeated inference (n=7) on the test set showed that the DenseNet was the most efficient backbone and achieved a mean AUC of 0.80 [0.76, 0.85] without WSO. Comparably, with input images manually adjusted to the emphysema window, the DenseNet model predicted COPD with a mean AUC of 0.86 [0.82, 0.89]. By adding a customized WSO layer to the DenseNet, an optimal window in the proximity of the emphysema window setting was learned automatically, and a mean AUC of 0.82 [0.78, 0.86] was achieved. Detection of COPD with DenseNet models was improved by WSO of CT data to the emphysema window setting range.

The purpose of this effort is to investigate if the use of quaternion mathematics can be used to better model and simulate the electromagnetic fields that occur from moving electromagnetic charges. One observed deficiency with the commonly used Maxwell's equations is the issue of polar versus axial vectors; the electromagnetic field E is a polar vector, whereas the magnetic field B is an axial vector, where the direction of rotation remains the same even after the axial vector is inverted. This effort first derived the rotation matrix for quaternion geometry. This rotation matrix was then applied to modeling the magnetic fields at a distance from a source, and comparing it to traditional Maxwell's equations. This effort was taken to model a series of moving charges, an observation aircraft observing a submarine, as well as an eddy current dynamic brake. It was clearly observed that when a moving observer is studying a moving target, differences in the magnetic direction and magnitude can be observed, demonstrating the effectiveness of quaternion mathematics in such applications.

Robots and intelligent systems that sense or interact with the world are increasingly being used to automate a wide array of tasks. The ability of these systems to complete these tasks depends on a large range of technologies such as the mechanical and electrical parts that make up the physical body of the robot and its sensors, perception algorithms to perceive the environment, and planning and control algorithms to produce meaningful actions. Therefore, it is often necessary to consider the interactions between these components when designing an embodied system. This thesis explores work on the task-driven co-optimization of robotics systems in an end-to-end manner, simultaneously optimizing the physical components of the system with inference or control algorithms directly for task performance. We start by considering the problem of optimizing a beacon-based localization system directly for localization accuracy. Designing such a system involves placing beacons throughout the environment and inferring location from sensor readings. In our work, we develop a deep learning approach to optimize both beacon placement and location inference directly for localization accuracy. We then turn our attention to the related problem of task-driven optimization of robots and their controllers. In our work, we start by proposing a data-efficient algorithm based on multi-task reinforcement learning. Our approach efficiently optimizes both physical design and control parameters directly for task performance by leveraging a design-conditioned controller capable of generalizing over the space of physical designs. We then follow this up with an extension to allow for the optimization over discrete morphological parameters such as the number and configuration of limbs. Finally, we conclude by exploring the fabrication and deployment of optimized soft robots., Comment: PhD thesis. arXiv admin note: text overlap with arXiv:2202.04575

A robot's ability to complete a task is heavily dependent on its physical design. However, identifying an optimal physical design and its corresponding control policy is inherently challenging. The freedom to choose the number of links, their type, and how they are connected results in a combinatorial design space, and the evaluation of any design in that space requires deriving its optimal controller. In this work, we present N-LIMB, an efficient approach to optimizing the design and control of a robot over large sets of morphologies. Central to our framework is a universal, design-conditioned control policy capable of controlling a diverse sets of designs. This policy greatly improves the sample efficiency of our approach by allowing the transfer of experience across designs and reducing the cost to evaluate new designs. We train this policy to maximize expected return over a distribution of designs, which is simultaneously updated towards higher performing designs under the universal policy. In this way, our approach converges towards a design distribution peaked around high-performing designs and a controller that is effectively fine-tuned for those designs. We demonstrate the potential of our approach on a series of locomotion tasks across varying terrains and show the discovery novel and high-performing design-control pairs., Comment: For code and videos, see https://sites.google.com/ttic.edu/nlimb

Automatic diagnosis of malignant prostate cancer patients from mpMRI has been studied heavily in the past years. Model interpretation and domain drift have been the main road blocks for clinical utilization. As an extension from our previous work where we trained a customized convolutional neural network on a public cohort with 201 patients and the cropped 2D patches around the region of interest were used as the input, the cropped 2.5D slices of the prostate glands were used as the input, and the optimal model were searched in the model space using autoKeras. Something different was peripheral zone (PZ) and central gland (CG) were trained and tested separately, the PZ detector and CG detector were demonstrated effectively in highlighting the most suspicious slices out of a sequence, hopefully to greatly ease the workload for the physicians., Comment: arXiv admin note: text overlap with arXiv:1903.12331

This work provides a complete framework for the simulation, co-optimization, and sim-to-real transfer of the design and control of soft legged robots. The compliance of soft robots provides a form of "mechanical intelligence" -- the ability to passively exhibit behaviors that would otherwise be difficult to program. Exploiting this capacity requires careful consideration of the coupling between mechanical design and control. Co-optimization provides a promising means to generate sophisticated soft robots by reasoning over this coupling. However, the complex nature of soft robot dynamics makes it difficult to provide a simulation environment that is both sufficiently accurate to allow for sim-to-real transfer, while also being fast enough for contemporary co-optimization algorithms. In this work, we show that finite element simulation combined with recent model order reduction techniques provide both the efficiency and the accuracy required to successfully learn effective soft robot design-control pairs that transfer to reality. We propose a reinforcement learning-based framework for co-optimization and demonstrate successful optimization, construction, and zero-shot sim-to-real transfer of several soft crawling robots. Our learned robot outperforms an expert-designed crawling robot, showing that our approach can generate novel, high-performing designs even in well-understood domains.

Purpose: We aimed to estimate the total lung volume (TLV) from real and synthetic frontal X-ray radiographs on a pixel level using lung thickness maps generated by a U-Net. Methods: 5,959 thorax X-ray computed tomography (CT) scans were retrieved from two publicly available datasets of the lung nodule analysis 2016 (n=656) and the RSNA pulmonary embolism detection challenge 2020 (n=5,303). Additionally, thorax CT scans from 72 subjects (33 healthy: 20 men, mean age [range] = 62.4 [34, 80]; 39 suffering from chronic obstructive pulmonary disease: 25 men, mean age [range] = 69.0 [47, 91]) were retrospectively selected (10.2018-12.2019) from our in-house dataset such that for each subject, a frontal chest X-ray radiograph no older than seven days was available. All CT scans and their corresponding lung segmentation were forward projected using a simulated X-ray spectrum to generate synthetic radiographs and lung thickness maps, respectively. A U-Net model was trained and tested on synthetic radiographs from the public datasets to predict lung thickness maps and consequently estimate TLV. Model performance was further assessed by evaluating the TLV estimations for the in-house synthetic and real radiograph pairs using Pearson correlation coefficient (r) and significance testing. Results: Strong correlations were measured between the predicted and CT-derived ground truth TLV values for test data from synthetic ($n_{Public}$=1,191, r=0.987, P < 0.001; $n_{In-house}$=72, r=0.973, P < 0.001) and real radiographs (n=72, r=0.908, P < 0.001). Conclusion: TLV from U-Net-generated pixel-level lung thickness maps were successfully estimated for synthetic and real radiographs.

Dexterous manipulation remains an open problem in robotics. To coordinate efforts of the research community towards tackling this problem, we propose a shared benchmark. We designed and built robotic platforms that are hosted at MPI for Intelligent Systems and can be accessed remotely. Each platform consists of three robotic fingers that are capable of dexterous object manipulation. Users are able to control the platforms remotely by submitting code that is executed automatically, akin to a computational cluster. Using this setup, i) we host robotics competitions, where teams from anywhere in the world access our platforms to tackle challenging tasks ii) we publish the datasets collected during these competitions (consisting of hundreds of robot hours), and iii) we give researchers access to these platforms for their own projects.

Dexterous manipulation is a challenging and important problem in robotics. While data-driven methods are a promising approach, current benchmarks require simulation or extensive engineering support due to the sample inefficiency of popular methods. We present benchmarks for the TriFinger system, an open-source robotic platform for dexterous manipulation and the focus of the 2020 Real Robot Challenge. The benchmarked methods, which were successful in the challenge, can be generally described as structured policies, as they combine elements of classical robotics and modern policy optimization. This inclusion of inductive biases facilitates sample efficiency, interpretability, reliability and high performance. The key aspects of this benchmarking is validation of the baselines across both simulation and the real system, thorough ablation study over the core features of each solution, and a retrospective analysis of the challenge as a manipulation benchmark. The code and demo videos for this work can be found on our website (https://sites.google.com/view/benchmark-rrc).

This report describes our winning submission to the Real Robot Challenge (https://real-robot-challenge.com/). The Real Robot Challenge is a three-phase dexterous manipulation competition that involves manipulating various rectangular objects with the TriFinger Platform. Our approach combines motion planning with several motion primitives to manipulate the object. For Phases 1 and 2, we additionally learn a residual policy in simulation that applies corrective actions on top of our controller. Our approach won first place in Phase 2 and Phase 3 of the competition. We were anonymously known as `ardentstork' on the competition leaderboard (https://real-robot-challenge.com/leader-board). Videos and our code can be found at https://github.com/ripl-ttic/real-robot-challenge., Comment: The winning submission to Real Robot Challenge (https://real-robot-challenge.com/)

In X-ray imaging, photons are transmitted through and absorbed by the subject, but are also scattered in significant quantities. Previous attempts to use scattered photons for biological imaging used pencil or fan beam illumination. Here we present 3D X-ray Scatter Tomography using full-field illumination. Synchrotron imaging experiments were performed of a phantom and the chest of a juvenile rat. Transmitted and scattered photons were simultaneously imaged with separate cameras; a scientific camera directly downstream of the sample stage, and a pixelated detector with a pinhole imaging system placed at 45${}^\circ$ to the beam axis. We obtained scatter tomogram feature fidelity sufficient for segmentation of the lung and major airways in the rat. The image contrast in scatter tomogram slices approached that of transmission imaging, indicating robustness to the amount of multiple scattering present in our case. This opens the possibility of augmenting full-field 2D imaging systems with additional scatter detectors to obtain complementary modes or to improve the fidelity of existing images without additional dose, potentially leading to single-shot or reduced-angle tomography or overall dose reduction for live animal studies., Comment: 11 pages, 6 figures, Accepted for publication in IEEE Transactions on Medical Imaging

Shared autonomy provides an effective framework for human-robot collaboration that takes advantage of the complementary strengths of humans and robots to achieve common goals. Many existing approaches to shared autonomy make restrictive assumptions that the goal space, environment dynamics, or human policy are known a priori, or are limited to discrete action spaces, preventing those methods from scaling to complicated real world environments. We propose a model-free, residual policy learning algorithm for shared autonomy that alleviates the need for these assumptions. Our agents are trained to minimally adjust the human's actions such that a set of goal-agnostic constraints are satisfied. We test our method in two continuous control environments: Lunar Lander, a 2D flight control domain, and a 6-DOF quadrotor reaching task. In experiments with human and surrogate pilots, our method significantly improves task performance without any knowledge of the human's goal beyond the constraints. These results highlight the ability of model-free deep reinforcement learning to realize assistive agents suited to continuous control settings with little knowledge of user intent., Comment: Published at Robotics: Science and Systems 2020 (RSS)

This study describes a new approach for material decomposition in x-ray imaging, utilising phase contrast to both increase sensitivity to weakly-attenuating samples and to act as a complementary measurement to attenuation, therefore allowing two overlaid materials to be separated. The measurements are captured using the single-exposure, single-grid x-ray phase contrast imaging technique, with a novel correction that aims to remove propagation-based phase effects seen at sharp edges in the attenuation image. The use of a single-exposure technique means that images could be collected in a high-speed sequence. Results are shown for both a known two-material sample and for a biological specimen.

Material decomposition in X-ray imaging uses the energy-dependence of attenuation to virtually decompose an object into specific constituent materials. X-ray phase contrast imaging is a developing technique that can enhance image contrast seen from weakly attenuating objects. In this paper, we combine spectral phase contrast imaging with material decomposition to both better visualise weakly attenuating features and separate them from overlying objects in radiography. We derive an algorithm that performs both tasks simultaneously and verify it against numerical simulations and experimental measurements of ideal two-component samples composed of pure aluminium and poly(methyl methacrylate). Additionally, we showcase first imaging results of a rabbit kitten's lung. The attenuation signal of a thorax, in particular, is dominated by the strongly attenuating bones of the ribcage, which combined with the weak soft tissue signal make it difficult to visualise the fine anatomical structures across the whole lung. In all cases, clean material decomposition was achieved, without residual phase contrast effects, from which we generate an un-obstructed image of the lung, free of bones. Spectral propagation-based phase contrast imaging has the potential to be a valuable tool, not only in future lung research, but also in other systems for which phase contrast imaging in combination with material decomposition proves to be advantageous., Comment: 11 pages, 5 figures

We present a compact atomic clock using cold rubidium atoms based on an isotropic light cooling, a Ramsey microwave interrogation and an absorption detection. Its technology readiness level is suitable to industrial transfer. We use a fibre optical bench, based on a frequency-doubled telecom laser. The isotropic light cooling technique allows us to cool down the atoms in 100 ms and works with a cycle time around 200 ms. We carried out measurements in simulated microgravity and obtained the narrowest fringes ever recorded in microgravity., Comment: 7 pages, 6 figures

The physical design of a robot and the policy that controls its motion are inherently coupled, and should be determined according to the task and environment. In an increasing number of applications, data-driven and learning-based approaches, such as deep reinforcement learning, have proven effective at designing control policies. For most tasks, the only way to evaluate a physical design with respect to such control policies is empirical--i.e., by picking a design and training a control policy for it. Since training these policies is time-consuming, it is computationally infeasible to train separate policies for all possible designs as a means to identify the best one. In this work, we address this limitation by introducing a method that performs simultaneous joint optimization of the physical design and control network. Our approach maintains a distribution over designs and uses reinforcement learning to optimize a control policy to maximize expected reward over the design distribution. We give the controller access to design parameters to allow it to tailor its policy to each design in the distribution. Throughout training, we shift the distribution towards higher-performing designs, eventually converging to a design and control policy that are jointly optimal. We evaluate our approach in the context of legged locomotion, and demonstrate that it discovers novel designs and walking gaits, outperforming baselines in both performance and efficiency.

The ability of robots to estimate their location is crucial for a wide variety of autonomous operations. In settings where GPS is unavailable, measurements of transmissions from fixed beacons provide an effective means of estimating a robot's location as it navigates. The accuracy of such a beacon-based localization system depends both on how beacons are distributed in the environment, and how the robot's location is inferred based on noisy and potentially ambiguous measurements. We propose an approach for making these design decisions automatically and without expert supervision, by explicitly searching for the placement and inference strategies that, together, are optimal for a given environment. Since this search is computationally expensive, our approach encodes beacon placement as a differential neural layer that interfaces with a neural network for inference. This formulation allows us to employ standard techniques for training neural networks to carry out the joint optimization. We evaluate this approach on a variety of environments and settings, and find that it is able to discover designs that enable high localization accuracy., Comment: Appeared at 2017 International Conference on Intelligent Robots and Systems (IROS)