Search

Purpose: This work presents FASTRL, a benchmark set of instrument manipulation tasks adapted to the domain of reinforcement learning and used in simulated surgical training. This benchmark enables and supports the design and training of human-centric reinforcement learning agents which assist and evaluate human trainees in surgical practice. Methods: Simulation tasks from the Fundamentals of Arthroscopic Surgery Training (FAST) program are adapted to the reinforcement learning setting for the purpose of training virtual agents that are capable of providing assistance and scoring to the surgical trainees. A skill performance assessment protocol is presented based on the trained virtual agents. Results: The proposed benchmark suite presents an API for training reinforcement learning agents in the context of arthroscopic skill training. The evaluation scheme based on both heuristic and learned reward functions robustly recovers the ground truth ranking on a diverse test set of human trajectories. Conclusion: The presented benchmark enables the exploration of a novel reinforcement learning-based approach to skill performance assessment and in-procedure assistance for simulated surgical training scenarios. The evaluation protocol based on the learned reward model demonstrates potential for evaluating the performance of surgical trainees in simulation.

The EU’s proposed Corporate Sustainability Due Diligence Directive seeks to enhance responsible business conduct through mandatory risk-based due diligence, administrative supervision by national authorities, and civil liability. If adopted the administrative accountability and remedy scheme envisaged by the proposal will exist alongside another corporate accountability mechanism, the National Contact Points established under the OECD Guidelines for Multinational Enterprises. Having two parallel systems will likely result in divergent findings on due diligence as well as remedial responses. Such inconsistency will neither benefit companies trying to understand what is expected and demanded in terms of risk-based due diligence, nor affected people who are victims of business-related societal harm. To avoid that, this article argues that a coordinated system in which NCPs are assigned some of the administrative supervision tasks envisaged by the CSDDD (or similar measures that may be introduced by states, if the CSDDD is not adopted) will serve accountability and legal certainty, including for remedy, better than two distinct systems whose overlapping competences may undermine those objectives.

Abstract: Background: This study aimed to evaluate the accuracy of functional cine-MRI in detecting abdominal adhesions. Methods: For this study, 89 consecutive patients with adhesion-related complaints after previous abdominal surgery underwent preoperative workup including cine-MRI in transverse and sagittal orientations for a dynamic examination of an induced visceral slide. An abdominal map consisting of nine segments was created to document the location and extent of the adhesion. Cine-MRI and intraoperative findings were correlated. Results: A total of 59 laparotomies and 30 laparoscopies were performed. Four cases required open surgery due to severe adhesions. The use of cine-MRI scan for the detection of adhesions showed an overall accuracy of 90%, a sensitivity of 93%, and a positive predictive value of 96%. The stronger the adhesions, the more accurate the scan findings. Of 44 patients with second-degree MRI scan findings, 50% had second-degree intraoperative findings. Of 35 patients with third- and fourth-degree adhesions on MRI scans, 74% had exactly the same intraabdominal findings at surgery. The MRI scan showed adhesions located in the small intestines (75%), large intestines (35%), abdominal cavity (42%), and reproductive organs (32%). Intraoperatively, adhesions were found in the small intestines (70%), large intestines (40%), abdominal cavity (42%), and reproductive organs (28%). Conclusions: Cine-MRI provides valid preoperative information with respect to extent, location, and strength of intraabdominal adhesions. Cine-MRI is a good alternative for diagnosing abdominal adhesions because objective findings of the scan and intraoperative findings correlate very well with each other.

Novel deep brain stimulation (DBS) systems allow directional and short-pulse stimulation to potentially improve symptoms and reduce side effects. The aim of this study was to investigate the effect of short-pulse and directional stimulation, in addition to a combination of both, in the ventral intermediate thalamus (VIM)/posterior subthalamic area (PSA) on tremor and stimulation-induced side effects in patients with essential tremor.

This short article is about our Giessener Mathematikum, a hands-on minds-on science centre which is unique in the world as an institution solely directed to reaching out to the general public in order to show that mathematics is an interesting and enjoyable subject for everybody. Given many unjustified prejudices against this subject from lay people, it was a challenge to create a large science centre that is attractive to everybody and aims to convince that the beauty, depth and usefulness of mathematics is comprehensible to all, independent of a particular education and particular scientific inclination. We believe we have succeeded, and my purpose is to outline the history and present operation of the Mathematikum, and its various ways of meeting the mentioned challenge.

Responsible innovation in artificial intelligence (AI) calls for public deliberation: well-informed “deep democratic” debate that involves actors from the public, private, and civil society sectors in joint efforts to critically address the goals and means of AI. Adopting such an approach constitutes a challenge, however, due to the opacity of AI and strong knowledge boundaries between experts and citizens. This undermines trust in AI and undercuts key conditions for deliberation. We approach this challenge as a problem of situating the knowledge of actors from the AI industry within a deliberative system. We develop a new framework of responsibilities for AI innovation as well as a deliberative governance approach for enacting these responsibilities. In elucidating this approach, we show how actors from the AI industry can most effectively engage with experts and nonexperts in different social venues to facilitate well-informed judgments on opaque AI systems and thus effectuate their democratic governance.

Based on data regarding the prevalence of Parkinson’s disease (PD), the prevalence of impulsive control disorders (ICD) in PD, and the percentage of PD patients driving a car, it has to be assumed that at least 50,000 PD patients with ICD in Germany actively drive a car. However, these patients might be at risk for unsafe driving due to ICD-related dysfunctions such as failure to resist an impulse or temptation, to control an act or other altered neurobehavioral processes. This study determines the influence of ICD on driving ability in PD. We prospectively compared driving simulator performance of 23 PD patients with and 23 matched patients without ICD. ICD had to be socially compensated and presence was defined clinically for primary and questionnaire-based (QUIP-RS) for post-hocanalyses. Furthermore, between-group comparisons of driving-relevant neuropsychological tests were executed. Except from a lower blinking frequency when changing lanes, overall driving safety of patients with ICD did not differ significantly from those without—regardless of the clinical or QUIP-RS-based ICD definition. ICD severity did not correlate with driving performance, but the latter correlated significantly with mean reaction times and certain neuropsychiatric tests (MoCA, TMT-A, TAP-M “flexibility” and DBQ “error”). Clinically compensated ICD does not seem to impair driving safety in PD patients. Rather, cognitive and attentional deficits appear to be clinical markers for driving uncertainty.

We present a novel low-temperature (30 °C) atomic layer deposition process for hafnium oxide and apply the layers as gate dielectric to fabricate devices out of the thermally sensitive topological insulator HgTe. The key to achieving self-limiting growth at these low temperatures is the incorporation of sufficiently long purge times ( ≥150 s) in the deposition cycles. We investigate the structural and compositional properties of these thin films using X-ray reflectometry and photoelectron spectroscopy, finding a growth rate of 1.6 Å per cycle and an atomic ratio of Hf/O of 1:1.85. In addition, we report on the transport properties of the microstructured devices, which are much enhanced compared to previous device generations. We determine a relative permittivity of ∼15 for our HfO2layers. Our process considerably reduces the thermal load of the samples during microfabrication and can be adapted to a broad range of materials, enabling the fabrication of high-quality gate insulators on various temperature-sensitive materials.

IMPORTANCE: Machine learning algorithms enable the automatic classification of cardiovascular diseases based on raw cardiac ultrasound imaging data. However, the utility of machine learning in distinguishing between takotsubo syndrome (TTS) and acute myocardial infarction (AMI) has not been studied. OBJECTIVES: To assess the utility of machine learning systems for automatic discrimination of TTS and AMI. DESIGN, SETTINGS, AND PARTICIPANTS: This cohort study included clinical data and transthoracic echocardiogram results of patients with AMI from the Zurich Acute Coronary Syndrome Registry and patients with TTS obtained from 7 cardiovascular centers in the International Takotsubo Registry. Data from the validation cohort were obtained from April 2011 to February 2017. Data from the training cohort were obtained from March 2017 to May 2019. Data were analyzed from September 2019 to June 2021. EXPOSURE: Transthoracic echocardiograms of 224 patients with TTS and 224 patients with AMI were analyzed. MAIN OUTCOMES AND MEASURES: Area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the machine learning system evaluated on an independent data set and 4 practicing cardiologists for comparison. Echocardiography videos of 228 patients were used in the development and training of a deep learning model. The performance of the automated echocardiogram video analysis method was evaluated on an independent data set consisting of 220 patients. Data were matched according to age, sex, and ST-segment elevation/non-ST-segment elevation (1 patient with AMI for each patient with TTS). Predictions were compared with echocardiographic-based interpretations from 4 practicing cardiologists in terms of sensitivity, specificity, and AUC calculated from confidence scores concerning their binary diagnosis. RESULTS: In this cohort study, apical 2-chamber and 4-chamber echocardiographic views of 110 patients with TTS (mean [SD] age, 68.4 [12.1] years; 103 [90.4%] were female) and 110 patients with AMI (mean [SD] age, 69.1 [12.2] years; 103 [90.4%] were female) from an independent data set were evaluated. This approach achieved a mean (SD) AUC of 0.79 (0.01) with an overall accuracy of 74.8 (0.7%). In comparison, cardiologists achieved a mean (SD) AUC of 0.71 (0.03) and accuracy of 64.4 (3.5%) on the same data set. In a subanalysis based on 61 patients with apical TTS and 56 patients with AMI due to occlusion of the left anterior descending coronary artery, the model achieved a mean (SD) AUC score of 0.84 (0.01) and an accuracy of 78.6 (1.6%), outperforming the 4 practicing cardiologists (mean [SD] AUC, 0.72 [0.02]) and accuracy of 66.9 (2.8%). CONCLUSIONS AND RELEVANCE: In this cohort study, a real-time system for fully automated interpretation of echocardiogram videos was established and trained to differentiate TTS from AMI. While this system was more accurate than cardiologists in echocardiography-based disease classification, further studies are warranted for clinical application.

Magneto-transport measurements on gated high-mobility heterostructures containing a 60 nm layer of tensile-strained HgTe, a three-dimensional topological insulator, show well-developed Hall quantization from surface states both in the n- as well as in the p-type regime. While the n-type behavior is due to transport in the topological surface state of the material, we find from 8-orbital k·pcalculations that the p-type transport results from massive Volkov–Pankratov states. Their formation prevents the Dirac point and thus the p-conducting topological surface state from being accessible in transport experiments. This interpretation is supported by low-field magneto-transport experiments demonstrating the coexistence of n-conducting topological surface states and p-conducting Volkov–Pankratov states at the relevant gate voltages.

Purpose:Virtual reality-based simulators have the potential to become an essential part of surgical education. To make full use of this potential, they must be able to automatically recognize activities performed by users and assess those. Since annotations of trajectories by human experts are expensive, there is a need for methods that can learn to recognize surgical activities in a data-efficient way. Methods:We use self-supervised training of deep encoder–decoder architectures to learn representations of surgical trajectories from video data. These representations allow for semi-automatic extraction of features that capture information about semantically important events in the trajectories. Such features are processed as inputs of an unsupervised surgical activity recognition pipeline. Results:Our experiments document that the performance of hidden semi-Markov models used for recognizing activities in a simulated myomectomy scenario benefits from using features extracted from representations learned while training a deep encoder–decoder network on the task of predicting the remaining surgery progress. Conclusion:Our work is an important first step in the direction of making efficient use of features obtained from deep representation learning for surgical activity recognition in settings where only a small fraction of the existing data is annotated by human domain experts and where those annotations are potentially incomplete.

Summarizing scheme of GRACE 2.0 and ML-based risk score for 1-year mortality and putative impact. Analyses of baseline variables in ACS patients for risk of mortality at 1 year. Easily available variables from clinical assessment build the basis for the GRACE score. The ML-derived multivariate SPUM-ACS risk score improves risk prediction compared to the GRACE 2.0 score.

We develop an automatic method for synaptic partner identification in insect brains and use it to predict synaptic partners in a whole-brain electron microscopy dataset of the fruit fly. The predictions can be used to infer a connectivity graph with high accuracy, thus allowing fast identification of neural pathways. To facilitate circuit reconstruction using our results, we develop CIRCUITMAP, a user interface add-on for the circuit annotation tool CATMAID.

The Langevin noise approach for quantization of macroscopic electromagnetics for three-dimensional, inhomogeneous environments is compared with normal-mode quantization. Recent works on the applicability of the method are discussed, and several examples are provided showing that for closed systems the Langevin noise approach reduces to the usual cavity mode expansion method when loss is eliminated.

Little is known about the patients’ view on treatment with medical cannabis (MC) for Parkinson’s disease (PD). To assess the PD community’s perception of MC and patients’ experience with MC. Applying a questionnaire-based survey, we evaluated general knowledge and interest in MC as well as the frequency, modalities, efficacy, and tolerability of application. Questionnaires were distributed nationwide via the membership journal of the German Parkinson Association and locally in our clinic to control for report bias. Overall, 1.348 questionnaires (1.123 nationwide, 225 local) were analysed. 51% of participants were aware of the legality of MC application, 28% of various routes of administration (ROA) and 9% of the difference between delta9-tetrahydrocannabinol (Δ9-THC) and cannabidiol (CBD). PD-related cannabis use was reported by 8.4% of patients and associated with younger age, living in large cities and better knowledge about the legal and clinical aspects of MC. Reduction of pain and muscle cramps was reported by more than 40% of cannabis users. Stiffness/akinesia, freezing, tremor, depression, anxiety and restless legs syndrome subjectively improved for more than 20% and overall tolerability was good. Improvement of symptoms was reported by 54% of users applying oral CBD and 68% inhaling THC-containing cannabis. Compared to CBD intake, inhalation of THC was more frequently reported to reduce akinesia and stiffness (50.0% vs. 35.4%; p < 0.05). Interest in using MC was reported by 65% of non-users. MC is considered as a therapeutic option by many PD patients. Nevertheless, efficacy and different ROA should further be investigated.

Gait disturbances and balance remain challenging issues in Parkinsonian patients (PD) with deep brain stimulation (DBS). Short pulse deep brain stimulation (spDBS) increases the therapeutic window in PD patients, yet the effect on gait and postural symptoms remains unknown. We assessed the efficacy of spDBS compared to conventional DBS (cDBS) within the subthalamic nucleus (STN) on Parkinsonian gait. The study was a single-centre, randomized, double-blind, clinical short-term trial. 20 PD patients were studied postoperatively in three different conditions (DBS stimulation switched off (off DBS), spDBS with 40μs pulse width, cDBS with 60μs pulse width) on regular medication. The primary endpoint was the relative difference of gait velocity at self-paced speed during quantitative gait analysis between stimulation conditions. Secondary endpoints were changes of further measures of quantitative gait analysis, Ziegler course, Berg balance scale, FOG questionnaire, MDS-UPDRS, PDQ-39, and HADS. Mixed-model analysis and post-hoc t-tests were performed. Both spDBS and cDBS improved gait velocity at self-paced speed compared to off DBS, however, there was no significant difference between both stimulation modes. Still, nearly half of the patients preferred spDBS over cDBS subjectively. Both stimulation modes were equally effective in improving secondary endpoints of gait, balance, motor and non-motor performances. The use of spDBS and cDBS is equally effective in improving gait and balance in PD and might be beneficial in specified cohorts of PD patients.

In a recent article in Public Relations Inquiry, Jenny Hou has fittingly argued for a stronger focus on agency and actorhood in PR research. We point to two crucial aspects in which we think her arguments need to be extended, namely: (a) embracing the constitutive role of communication for organizational actorhood and agency, and (b) rethinking the role of PR in the constitution of organizational actors. We argue that such extension would allow for an important and radical twist in perspective that highlights a widely neglected question in PR research: What if the collective actorhood status of organizations is not treated as a given but rather arises from communicative attributionsof such actorhood status to social entities? Finally, we develop key implications from this shift in perspective for PR scholarship, education, and practice.

The measurement of machined surfaces directly on ultra-precision turning machines yields to notable advantages, e.g. in correcting formdeviations without re-clamping. In the present work, a one-dimensional probe is moved by the machine axes to optically scan the workpiece. The integration and positioning methodology of the probe within the machine tool coordinate system as well as a measurement simulation model are presented. Simulations and experiments are conducted to identify the influence of positioning errors of the probe regarding the measurement results. The measurement error along the spindle axis, which mainly affects the measured form deviation, depends on the gradient of the geometry to be measured.

Pain is a very frequent symptom with influence on the quality of life in Parkinson’s disease (PD), but is still underdiagnosed and commonly treated only unsystematically. Pain etiology and pain character are often complex and multi-causal, and data regarding treatment recommendations are limited. Pain can be primarily related to PD but frequently it is associated with secondary diseases, such as arthrosis of the spine or joints. However, even basically PD-unrelated pain often is amplified by motor- or non-motor PD symptoms, such as akinesia or depression. Beyond an optimization of anti-parkinsonian treatment, additional pain treatment strategies are usually needed to properly address pain in PD. A careful pain history and diagnostic work-up is essential to rate the underlying pain pathophysiology and to develop a targeted therapeutic concept. This review gives an overview on how pain is treated in PD patients and how patients assess the effectiveness of these therapies; here, the manuscript focuses on pathophysiology-driven suggestions for a multimodal pain management in clinical practice.

In view of the vital role of water, exact knowledge of its dielectric function over a large frequency range is important. We report on currently available measurements of the dielectric function of water at room temperature (25 °C) across the full spectrum: microwave, IR, UV, and X-ray (up to 100 eV). We parameterize the complex dielectric function of water with two Debye (microwave) oscillators and high resolution of IR and UV/X-ray oscillators. We also report dielectric parameters for ice-cold water with a microwave/IR spectrum measured at 0.4 °C, while taking the UV spectrum at 25 °C (assuming negligible temperature dependence in UV). We employ van der Waals dispersion interactions to contrast our model of ice-cold water with earlier models. Air bubbles in water and dissolved gas molecules show attraction toward interfaces rather than repulsion. The van der Waals interaction promotes complete freezing rather than supporting a thin layer of water on ice. We infer that premelting is driven by charge and ion adsorption. Density-based extrapolation from warm to cold water of the dielectric function is satisfactory in microwave but poor (40% error) at IR frequencies.

We present a critical overview comparing theoretical predictions and measurements of van der Waals dispersion forces in media on the basis of the respective Hamaker constants. To quantify the agreement, we complement the reported experimental errors with those for the theoretical predictions, which are because of the uncertainties in the underlying spectroscopic data. Our main finding is that the theoretical errors are often larger than their experimental counterparts. Within these uncertainties, the comparison confirms the standard Lifshitz theory based on the Abraham electromagnetic stress tensor against the recently suggested alternative account on the basis of the Maxwell stress tensor.

ABSTRACTThe business and human rights (BHR) regime defines a corporate responsibility to respect human rights, which applies to all business enterprises, regardless of form, business model, or other characteristics. Its global applicability provides that responsibility with important potential to prevent and address privacy infringements related to social media platforms. This article contributes to scholarship on human rights and technology by focusing on social media platforms’ enabling of posting photos of individuals. Applying the BHR typology of forms of business involvement in harmful human rights impacts—causing them, contributing to them, or being directly linked through business relationships—we identify situations in which social media providers can be involved in privacy infringements. We argue that the particularities of the business model applied by social media providers can turn the connection between social media providers and users into a business relationship, in the sense of the BHR regime. We identify the human rights due diligence implications of infringements caused by users, when the social media provider can be seen to have contributed to or been directly linked to the impact, as well as those caused by the social media provider, and discuss appropriate responses.

Electrical currents in a quantum spin Hall insulator are confined to the boundary of the system. The charge carriers behave as massless relativistic particles whose spin and momentum are coupled to each other. Although the helical character of those states is already established by experiments, there is an open question regarding how those edge states interact with each other when they are brought into close spatial proximity. We employ an inverted HgTe quantum well to guide edge channels from opposite sides of a device into a quasi-one-dimensional constriction. Our transport measurements show that, apart from the expected quantization in integer steps of 2e2/h, we find an additional plateau at e2/h. We combine band structure calculations and repulsive electron–electron interaction effects captured within the Tomonaga–Luttinger liquid model and Rashba spin–orbit coupling to explain our observation in terms of the opening of a spin gap. These results may have direct implications for the study of one-dimensional helical electron quantum optics, and for understanding Majorana and para fermions.