Search

Background. Poor balance is one of the risk factors for falls in patients with knee osteoarthritis (KOA), which is related to the symptoms. Electroacupuncture (EA) is one of the traditional Chinese conservative methods commonly used to improve the symptoms in patients with KOA. Objective. To assess whether EA increases the dynamic balance during stair negotiation among patients with KOA. Methods. A total of 40 KOA patients were assigned to two groups randomly (true electroacupuncture vs. mock electroacupuncture). Acupoints around the knee were selected in the true electroacupuncture (TEA) group with electrical stimulation (2 Hz). In the mock electroacupuncture (MEA) group, about 2 cm next to the above acupoints, the needles were inserted superficially without electrical stimulation. All the participants received 11 sessions of stimulation treatment in three weeks. The primary outcome was margin of stability (MOS). Secondary outcomes included hip kinematics and kinetics as well as pain. Results. There was no significant difference between the two groups for self-reported pain (p=0.585). During ascent, there was no difference between groups in MOS value in both directions, which was the anterior-posterior (A/P) direction and medial-lateral (M/L) direction at initial contact and toe-off as well as the midstance in the gait cycle, and no difference for the hip kinematics and kinetics between the groups was detected (p>0.05). For descent, at the toe-off event, the TEA group was more unstable as compared to the MEA group in the A/P direction (p=0.029) but not in the M/L direction, and the hip showed a larger internal rotator moment (p=0.049); at the midstance, the TEA group showed a lower abductor moment than the MEA group (p=0.003). Conclusions. Based on the assessment results from the chosen patients with KOA, the TEA did not demonstrate a significant effect in improving the dynamic balance during stair negotiation in comparison with the MEA. This finding does not support EA as a conservative treatment to improve the dynamic balance in such patients.

Traffic simulators act as an essential component in the operating and planning of transportation systems. Conventional traffic simulators usually employ a calibrated physical car-following model to describe vehicles' behaviors and their interactions with traffic environment. However, there is no universal physical model that can accurately predict the pattern of vehicle's behaviors in different situations. A fixed physical model tends to be less effective in a complicated environment given the non-stationary nature of traffic dynamics. In this paper, we formulate traffic simulation as an inverse reinforcement learning problem, and propose a parameter sharing adversarial inverse reinforcement learning model for dynamics-robust simulation learning. Our proposed model is able to imitate a vehicle's trajectories in the real world while simultaneously recovering the reward function that reveals the vehicle's true objective which is invariant to different dynamics. Extensive experiments on synthetic and real-world datasets show the superior performance of our approach compared to state-of-the-art methods and its robustness to variant dynamics of traffic., Accepted for publication by IJCAI 2021

In this paper, a hierarchical energy management strategy (EMS) based on low-pass filter and equivalent consumption minimization strategy (ECMS) is proposed in order to lift energy sources lifespan, power performance and fuel economy for hybrid electrical vehicles equipped with fuel cell, battery and supercapacitor. As for the considered powertrain configuration, fuel cell serves as main energy source, and battery and supercapacitor are regarded as energy support and storage system. Supercapacitor with high power density and dynamic response acts during great power fluctuations, which relives stress on fuel cell and battery. Meanwhile, battery is used to lift the economy of hydrogen fuel. In higher layer strategy of the proposed EMS, supercapacitor is employed to supply peak power and recycle braking energy by using the adaptive low-pass filter method. Meantime, an ECMS is designed to allocate power of fuel cell and battery such that fuel cell can work in a high efficient range to minimize hydrogen consumption in lower layer. The proposed EMS for hybrid electrical vehicles is modeled and verified by advisor-simulink and experiment bench. Simulation and experiment results are given to confirm effectiveness of the proposed EMS of this paper.

In this paper, in order to extend battery lifespan and lift power performance in hybrid batteries/supercapacitors electrical vehicles, a new energy management strategy is proposed based on...

Additional file 1: Supplementary Methods.

Additional file 1: Figure S1. Study profile and IHC tiles dataset generation. Figure S2���S4. Kaplan���Meier survival analysis for patients stratified by TNM stage, age, sex, tumor site, MSI status, and treatment modalities. Figure S5. Using the pipeline to predict the prognosis of one patient with CD3 and CD8 WSIs. Figure S6. Kaplan���Meier survival curve of overall survival of stroma-low vs. stroma-high groups. Figure S7. Stroma-immune score vs. intraepithelial-immune score.

Simulation of the real-world traffic can be used to help validate the transportation policies. A good simulator means the simulated traffic is similar to real-world traffic, which often requires dense traffic trajectories (i.e., with a high sampling rate) to cover dynamic situations in the real world. However, in most cases, the real-world trajectories are sparse, which makes simulation challenging. In this paper, we present a novel framework ImInGAIL to address the problem of learning to simulate the driving behavior from sparse real-world data. The proposed architecture incorporates data interpolation with the behavior learning process of imitation learning. To the best of our knowledge, we are the first to tackle the data sparsity issue for behavior learning problems. We investigate our framework on both synthetic and real-world trajectory datasets of driving vehicles, showing that our method outperforms various baselines and state-of-the-art methods., Comment: Accepted by ECML-PKDD 2020, Best Applied Data Science Paper. 16 pages, 6 figures

The heavy traffic and related issues have always been concerns for modern cities. With the help of deep learning and reinforcement learning, people have proposed various policies to solve these traffic-related problems, such as smart traffic signal control systems and taxi dispatching systems. People usually validate these policies in a city simulator, since directly applying them in the real city introduces real cost. However, these policies validated in the city simulator may fail in the real city if the simulator is significantly different from the real world. To tackle this problem, we need to build a real-like traffic simulation system. Therefore, in this paper, we propose to learn the human routing model, which is one of the most essential part in the traffic simulator. This problem has two major challenges. First, human routing decisions are determined by multiple factors, besides the common time and distance factor. Second, current historical routes data usually covers just a small portion of vehicles, due to privacy and device availability issues. To address these problems, we propose a theory-guided residual network model, where the theoretical part can emphasize the general principles for human routing decisions (e.g., fastest route), and the residual part can capture drivable condition preferences (e.g., local road or highway). Since the theoretical part is composed of traditional shortest path algorithms that do not need data to train, our residual network can learn human routing models from limited data. We have conducted extensive experiments on multiple real-world datasets to show the superior performance of our model, especially with small data. Besides, we have also illustrated why our model is better at recovering real routes through case studies.

Historical features are important in ads click-through rate (CTR) prediction, because they account for past engagements between users and ads. In this paper, we study how to efficiently construct historical features through counting features. The key challenge of such problem lies in how to automatically identify counting keys. We propose a tree-based method for counting key selection. The intuition is that a decision tree naturally provides various combinations of features, which could be used as counting key candidate. In order to select personalized counting features, we train one decision tree model per user, and the counting keys are selected across different users with a frequency-based importance measure. To validate the effectiveness of proposed solution, we conduct large scale experiments on Twitter video advertising data. In both online learning and offline training settings, the automatically identified counting features outperform the manually curated counting features., 7 pages, KDD'19 AutoML Workshop