| 1. |
- Kristan, Matej, et al.
(författare)
-
The first visual object tracking segmentation VOTS2023 challenge results
- 2023
-
Ingår i: 2023 IEEE/CVF International conference on computer vision workshops (ICCVW). - : Institute of Electrical and Electronics Engineers Inc.. - 9798350307443 - 9798350307450 ; , s. 1788-1810
-
Konferensbidrag (refereegranskat)abstract
- The Visual Object Tracking Segmentation VOTS2023 challenge is the eleventh annual tracker benchmarking activity of the VOT initiative. This challenge is the first to merge short-term and long-term as well as single-target and multiple-target tracking with segmentation masks as the only target location specification. A new dataset was created; the ground truth has been withheld to prevent overfitting. New performance measures and evaluation protocols have been created along with a new toolkit and an evaluation server. Results of the presented 47 trackers indicate that modern tracking frameworks are well-suited to deal with convergence of short-term and long-term tracking and that multiple and single target tracking can be considered a single problem. A leaderboard, with participating trackers details, the source code, the datasets, and the evaluation kit are publicly available at the challenge website1
|
|
| 2. |
- Zhang, Liang, et al.
(författare)
-
Deep Learning for Additive Screening in Perovskite Light-Emitting Diodes
- 2022
-
Ingår i: Angewandte Chemie International Edition. - : WILEY-V C H VERLAG GMBH. - 1433-7851 .- 1521-3773. ; 61:37
-
Tidskriftsartikel (refereegranskat)abstract
- Additive engineering with organic molecules is of critical importance for achieving high-performance perovskite optoelectronic devices. However, experimentally finding suitable additives is costly and time consuming, while conventional machine learning (ML) is difficult to predict accurately due to the limited experimental data available in this relatively new field. Here, we demonstrate a deep learning method that can predict the effectiveness of additives in perovskite light-emitting diodes (PeLEDs) with a high accuracy up to 96 % by using a small dataset of 132 molecules. This model can maximize the information of the molecules and significantly mitigate the duplicated problem that usually happened with previous models in ML for molecular screening. Very high efficiency PeLEDs with a peak external quantum efficiency up to 22.7 % can be achieved by using the predicated additive. Our work opens a new avenue for further boosting the performance of perovskite optoelectronic devices.
|
|
| 3. |
- Cao, Qi, et al.
(författare)
-
Jointly estimating the most likely driving paths and destination locations with incomplete vehicular trajectory data
- 2023
-
Ingår i: Transportation Research, Part C: Emerging Technologies. - 0968-090X. ; 155
-
Tidskriftsartikel (refereegranskat)abstract
- With an ever-increasing deployment density of probe and fixed sensors, massive vehicular trajectory data is available and show a promising foundation to improve the observability of dynamic traffic demand pattern. However, due to technical and privacy issues, the raw trajectories are not always complete and the paths and destinations between discontinuous trajectory nodes are usually missing. This paper proposes a probabilistic method to jointly reconstruct the missing driving path and destination location of vehicles with incomplete trajectory data. One problem-specific HMM-structured model incorporating spatial and temporal analysis (ST-HMM) is constructed to define the matching probability between observed data and possible movement. Two algorithms, namely candidate set generation and best-match search algorithms, are developed to seek the most possible one as matching result. It can implement end-to-end processing from incomplete trajectory data to complete and connective paths and destinations for the target vehicle. The proposed method is tested based on field-test data and city-wide road network. Compared with two benchmark methods, the proposed method improved the matching accuracy in terms of both path identification and destination inference. Additionally, sensitivity analyses on the size of training dataset and candidate set were performed. We believe that experiment results of these sensitivity analyses can help to provide guidance on data sensing and candidate generation.
|
|
| 4. |
- Cheng, Anying, et al.
(författare)
-
Diagnostic performance of initial blood urea nitrogen combined with D-dimer levels for predicting in-hospital mortality in COVID-19 patients
- 2020
-
Ingår i: International Journal of Antimicrobial Agents. - : ELSEVIER. - 0924-8579 .- 1872-7913. ; 56:3
-
Tidskriftsartikel (refereegranskat)abstract
- The crude mortality rate in critical pneumonia cases with coronavirus disease 2019 (COVID-19) reaches 49%. This study aimed to test whether levels of blood urea nitrogen (BUN) in combination with D-dimer were predictors of in-hospital mortality in COVID-19 patients. The clinical characteristics of 305 COVID19 patients were analysed and were compared between the survivor and non-survivor groups. Of the 305 patients, 85 (27.9%) died and 220 (72.1%) were discharged from hospital. Compared with discharged cases, non-survivor cases were older and their BUN and D-dimer levels were significantly higher ( P < 0.0 0 01). Least absolute shrinkage and selection operator (LASSO) and multivariable Cox regression analyses identified BUN and D-dimer levels as independent risk factors for poor prognosis. Kaplan-Meier analysis showed that elevated levels of BUN and D-dimer were associated with increased mortality (logrank, P 0.0 0 01). The area under the curve for BUN combined with D-dimer was 0.94 (95% CI 0.90-0.97), with a sensitivity of 85% and specificity of 91%. Based on BUN and D-dimer levels on admission, a nomogram model was developed that showed good discrimination, with a concordance index of 0.94. Together, initial BUN and D-dimer levels were associated with mortality in COVID-19 patients. The combination of BUN 4.6 mmol/L and D-dimer > 0.845 mu g/mL appears to identify patients at high risk of in-hospital mortality, therefore it may prove to be a powerful risk assessment tool for severe COVID-19 patients. (c) 2020 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.
|
|
| 5. |
- Liu, Dongjie, et al.
(författare)
-
Enhancing choice-set generation and route choice modeling with data- and knowledge-driven approach
- 2024
-
Ingår i: Transportation Research, Part C: Emerging Technologies. - 0968-090X. ; 162
-
Tidskriftsartikel (refereegranskat)abstract
- Two central and interconnected problems arise in the specification of a ‘‘complete’’ path-based route choice model: choice-set generation and choice from a choice set. Choice-set generation poses a significant challenge in personalization and the enumeration of the full choice set with large size. Despite the continued prevalence of classic econometric models for modeling choices within a given set, this requirement of knowledge-driven modeling necessitates explicit model structures and intricate domain knowledge, which may result in practical biases. In this study, a Conditional Variational AutoEncoder (CVAE)-based choice set generation model is developed, which approximates the probability distribution of the underlying choice set generation process conditional on individual and OD characteristics without relying on expert knowledge. In order to facilitate a friendly integration between knowledge-driven econometric and machine learning approaches, a neural-embedded route choice model (IAP-NERCM) with implicit availability/perception (IAP) of choice alternatives is proposed to automatically capture the heterogeneity of taste parameters without assuming any a priori relationship. Results based on synthetic data show that the proposed models are capable of reproducing the pre-defined coefficients. Field data of GPS data collected in Toyota City is used to future test the proposed models compared to classical statistical models. Results indicate that IAP-NERCM exhibits the ability to recover underlying taste function and achieves the best performance in terms of goodness-of-fit, predictability, and estimation time.
|
|
| 6. |
- Mashausi, Dhahiri Saidi, et al.
(författare)
-
A high efficient FVIII variant corrects bleeding in hemophilia A mouse model
- 2022
-
Ingår i: Biochemical and Biophysical Research Communications - BBRC. - : Elsevier. - 0006-291X .- 1090-2104. ; 637, s. 358-364
-
Tidskriftsartikel (refereegranskat)abstract
- Hemophilia A is a bleeding disorder caused by quantitative or qualitative deficiencies in coagulation factor VIII (FVIII). Low FVIII expression due to its unstable mRNA and binding with immunoglobulin-binding protein (BiP) compromises gene therapy endeavors in hemophilia A. Site-directed mutagenesis have demonstrated an improvement in the expression of FVIII proteins. In this study, a targeted point mutation of Pro at position 290 to Thr (P290T) enhances the in vitro specific activity of B-domain-deleted factor VIII (BDD-FVIII). Hydrodynamic gene delivery of P290T cDNA into FVIII-deficient (FVIII-/-) mice corrected bleeding symptoms. P290T variant resulted in high plasma FVIII coagulant activity 24 h post-gene delivery. Furthermore, bleeding time and average blood loss was significantly reduced in FVIII-/- mice injected with P290T variant, whereas BDD-FVIII and PBS-injected mice experienced prolonged bleeding and excessive blood loss. Histological analysis of the liver biopsies revealed no apparent signs of liver damage. No signs of potential toxicity were seen in mice following mice bodyweights assessment. Altogether, our results demonstrate that the introduction of P290T mutation increases both in vitro and in vivo FVIII coagulant activity, supporting ongoing efforts to develop more effective replacement therapy for hemophilia A.
|
|
| 7. |
- Merugu, Siva Bharath, et al.
(författare)
-
Extracellular AGR2 activates neighboring fibroblasts through endocytosis and direct binding to β-catenin that requires AGR2 dimerization and adhesion domains
- 2021
-
Ingår i: Biochemical and Biophysical Research Communications - BBRC. - : Elsevier. - 0006-291X .- 1090-2104. ; 573, s. 86-92
-
Tidskriftsartikel (refereegranskat)abstract
- Anterior gradient 2 (AGR2) is often overexpressed in several types of cancer. AGR2 is cytoplasmic or secreted as an extracellular signal. Intracellular AGR2 properties and role in cancer have been well studied, but its extracellular function is largely unclear. It has been shown that extracellular AGR2 activates endothelial cells and fibroblasts in culture, but the mechanism of AGR2 signaling is not well elucidated. Here, we report that tumor secreted or externally added AGR2 translocates into cytoplasm by endocytosis, binds to β-catenin and further co-translocates to the nucleus in NIH3T3 fibroblasts. Externally added AGR2 also increased β-catenin expression, stability, and accumulation in the nucleus in both fibroblasts and cancer cells. External AGR2 rescued the expression of β-catenin, which was suppressed by EGFR inhibitor AG1478 indicating an alternative pathway to regulate β-catenin independent of EGFR signal. These effects were abolished when a monoclonal antibody against AGR2 was added to the experiments, confirming the effects are caused by AGR2 only. Putting together, our results show that extracellular AGR2 signaling pathway involves endocytosis mediated cellular translocation, direct binding and regulating β-catenin nuclear accumulation. It is also a target against tumor initiated AGR2 signaling to form and maintain tumor microenvironment.
|
|
| 8. |
- Song, Yuchen, et al.
(författare)
-
A state-based inverse reinforcement learning approach to model activity-travel choices behavior with reward function recovery
- 2024
-
Ingår i: Transportation Research Part C. - : Elsevier BV. - 0968-090X .- 1879-2359. ; 158
-
Tidskriftsartikel (refereegranskat)abstract
- Behaviorally oriented activity-travel choices (ATC) modeling is a principal part of travel demand analysis. Traditional econometric and rule-based methods require explicit model structures and complex domain knowledge. While several recent studies used machine learning models, especially adversarial inverse reinforcement learning (IRL) models, to learn potential ATC patterns with less expert-designed settings, they lack a clear representation of rational ATC behavior. In this study, we propose a data-driven IRL framework based on the maximum causal approach to minimize f-divergences between expert and agent state marginal distributions, which provides a more sample-efficient measurement. In addition, we specify a separate state-only reward function and derive an analytical gradient of the f-divergence objective with respect to reward parameters to ensure good convergences. The method can recover a stationary reward function, which assures the agent to get close to the expert behavior when training from scratch. We validate the proposed model using cellular signaling data from Chongqing, China by comparing with baseline models (behavior cloning, policy-based, and reward-based models) in aspects of policy performance comparison, reward recovery, and reward transfer tasks. The experiment results indicate that the proposed model outperforms existing methods and is relatively less sensitive to the number of expert demonstrations. Qualitative analyses are provided on the fundamental ATC preferences on different features given the reward function recovered from the observed mobility trajectories, and on the learning behaviors under different choices of f-divergence.
|
|
| 9. |
- Wu, Yitong, et al.
(författare)
-
Numerical and experiment study on ventilation performance of the equipment compartment of Alpine high-speed train
- 2023
-
Ingår i: Engineering Applications of Computational Fluid Mechanics. - 1994-2060 .- 1997-003X. ; 17:1
-
Tidskriftsartikel (refereegranskat)abstract
- Alpine High-Speed Train serves on the Lanzhou-Urumqi Line in northwest China, where the terrain is mainly the Gobi Desert. To adapt to this complex environment, the independent-air duct is mainly used for the electrical facilities inside the equipment compartment to prevent the spread of sand particles. This isolated air duct makes ventilation characteristics of the equipment susceptible to the external environment. For this reason, this work aims to clarify and investigate the ventilation characteristic of electrical facilities. A two-step simulation method using IDDES (Improved Delayed Detached Eddy Simulation) and a real-vehicle tracking test using the T-typed pitot tubes were conducted. In the simulation, it is found that the ventilation performance can be influenced by the location of the equipment compartment, facilities and the fan mounted inside. By comparing the results of the test and simulation, they share the same characteristic that the air outlet volume of the converter near the head car is being promoted while it near the tail car is being inhibited. The maximum deviation ratio between the test and simulation is 9%. Therefore, the measurement method in this study is relatively reliable.
|
|
| 10. |
|
|
