Most Down Articles

    Published in last 1 year| In last 2 years| In last 3 years| All| Most Downloaded in Recent Month | Most Downloaded in Recent Year|

    Published in last 1 year
    Please wait a minute...
    For Selected: Toggle Thumbnails
    Numerical Study on Separation of Circulating Tumor Cell Using Dielectrophoresis in a Four-Electrode Microfluidic Device
    WANG Yukuna (王雨坤), DING Xiantingb (丁显廷), ZHANG Zhinana (张执南)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 391-.   DOI: 10.1007/s12204-022-2459-9
    Abstract217)      PDF(pc) (1462KB)(232)       Save
    This numerical study proposes a cell sorting technique based on dielectrophoresis (DEP) in a microfluidic chip. Under the joint effect of DEP and fluid drag, white blood cells and circulating tumor cells are separated because of different dielectric properties. First, the mathematical models of device geometry, single cell, DEP force, electric field, and flow field are established to simulate the cell motion. Based on the simulation model, important boundary parameters are discussed to optimize the cell sorting ability of the device. A proper matching relationship between voltage and flow rate is then provided. The inlet and outlet conditions are also investigated to control the particle motion in the flow field. The significance of this study is to verify the cell separating ability of the microfluidic chip, and to provide a logistic design for the separation of rare diseased cells.
    Reference | Related Articles | Metrics | Comments0
    Integrated Hydraulic-Driven Wearable Robot for Knee Assistance
    ZHAO Yafei (赵亚飞), HUANG Chaoyi (黄超逸), ZOU Yuging(邹玉莹), ZOUKehan(邹可涵), zoU Xiaogang(邹笑阳), XUE .Jiaqi(薛嘉琦), LI Xiaoting(李晓婷), KOH Keng Huat, WANG Xiaojun(王小军), LAI Wai Chiu King(赖伟超), HU Yong(胡勇), XI Ning(席宁), WANG Zheng(王峥)
    J Shanghai Jiaotong Univ Sci    2023, 28 (3): 289-295.   DOI: 10.1007/s12204-023-2602-2
    Abstract333)      PDF(pc) (1156KB)(139)       Save
    Age-related diseases can lead to knee joint misfunction, making knee assistance necessary through the use of robotic wearable braces. However, existing wearable robots face challenges in force transmission and human motion adaptation, particularly among the elderly. Although soft actuators have been used in wearable robots, achieving rapid response and motion control while maintaining portability remains challenging. To address these issues, we propose a soft-robotic knee brace system integrated with multiple sensors and a direct-drive hydraulic actuation system. This approach allows for controlled and rapid force output on the portable hydraulic system. The multi-sensor feedback structure enables the robotic system to collaborate with the human body through human physiological signal and body motion information. The human user tests demonstrate that the knee robot provides assistive torques to the knee joint by being triggered by the electromyography signal and under human motion control.
    Reference | Related Articles | Metrics | Comments0
    Early Detection Methods of Deep Tissue Pressure Injuries: A Systematic Review
    JIA Jingyil (贾菁怡),LI Zhengyi1,2 (李正裔),PENG Linjing1 (彭琳晶),YAO Yifeil* (姚怡飞)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 526-.   DOI: 10.1007/s12204-022-2518-2
    Abstract126)      PDF(pc) (552KB)(120)       Save
    Deep tissue pressure injuries (DTPIs) have witnessed a growing prevalence in hospitals and other health care units especially among individuals with pathological conditions that give rise to restricted mobility, impaired sensation, and reduced tissue tolerance. The etiology of DTPIs has been a subject of controversy, to which several explanatory models have been proposed, including direct mechanical insult, ischemia-reperfusion, lymphatic occlusion, and inflammatory cytokines. In line with these pathophysiological scenarios, ultrasound, subepidermal moisture detection, and biomarker technologies have been proposed as potential early detection methods of DTPIs. This paper provides a systematic review involving these three methods. The conclusion is that combining and implementing these methods at different time periods during DTPIs development and progression respectively is likely to be the most universal, effective and promising way for DTPIs diagnosis.
    Reference | Related Articles | Metrics | Comments0
    Entity Relationship Explanation via Conceptualization
    XIE Chenhao(谢晨昊), LIANG Jiaqing(梁家卿), XIA Yanghua(肖仰华), HWANG Seung-won
    J Shanghai Jiaotong Univ Sci    2023, 28 (6): 695-702.   DOI: 10.1007/s12204-021-2394-1
    Abstract144)      PDF(pc) (608KB)(99)       Save
    Finding an attribute to explain the relationships between a given pair of entities is valuable in many applications. However, many direct solutions fail, owing to its low precision caused by heavy dependence on text and low recall by evidence scarcity. Thus, we propose a generalization-and-inference framework and implement it to build a system: entity-relationship finder (ERF). Our main idea is conceptualizing entity pairs into proper concept pairs, as intermediate random variables to form the explanation. Although entity conceptualization has been studied, it has new challenges of collective optimization for multiple relationship instances, joint optimization for both entities, and aggregation of diluted observations into the head concepts defining the relationship. We propose conceptualization solutions and validate them as well as the framework with extensive experiments.
    Reference | Related Articles | Metrics | Comments0
    Establishment of a Refined Brain Model for Evaluating Implantation Behavior of Neural Electrode and Research of its Simulated Behavior
    HE Yuxcin (贺雨欣), ZHANG Wenguang (张文光), XU Haotian (胥浩天), XU Yifan (徐倚帆), XU Liyue (许李悦)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 401-.   DOI: 10.1007/s12204-022-2523-5
    Abstract159)      PDF(pc) (1512KB)(98)       Save
    The long-term reliability of the neural electrode is closely related to its implantation behavior. In orderto realize the quantitative research of the implantation behavior in a low-cost and accurate way, a refined brainmodel containing meninges is proposed. First, the expected simulation material was selected through measuringthe elastic modulus based on the method of atomic force microscope indentation technique. As a result, the 2%(mass fraction) agarose gel simulated the gray and white matter, the 7 : 1 (volume ratio) polydimethylsiloxane(PDMS) sheet simulated the pia mater, and the polyvinyl chloride (PVC) film simulated the dura mater. Second,based on designing a three-layer structure mold, the brain model was prepared by inverted pouring to realizea flat implantation surface. Finally, the simulation behavior of the brain model was investigated with the ratbrain as a reference. For mechanical behavior of implantation, the implantation force experienced two peaks bothin the brain model and the rat brain, maximum values of which were 10.17 mN and 7.69 mN respectively. Thelarger implantation force in the brain model will increase the strength requirement for the electrode, but reducethe risk of buckling of that in practical application. For humoral dissolution behavior, the dissolution rates ofthe polyethylene glycol (PEG) coating of the electrode in the brain model and rat brain were 7 000 μm3/s and5 600 μm3/s, respectively. The faster dissolution rate in the brain model will cause the larger thickness of thecoating design but provide sufficient implantable time in practical application. The establishment of the brainmodel and the research of its simulated behavior are beneficial to the size design of the electrode substrate andcoating, and research of the implantation mechanism, and further increase the functional life of the electrode.
    Reference | Related Articles | Metrics | Comments0
    Review of Power-Assisted Lower Limb Exoskeleton Robot
    HE Guisong (贺贵松), HUANG Xuegong (黄学功), LI Feng (李峰), WANG Huixing (汪辉兴)
    J Shanghai Jiaotong Univ Sci    2024, 29 (1): 1-15.   DOI: 10.1007/s12204-022-2489-3
    Abstract242)      PDF(pc) (1195KB)(93)       Save
    Power-assisted lower limb exoskeleton robot is a wearable intelligent robot system involving mechanics,materials, electronics, control, robotics, and many other fields. The system can use external energy to provide additional power to humans, enhance the function of the human body, and help the wearer to bear weight that is previously unbearable. At the same time, employing reasonable structure design and passive energy storage can also assist in specific actions. First, this paper introduces the research status of power-assisted lower limb exoskeleton robots at home and abroad, and analyzes several typical prototypes in detail. Then, the key technologies such as structure design, driving mode, sensing technology, control method, energy management, and human-machine coupling are summarized, and some common design methods of the exoskeleton robot are summarized and compared. Finally, the existing problems and possible solutions in the research of power-assisted lower limb exoskeleton robots are summarized, and the prospect of future development trend has been analyzed.
    Reference | Related Articles | Metrics | Comments0
    Progress in Force-Sensing Techniques for Surgical Robots
    GAO Hongyan1,2(高红岩),AI Xiaojie1,2(艾孝杰),SUN Zhenglong3(孙正隆),CHEN Weidong1,2(陈卫东),GAO Anzhu1,2*(高安柱)
    J Shanghai Jiaotong Univ Sci    2023, 28 (3): 370-381.   DOI: 10.1007/s12204-023-2607-x
    Abstract267)      PDF(pc) (1017KB)(87)       Save
    Force sensing is vital for situational awareness and safe interaction during minimally invasive surgery. Consequently, surgical robots with integrated force-sensing techniques ensure precise and safe operations. Over the past few decades, there has been considerable progress in force-sensing techniques for surgical robots. This review summarizes the existing electrically- and optically-based force sensors for surgical robots, including piezoresistive, piezoelectric, capacitive, intensity/phase-modulated, and fiber Bragg gratings. Their principles, applications, advantages, and limitations are also discussed. Finally, we summarize our conclusions regarding state-of-the-art force-sensing technologies for surgical robotics.
    Reference | Related Articles | Metrics | Comments0
    Novel Visualization Tool for Percutaneous Renal Puncture Training Using Augmented Reality Technology
    YU Jiaqi1 (于佳琪),WANG Shuyi1* (王殊轶),WANG Yuqil (王浴屺),XIE Hua2 (谢华), WU Zhangleil (吴张檑),FU Xiaonil (付小妮),MA Bangfeng1 (马邦峰)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 517-.   DOI: 10.1007/s12204-022-2554-y
    Abstract100)      PDF(pc) (1512KB)(84)       Save
    We aim to develop a novel visualization tool for percutaneous renal puncture training based on augmented reality (AR) and compare the needle placement performance of this AR system with ultrasound-guidedfreehand navigation in phantoms. A head-mounted display-based AR navigation system was developed usingthe Unity3D software and Visual Studio to enable the overlay of the preoperative needle path and the complexanatomical structures onto a phantom in real time. The spatial location of the stationary phantom and the percutaneous instrument motion were traced by a Qualisys motion capture system. To evaluate the tracking accuracy,15 participants (7 males and 8 females) performed a single needle insertion using AR navigation (the number ofpunctures n = 75) and ultrasound-guided freehand navigation (n = 75). The needle placement error was measuredas the Euclidean distance between the actual needle tip and the virtual target by MicronTracker. All participantsdemonstrated a superior needle insertion efficiency when using the AR-assisted puncture method compared withthe ultrasound-guided freehand method. The needle insertion error of the ultrasound-guided method showed anincreased error compared with the AR method (5.54 mm ± 2.59 mm, 4.34 mm ± 2.10 mm, respectively, p < 0.05).The ultrasound-guided needle placements showed an increased time compared with the AR method (19.08 s ±3.59 s, 15.14 s ± 2.72 s, respectively, p < 0.000 1). Our AR training system facilitates the needle placement performance and solves hand-eye coordination problems. The system has the potential to increase efficiency andeffectiveness of percutaneous renal puncture training.
    Reference | Related Articles | Metrics | Comments0
    Calibration Technology of Optical Fiber Strain Sensor
    CHEN Gang(陈刚), LIU Hongyue(刘宏月), GAO Ruiriang(高瑞翔)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 551-559.   DOI: 10.1007/s12204-022-2406-9
    Abstract133)      PDF(pc) (1122KB)(81)       Save
    As one of the hotspots of sensing technology at present, optical fiber sensor has the characteristics of small size, anti-electromagnetic interference, and easy networking, which plays an irreplaceable role in multiphysics parameter monitoring of complex electromagnetic environments. The precise calibration of the optical fiber strain sensor has great practical value in prolonging the survival rate of the sensor, improving the measurement accuracy, and meeting the needs of long-term monitoring. By reviewing the research status of strain sensor calibration method and fiber optic strain sensor calibration method, the advantages and disadvantages of the main methods are analyzed separately from the static and dynamic perspectives, and the development prospect of the calibration technology of optic fiber strain sensor is summarized.
    Reference | Related Articles | Metrics | Comments0
    Dlung: Unsupervised Few-Shot Diffeomorphic Respiratory Motion Modeling
    CHEN Peizhi1,2* (陈培芝), GUO Yifan1 (郭逸凡),WANG Dahan1,2 (王大寒), CHEN Chinling1,3,4* (陈金铃)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 536-.   DOI: 10.1007/s12204-022-2525-3
    Abstract112)      PDF(pc) (1720KB)(78)       Save
    Lung image registration plays an important role in lung analysis applications, such as respiratory motion modeling. Unsupervised learning-based image registration methods that can compute the deformation without the requirement of supervision attract much attention. However, it is noteworthy that they have two drawbacks: they do not handle the problem of limited data and do not guarantee diffeomorphic (topologypreserving) properties, especially when large deformation exists in lung scans. In this paper, we present an unsupervised few-shot learning-based diffeomorphic lung image registration, namely Dlung. We employ fine-tuning techniques to solve the problem of limited data and apply the scaling and squaring method to accomplish the diffeomorphic registration. Furthermore, atlas-based registration on spatio-temporal (4D) images is performed and thoroughly compared with baseline methods. Dlung achieves the highest accuracy with diffeomorphic properties. It constructs accurate and fast respiratory motion models with limited data. This research extends our knowledge of respiratory motion modeling.
    Reference | Related Articles | Metrics | Comments0
    Numerical Simulation Method of Scoliosis Orthosis Considering Muscle Factor
    LI Jian1,2(李健),ZHU Ye1 (朱晔),GUAN Tianmin1*(关天民)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 486-.   DOI: 10.1007/s12204-022-2535-1
    Abstract73)      PDF(pc) (1382KB)(76)       Save
    Adolescent idiopathic scoliosis seriously affects the physical and mental health of adolescents. In thepast, the research on therapeutic orthosis ignored the influence of muscle factors. Aimed at this problem, basedon the principle of reverse engineering, through the spine computed tomography data model of three-dimensionalreconstruction, muscle forces around the spine are imported into the spinal muscle force model and AnyBodysoftware is used for simulation. The geometric similarity and biomechanical effectiveness of the established modelare verified. In order to obtain the relationship among the applied orthopedic force, Cobb angle and vertebraldisplacement, a finite element model conforming to spinal anatomy is established, and then the biomechanicalanalysis of the finite element model of the scoliosis is carried out. Reasonable control of paravertebral muscles canplay a positive role in orthopedic treatment, and the fitting equation can provide a reference for doctors to applythe orthopedic force on patient.
    Reference | Related Articles | Metrics | Comments0
    Real-Time Calculation Method for Temperature Distribution of Temperature-Controlled Radiofrequency Ablation
    WANG Xuewei (王雪维),WANG Yifei (王逸飞),ZHANG Aili* (张爱丽)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 411-.   DOI: 10.1007/s12204-022-2481-y
    Abstract138)      PDF(pc) (1220KB)(72)       Save
    Precise temperature control and temperature distribution prediction are of great significance forradiofrequency ablation. This research proposes a real-time calculation method for the temperature distribution of radiofrequency ablation combined with proportional-integral temperature control. The thermo-electricalcoupling was simplified into a linear relationship based on the study of the influence of temperature-dependentelectrical conductivity and thermal conductivity on the PI-controlled radiofrequency ablation temperature distribution, which increases the computational efficiency by 150 times. The average calculation time for radiofrequencyablation of 10 min is about 23 s, and the difference between the calculation results of this method and that fromCOMSOL Multiphysics is no more than 1 ?C. This method is not only used for single-probe, but also for doubleprobe radiofrequency ablation in this paper.
    Reference | Related Articles | Metrics | Comments0
    High Curvature Stripe Profile Extraction Algorithm of Line Structured Light Measuring System
    SUN Hao (孙昊), DU Xuan (杜宣), LÜ Na(吕娜), CUI Bin(崔斌), ZHA Hui(赵辉)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 560-568.   DOI: 10.1007/s12204-022-2476-8
    Abstract87)      PDF(pc) (1375KB)(65)       Save
    In the line structured light measuring system, the accuracy of the process of laser stripe directly affects the measurement results. Therefore, the extraction algorithm for the laser stripe, especially the surface with high reflection and high curvature, is very important. The imaging principle of line structured light, the light intensity distribution law of laser stripe and the extraction algorithm have been studied, and a stripe profile extraction method based on real light intensity distribution has been proposed. In this algorithm, fast region of interest extraction, stripe width estimation, and adaptive filtering on the striped image are performed. Then the energy center of the stripe at the sub-pixel level is extracted. Finally, the low-quality center points are eliminated, and the context information is used to recover the missing central points. Simulated images generated based on the imaging principle of line structured light and real experimental images were used to evaluate the accuracy and repeatability of the proposed method. The results show that the method behaves excellently at the edges of high-curvature stripes; the maximum error is only 1.6 pixels, which is 1/10 of the classic Steger algorithm; the experiment repeatability is only 8.8 μm, which is 2.7 times that of the Steger method. Therefore, the proposed method improves the accuracy of object contour extraction, and it is especially suitable for contour detection of objects with high curvature.
    Reference | Related Articles | Metrics | Comments0
    Boosting Unsupervised Domain Adaptation with Soft Pseudo-Label and Curriculum Learning
    ZHANG Shengjia(张晟嘉), LIN Tiancheng(林天成), XU Yi(徐奕)
    J Shanghai Jiaotong Univ Sci    2023, 28 (6): 703-716.   DOI: 10.1007/s12204-022-2487-5
    Abstract92)      PDF(pc) (963KB)(65)       Save
    By leveraging data from a fully labeled source domain, unsupervised domain adaptation (UDA) improves classification performance on an unlabeled target domain through explicit discrepancy minimization of data distribution or adversarial learning. As an enhancement, category alignment is involved during adaptation to reinforce target feature discrimination by utilizing model prediction. However, there remain unexplored problems about pseudo-label inaccuracy incurred by wrong category predictions on target domain, and distribution deviation caused by overfitting on source domain. In this paper, we propose a model-agnostic two-stage learning framework, which greatly reduces flawed model predictions using soft pseudo-label strategy and avoids overfitting on source domain with a curriculum learning strategy. Theoretically, it successfully decreases the combined risk in the upper bound of expected error on the target domain. In the first stage, we train a model with distribution alignment-based UDA method to obtain soft semantic label on target domain with rather high confidence. To avoid overfitting on source domain, in the second stage, we propose a curriculum learning strategy to adaptively control the weighting between losses from the two domains so that the focus of the training stage is gradually shifted from source distribution to target distribution with prediction confidence boosted on the target domain. Extensive experiments on two well-known benchmark datasets validate the universal effectiveness of our proposed framework on promoting the performance of the top-ranked UDA algorithms and demonstrate its consistent superior performance.
    Reference | Related Articles | Metrics | Comments0
    Novel Concentric Tube Robot Based on Double-Threaded Helical Gear Tube
    CHEN Weichi(陈韦池), LIU Haocheng(刘浩城), LI Zijian(李子建), GUO Jing, (郭靖), ZHAI Zhenkun(翟振坤), MENG Wei(孟伟)
    J Shanghai Jiaotong Univ Sci    2023, 28 (3): 296-306.   DOI: 10.1007/s12204-023-2595-x
    Abstract217)      PDF(pc) (2087KB)(63)       Save
    Nasopharyngeal carcinoma is a malignant tumor originating from the nasal mucosa. It is a malignant tumor of the head and neck. Concentric tube robot (CTR), as it can form a complicated shape and access hardto-reach lesions, is often used in minimally invasive surgeries. However, some CTRs are bulky because of their transmission design. In this paper, a light CTR based on double-threaded helical gear tube is proposed. Such a CTR is less cumbersome than the traditional CTR as its actuation unit is compact and miniaturized. The mapping relationship between the gear tube attitude and motor output angle is obtained by kinematic analysis. The precision, stability, and repeatability of the driving mechanism are tested. The experimental results show that the positioning error in the translation test is less than 0.3 mm, the rolling angle error in the stability test is less than 0.6?, and the error in the translation repeatability test is less than 0.005 mm. Finally, a tip-targeting test is performed using the new CTR, which verifies the feasibility of the CTR for surgeries.
    Reference | Related Articles | Metrics | Comments0
    Novel Scheme for Essential Proteins Identification Based on Improved Multicriteria Decision Making
    LU Pengli1* (卢鹏丽),CHEN Yuntian1 (陈云天), LIAO Yonggang2 (廖永刚)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 418-.   DOI: 10.1007/s12204-023-2584-0
    Abstract91)      PDF(pc) (1360KB)(63)       Save
    Identifying essential proteins from protein-protein interaction networks is important for studies onbiological evolution and new drug’s development. Most of the presented criteria for prioritizing essential proteinsonly focus on a certain attribute of the proteins in the network, which suffer from information loss. In order toovercome this problem, a relatively comprehensive and effective novel method for essential proteins identificationbased on improved multicriteria decision making (MCDM), called essential proteins identification-technique fororder preference by similarity to ideal solution (EPI-TOPSIS), is proposed. First, considering different attributes ofproteins, we propose three methods from different aspects to evaluate the significance of the proteins: gene-degreecentrality (GDC) for gene expression sequence; subcellular-neighbor-degree centrality (SNDC) and subcellular-indegree centrality (SIDC) for subcellular location information and protein complexes. Then, betweenness centrality(BC) and these three methods are considered together as the multiple criteria of the decision-making model.Analytic hierarchy process is used to evaluate the weights of each criterion, and the essential proteins are prioritizedby an ideal solution of MCDM, i.e., TOPSIS. Experiments are conducted on YDIP, YMIPS, Krogan and BioGRIDnetworks. The results indicate that EPI-TOPSIS outperforms several state-of-the-art approaches for identifyingthe essential proteins through the performance measures.
    Reference | Related Articles | Metrics | Comments0
    Gene Expression Profiling Identifies Potential Biomarkers for Colorectal Cancer Using NanoString nCounter Assay
    ZHAO Hui (赵晖),WEN Baiqing (文柏清),KANG Yani*(康亚妮)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 432-.   DOI: 10.1007/s12204-022-2440-7
    Abstract91)      PDF(pc) (627KB)(63)       Save
    Colorectal cancer (CRC) is one of the most common malignancies worldwide. Development of predictivemolecular markers may help to achieve the best outcome in clinic. The purpose of this study is to identify thedifferentially expressed genes and new potential predictive and prognostic molecular biomarkers for CRC. Inthis study, CRC and matched normal tissues acquired from the same patient were used to extract total RNA.The NanoString nCounter assay was applied to determine the differentially expressed genes. The results werethen validated by using the Cancer Genome Atlas data. Finally, we identified 27 genes that revealed significantcorrelation with CRC in the tumor tissue. Several genes in the pan-cancer panel showed significant differentialexpression, which were more universal than others in the CRC tissue. Since some of them have not been reportedas being directly related to CRC yet, future mechanism studies can be designed based on this study. Our studydemonstrated NanoString nCounter assay could serve as an alternative approach for gene expression analysis andidentified several unreported differently expressed genes in CRC patients, which may provide some importantclues for more in-depth study of CRC and serve as potential predictive molecular biomarkers for clinical diagnosisapplication.
    Reference | Related Articles | Metrics | Comments0
    Improving Colonoscopy Polyp Detection Rate Using Semi-Supervised Learning
    YAO Leyul (姚乐宇),HE Fan1,3 (何凡), PENG Haixia2* (彭海霞), WANG Xiaofeng2 (王晓峰),ZHOU Lu2(周璐), HUANG Xiaolin1,3* (黄晓霖)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 441-.   DOI: 10.1007/s12204-022-2519-1
    Abstract126)      PDF(pc) (497KB)(61)       Save
    Colorectal cancer is one of the biggest health threats to humans and takes thousands of lives every year.Colonoscopy is the gold standard in clinical practice to inspect the intestinal wall, detect polyps and remove polypsin early stages, preventing polyps from becoming malignant and forming colorectal cancer instances. In recentyears, computer-aided polyp detection systems have been widely used in colonoscopies to improve the qualityof colonoscopy examination and increase the polyp detection rate. Currently, the most efficient computer-aidedsystems are built with machine learning methods. However, developing such a computer-aided detection systemrequires experienced doctors to label a large number of image data from colonoscopy videos, which is extremelytime-consuming, laborious and expensive. One possible solution is to adopt a semi-supervised learning, which canbuild a detection system on a dataset where part of its data is not necessary to be labeled. In this paper, on thebasis of state-of-the-art object detection method and semi-supervised learning technique, we design and implementa semi-supervised colonoscopy polyp detection system containing four main steps: running standard supervisedtraining with all labeled data; running inference on unlabeled data to obtain pseudo labels; applying a set ofstrong augmentation to both unlabeled data and pseudo label; combining labeled data, and unlabeled data withits pseudo labels to retrain the detector. The semi-supervised learning system is evaluated both on public datasetand our original private dataset and proves its effectiveness. Also, the inference speed of the semi-supervisedlearning system can meet the requirement of real-time operation.
    Reference | Related Articles | Metrics | Comments0
    Lateral Force Characteristics of Cartridge Valve Core
    HAN Heyong1(韩贺永),ZHANG Jianru1(张建茹), PAN Siji1(潘思意),LI Yugui2*(李玉贵),MA Lifeng1(马立峰),LIU Shirui3(刘实睿)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 604-610.   DOI: 10.1007/s12204-021-2402-5
    Abstract45)      PDF(pc) (978KB)(59)       Save
    he valve core is readily subjected to a large lateral force which affects the dynamic response speed. Here, a new type of cartridge valve core structure is proposed to solve this problem. The numerical simulation method is applied to analyze the flow characteristics of clearance flow field on velocity distribution, pressure distribution, valve core motion speed, and leakage. The results using computational fluid dynamics (CFD) show that the guide groove is set on the surface of the cartridge valve core, increasing the connecting length of the valve core, forming a uniform radial pressure distribution and velocity distribution, effectively reducing the lateral force, and at the same time ensuring that the leak is not too big. These findings provide theoretical guidance and a basis for optimizing cartridge valve to reduce the occurrence of jamming and improve the response frequency.
    Reference | Related Articles | Metrics | Comments0
    Fault Diagnosis for Rolling Element Bearing in Dataset Bias Scenario
    HOU Liangsheng(侯良生),ZHANG Jundong*(张均东)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 638-651.   DOI: 10.1007/s12204-021-2320-6
    Abstract25)      PDF(pc) (1575KB)(59)       Save
    Recently, data-driven methods, especially deep learning, outperform other methods for rolling element bearing (REB) fault diagnosis. Nevertheless, most research work assumes that REB dataset is unbiased. In the real industry applications, the dataset bias exists with REB owing to varying REB working conditions and noise interference. Recently proposed adversarial discriminative domain adaptation (ADDA) is an increasingly popular incarnation to solve dataset bias problem. However, it mainly devotes to realizing domain alignments, and ignores class-level alignments; it can cause degradation of classification performance. In this study, we propose a new REB fault diagnosis model based on improved ADDA to address dataset bias. The proposed diagnosis model realizes domain- and class-level alignments in dataset bias scenario; it consists of two feature extractors, a domain discriminator, and two label classifiers. The feature extractors and domain discriminator are trained in an adversarial manner to minimize the domain difference in feature extractors. The domain discrepancy in label classifier is reduced by minimizing correlation alignment (CORAL) loss. We evaluate the proposed model on the Case Western Reserve University (CWRU) bearing dataset and Paderborn University bearing dataset. The proposed method yields better results than other methods and has good prospects for industrial applications.
    Reference | Related Articles | Metrics | Comments0
    Medicine-Engineering Interdisciplinary Research Based on Bibliometric Analysis: A Case Study on Medicine-Engineering Institutional Cooperation of Shanghai Jiao Tong University
    WANG Qingwen (王庆稳),CUI Tingting (崔婷婷),DENG Peiwen* (邓珮雯)
    J Shanghai Jiaotong Univ Sci    2023, 28 (6): 841-856.   DOI: 10.1007/s12204-022-2418-5
    Abstract44)      PDF(pc) (1829KB)(52)       Save
    This article aims to provide reference for medicine-engineering interdisciplinary research. Targeted at the scientific literature and patent literature published by Shanghai Jiao Tong University, this article attempts to set up co-occurrence matrix of medicine-engineering institutional information which was extracted from address fields of the papers, so as to construct the medicine-engineering intersection datasets. The dataset of scientific literature was analyzed using bibliometrics and visualization methods from multiple dimensions, and the most active factors, such as trends of output, journal and subject distribution, were identified from the indicators of category normalized citation impact (CNCI), times cited, keywords, citation topics and the degree of medicineengineering interdisplinary. Research on hotspots and trends was discussed in detail. Analyses of the dataset of patent literature showed research themes and measured the degree for technology convergence of medicineengineering.
    Reference | Related Articles | Metrics | Comments0
    Review of Key Technologies for Developing Personalized Lower Limb Rehabilitative Exoskeleton Robots
    TAO Jing, (陶璟), ZHOU Zhenhuan (周振欢)
    J Shanghai Jiaotong Univ Sci    2024, 29 (1): 16-28.   DOI: 10.1007/s12204-022-2452-3
    Abstract83)      PDF(pc) (1179KB)(51)       Save
    Rehabilitative training and assistance to daily living activities play critical roles in improving the life quality of lower limb dyskinesia patients and older people with motor function degeneration. Lower limb rehabilitative exoskeleton has a promising application prospect in support of the above population. In this paper, critical technologies for developing lower limb rehabilitative exoskeleton for individualized user needs are identi- fied and reviewed, including exoskeleton hardware modularization, bionic compliant driving, individualized gait planning and individual-oriented motion intention recognition. Inspired by the idea of servitization, potentials in exoskeleton product-service system design and its enabling technologies are then discussed. It is suggested that future research will focus on exoskeleton technology and exoskeleton-based service development oriented to an individual’s physical features and personalized requirements to realize better human-exoskeleton coordination in terms of technology, as well as accessible and high-quality rehabilitation and living assistance in terms of utility.
    Reference | Related Articles | Metrics | Comments0
    Enhancement of Pinching Grasping Robustness Using a Multi-Structure Soft Gripper
    LI Linlin (李林霖), GAO Feiyang (高飞扬), ZHENG Xiongfei(郑雄飞), ZHANG Liming(张黎明), LI Shijie (李世杰), WANG Heran(王赫然)
    J Shanghai Jiaotong Univ Sci    2023, 28 (3): 307-311.   DOI: 10.1007/s12204-022-2508-4
    Abstract132)      PDF(pc) (2071KB)(46)       Save
    Recently, soft grippers have garnered considerable interest in various fields, such as medical rehabilitation, due to their high compliance. However, the traditional PneuNet only reliably grasps medium and large objects via enveloping grasping (EG), and cannot realize pinching grasping (PG) to stably grasp small and thin objects as EG requires a large bending angle whereas PG requires a much smaller one. Therefore, we proposed a multi-structure soft gripper (MSSG) with only one vent per finger which combines the PneuNet in the proximal segment with the normal soft pneumatic actuator (NSPA) in the distal segment, allowing PG to be realized without a loss in EG and enhancing the robustness of PG due to the height difference between the distal and proximal segments. Grasping was characterized on the basis of the stability (finger bending angle describes) and robustness (pull-out force describes), and the bending angle and pull-out force of MSSG were analyzed using the finite element method. Furthermore, the grasping performance was validated using experiments, and the results demonstrated that the MSSG with one vent per finger was able to realize PG without a loss in EG and effectively enhance the PG robustness.
    Reference | Related Articles | Metrics | Comments0
    Hysteresis Modeling and Compensation for Distal Shaft Deflection of Flexible Ureteroscope
    HUA Penga (华鹏), SHU Xiongpenga (舒雄鹏),XIE Lea,b* (谢叻)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 507-.   DOI: 10.1007/s12204-022-2505-7
    Abstract73)      PDF(pc) (1652KB)(44)       Save
    Flexible ureteroscopy (FURS) has been widely used in the diagnosis and treatment of upper urinarytract diseases. The key operation of FURS is that the surgeon manipulates the distal shaft of flexible ureteroscopeto a specific target for diagnosis and treatment. However, the hysteresis of flexible ureteroscope may be one ofthe most important factors that degrade the manipulation accuracy and the surgeon usually spends a long timenavigating the distal shaft during surgery. In this study, we obtained hysteresis curves of distal shaft deflectionfor the flexible ureteroscope through extensive repeated experiments. Then, two methods based on piecewiselinear approximation and long short-term memory neural network were employed to model the hysteresis curves.On this basis, we proposed two hysteresis compensation strategies for the distal shaft deflection. Finally, wecarried out hysteresis compensation experiments to verify the two proposed compensation strategies. Experimentalresults showed that the hysteresis compensation strategies can significantly improve position accuracy with meancompensation errors of no more than 5?.
    Reference | Related Articles | Metrics | Comments0
    Numerical Study of Bifurcated Blood Flow in Three Different Blood Viscosity Models
    WU Hui(吴 辉), FU Rongchang* (富荣昌), YANG Xiaoyu (杨晓玉), LI Xianzheng (李现政), WANG Zhaoyao (王召耀)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 450-.   DOI: 10.1007/s12204-022-2512-8
    Abstract83)      PDF(pc) (1759KB)(44)       Save
    This study aimed to analyze the hemodynamic effects of bifurcated vessels using different bloodviscosity models. Three-dimensional models of bifurcated vessels in the popliteal artery were constructed basedon CT images, and hemodynamic parameters of the Newtonian, Casson, and two-phase models were calculatedby the computational fluid dynamics method. Blood flowed through the popliteal artery. Blood flow velocitychanged after the bifurcated vessel, with accelerated blood flow velocity in the anterior tibial artery. A lowvelocity vortex region with a region of low wall shear stress (WSS) was generated outside the bifurcated vessel.Local non-Newtonian importance factors of great than 1 (i.e., IL > 1) occurred during the cardiac cycle, andIL > 1.75 occurred at the beginning and end of the cycle. Compared with the Casson and two-phase models, theNewtonian model has a larger vortex region and lower WSS. Low-velocity vortex regions and low WSS regionsin the bifurcated vessels may contribute to the development of atherosclerosis. Blood exhibited non-Newtonianfluid properties in bifurcated vessels (IL > 1), and the effect of non-Newtonian properties was more pronouncedat the beginning and end of heartbeats (IL > 1.75). The Newtonian model predicts a higher risk of atherosclerosisformation and the effect of non-Newtonian properties of blood should be considered in hemodynamic studies.
    Reference | Related Articles | Metrics | Comments0
    Parameter Optimization and Precision Enhancement of Dual-Coil Eddy Current Sensor
    ZHANG Zhenning1(张振宁),LIU Qiang2(刘强), Lü Chunfeng3(吕春峰),MAO Yimeil(毛义梅),TAo Weil(陶卫),ZHAO Huil*(赵辉)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 596-603.   DOI: 10.1007/s12204-022-2511-9
    Abstract71)      PDF(pc) (948KB)(43)       Save
    To enhance the measurement precision of eddy current sensor in particular environments such as extreme temperature changes and limited available space in aerospace, we optimized the structural parameters of the traditional dual-coil eddy current sensor probe by electromagnetic field analysis and finite element simulation modeling, and further presented the criteria for determining the optimal coil distance of the dual-coil probe. The simulation results are verified by setting up an experimental platform. For the extreme temperature environment, the displacement measurement error caused by the full range temperature variation of the dual-coil sensor under the optimal distance is less than 21.0% of that of the single-coil sensor. On this basis, we analyzed and verified the thermal stability of the structurally optimized dual-coil eddy current sensor. After temperature compensation, the displacement measurement accuracy can reach 14.9 times more accurate than that of the single-coil sensor. The method proposed in this paper can provide a design reference for the structural optimization of the axial dual-coil eddy current sensor probe.
    Reference | Related Articles | Metrics | Comments0
    Applications of Polypeptide Hydrogels in Cartilage-Regeneration Engineering
    HU Yinghan1 (胡颖涵),ZHU Zegu1 (朱泽宇), TENG Lin2 (滕林), HE Yushi3 (何雨石),ZOU Derong1 (邹德荣),LU Jiayu1*(陆家瑜)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 468-.   DOI: 10.1007/s12204-022-2507-5
    Abstract92)      PDF(pc) (3066KB)(41)       Save
    Articular cartilage defects are considered to be associated with the development of osteoarthritis. Research on relevant tissue regeneration is important in the treatment of osteoarthritis. The scaffolds applied incartilage regeneration should have good histocompatibility and mechanical properties, as well as no cytotoxicity,and promote the proliferation and differentiation of seed cells. Different combinations of peptide sequences inpolypeptide hydrogels endow them with unique characteristics including excellent biodegradability and accuratesimulation of the extracellular matrix of chondrocytes to maintain the stability of the chondrogenic phenotypeand facilitate articular hyaline cartilage regeneration. Thus, the application of polypeptide hydrogels for cartilage regeneration has a bright future. In this study, the research progress of polypeptide hydrogels used incartilage-regeneration engineering is systematically reviewed. The characteristics, limitations, and prospects ofthese materials are evaluated.
    Reference | Related Articles | Metrics | Comments0
    Multiple Detection Model Fusion Framework for Printed Circuit Board Defect Detection
    WU Xingl(武星), ZHANG Qingfeng(张庆丰), WANG Jianjia(王健嘉), YAO Junfeng(姚骏峰), Guo Yike.(郭毅可)
    J Shanghai Jiaotong Univ Sci    2023, 28 (6): 717-727.   DOI: 10.1007/s12204-022-2471-0
    Abstract60)      PDF(pc) (1870KB)(39)       Save
    The printed circuit board (PCB) is an indispensable component of electronic products, which determines the quality of these products. With the development and advancement of manufacturing technology, the layout and structure of PCB are getting complicated. However, there are few effective and accurate PCB defect detection methods. There are high requirements for the accuracy of PCB defect detection in the actual production environment, so we propose two PCB defect detection frameworks with multiple model fusion including the defect detection by multi-model voting method (DDMV) and the defect detection by multi-model learning method (DDML). With the purpose of reducing wrong and missing detection, the DDMV and DDML integrate multiple defect detection networks with different fusion strategies. The effectiveness and accuracy of the proposed framework are verified with extensive experiments on two open-source PCB datasets. The experimental results demonstrate that the proposed DDMV and DDML are better than any other individual state-of-the-art PCB defect detection model in F1-score, and the area under curve value of DDML is also higher than that of any other individual detection model. Furthermore, compared with DDMV, the DDML with an automatic machine learning method achieves the best performance in PCB defect detection, and the F1-score on the two datasets can reach 99.7% and 95.6% respectively.
    Reference | Related Articles | Metrics | Comments0
    Development of Rehabilitation and Assistive Robots in China: Dilemmas and Solutions
    ZHAO Lingling1*(赵玲玲),GUO Yao2(郭遥)
    J Shanghai Jiaotong Univ Sci    2023, 28 (3): 382-390.   DOI: 10.1007/s12204-023-2596-9
    Abstract154)      PDF(pc) (367KB)(37)       Save
    China is rapidly becoming an aging society, leading to a significant demand for chronic disease management and personalized healthcare. The development of rehabilitation and assistive robotics in China has gathered significant attention not only in research fields but also in industries. Such robots aim to either guide patients in completing therapeutic training or assist people with impaired functions in performing their daily activities. In the past decades, we have witnessed the advancement in rehabilitation and assistive robotics, with diverse mechanical designs, functionalities, and purposes. However, the construction of dedicated regulations and policies is relatively lagged compared with the flourishing development in research fields. Moreover, these kinds of robots are working or collaborating closely with human beings, bringing unprecedented considerations on ethical issues. This paper aims to provide an overview of major dilemmas in the development of rehabilitation and assistive robotics in China and propose several potential solutions.
    Reference | Related Articles | Metrics | Comments0
    Shape Sensing for Single-Port Continuum Surgical Robot Using Few Multicore Fiber Bragg Grating Sensors
    LI Dingjia1,2,3,4(黎定佳),WANG Chongang1,2,3(王重阳),GUO Wei5(郭伟),WANG Zhidong6(王志东),ZHANG Zhongtao5(张忠涛),LIU Hao1,2,3*(刘浩)
    J Shanghai Jiaotong Univ Sci    2023, 28 (3): 312-322.   DOI: 10.1007/s12204-023-2579-x
    Abstract195)      PDF(pc) (2606KB)(37)       Save
    We proposed a method for shape sensing using a few multicore fiber Bragg grating (FBG) sensors in a single-port continuum surgical robot (CSR). The traditional method of utilizing a forward kinematic model to calculate the shape of a single-port CSR is limited by the accuracy of the model. If FBG sensors are used for shape sensing, their accuracy will be affected by their number, especially in long and flexible CSRs. A fusion method based on an extended Kalman filter (EKF) was proposed to solve this problem. Shape reconstruction was performed using the CSR forward kinematic model and FBG sensors, and the two results were fused using an EKF. The CSR reconstruction method adopted the incremental form of the forward kinematic model, while the FBG sensor method adopted the discrete arc-segment assumption method. The fusion method can eliminate the inaccuracy of the kinematic model and obtain more accurate shape reconstruction results using only a small number of FBG sensors. We validated our algorithm through experiments on multiple bending shapes under different load conditions. The results show that our method significantly outperformed the traditional methods in terms of robustness and effectiveness.
    Reference | Related Articles | Metrics | Comments0
    Visual Positioning of Nasal Swab Robot Based on Hierarchical Decision
    LI Guozhia a(李国志),ZOU Shuizhong b*(邹水中),DING Shuacue a(丁数学)
    J Shanghai Jiaotong Univ Sci    2023, 28 (3): 323-329.   DOI: 10.1007/s12204-023-2581-3
    Abstract180)      PDF(pc) (743KB)(36)       Save
    This study focuses on a robot vision localization method for coping with the operational task of automatic nasal swab sampling. The application is important in the detection and epidemic prevention of Corona Virus Disease 2019 (COVID-19) to alleviate the large-scale negative impact of individuals suffering from pneumonia owing to COVID-19. In this method, the idea of a hierarchical decision network is used to consider the strong infectious characteristics of the COVID-19, which is followed by processing the robot behavior constraint condition. The visual navigation and positioning method using a single-arm robot for sampling is also planned, which considers the operation characteristics of medical staff. In the decision network, the risk factor for potential contact infection caused by swab sampling operations is established to avoid the spread among personnel. A robot visual servo control with artificial intelligence characteristics is developed to achieve a stable and safe nasal swab sampling operation. Experiments demonstrate that the proposed method can achieve good vision positioning for the robots and provide technical support for managing new major public health situations.
    Reference | Related Articles | Metrics | Comments0
    Foreground Segmentation Network with Enhanced Attention
    JIANG Rui1*(姜﹐锐),ZHU Ruiriang1(朱瑞祥),CAI Xiaocui1(蔡萧萃),SU Hu2(苏虎)
    J Shanghai Jiaotong Univ Sci    2023, 28 (3): 360-369.   DOI: 10.1007/s12204-023-2603-1
    Abstract247)      PDF(pc) (734KB)(36)       Save
    Moving object segmentation (MOS) is one of the essential functions of the vision system of all robots,including medical robots. Deep learning-based MOS methods, especially deep end-to-end MOS methods, are actively investigated in this field. Foreground segmentation networks (FgSegNets) are representative deep end-to-end MOS methods proposed recently. This study explores a new mechanism to improve the spatial feature learning capability of FgSegNets with relatively few brought parameters. Specifically, we propose an enhanced attention (EA) module, a parallel connection of an attention module and a lightweight enhancement module, with sequential attention and residual attention as special cases. We also propose integrating EA with FgSegNet v2 by taking the lightweight convolutional block attention module as the attention module and plugging EA module after the two Maxpooling layers of the encoder. The derived new model is named FgSegNet v2 EA. The ablation study verifies the effectiveness of the proposed EA module and integration strategy. The results on the CDnet2014 dataset, which depicts human activities and vehicles captured in different scenes, show that FgSegNet v2 EA outperforms FgSegNet v2 by 0.08% and 14.5% under the settings of scene dependent evaluation and scene independent evaluation, respectively, which indicates the positive effect of EA on improving spatial feature learning capability of FgSegNet v2.
    Reference | Related Articles | Metrics | Comments0
    Time-Resolved Imaging in Short-Wave Infrared Region
    XU Yang (徐杨), LI Wanwan∗ (李万万)
    J Shanghai Jiaotong Univ Sci    2024, 29 (1): 29-36.   DOI: 10.1007/s12204-022-2547-x
    Abstract66)      PDF(pc) (810KB)(35)       Save
    Compared with the conventional first near-infrared (NIR-I, 700—900 nm) window, the short-wave infrared region (SWIR, 900—1 700 nm) possesses the merits of the increasing tissue penetration depths and the suppression of scattering background, leading to great potential for in vivo imaging. Based on the limitations of the common spectral domain, and the superiority of the time-dimension, time-resolved imaging eliminates the auto-fluorescence in the biological tissue, thus supporting higher signal-to-noise ratio and sensitivities. The imaging technique is not affected by the difference in tissue composition or thickness and has the practical value of quantitative in vivo detection. Almost all the relevant time-resolved imaging was carried out around lanthanide-doped upconversion nanomaterials, owing to the advantages of ultralong luminescence lifetime, excellent photostability, controllable morphology, easy surface modification and various strategies of regulating lifetime. Therefore, this review presents the research progress of SWIR time-resolved imaging technology based on nanomaterials doped with lanthanide ions as luminescence centers in recent years.
    Reference | Related Articles | Metrics | Comments0
    High Curvature Stripe Profile Extraction Algorithm of Line Structured Light Measuring System
    CHEN Qiheng1(陈启恒),PENG Na2(彭娜),Lü Na1(吕娜),TAO Wei(陶卫),ZHAO Huil*(赵辉)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 577-586.   DOI: 10.1007/s12204-022-2414-9
    Abstract67)      PDF(pc) (1586KB)(32)       Save
    Non-uniformity of light sources is one of the inevitable error factors causing poor shape recovery accuracy of photometric stereo methods under close-range lighting with quasi point lights. Semi-calibrated photometric stereo methods are required to avoid repeated, tedious and impractical photometric calibration. In this paper, two simple, concise but effective mesh-based semi-calibrated photometric stereo methods are proposed. The proposed methods extend the traditional mesh-based photometric stereo methods and further allow joint and accurate estimation of normals and non-uniform light intensities by alternatively updating normals, depth maps and intensities. Extensive experiments are conducted to validate the effectiveness and robustness of the proposed algorithms. Even under extremely severe non-uniform lighting, the proposed methods can still suppress the error and improve the shape recovery accuracy by up to 65.6% in real-world experiments.
    Reference | Related Articles | Metrics | Comments0
    CFD-Aided Investigation of Combined Flow Conditioners for Gas Ultrasonic Flow Meter
    YUAN Yaqi1(袁亚琦),LI Shiyangl*(李世阳),ZHENG Jia1(郑佳),LI Mingrui2(李明睿)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 611-620.   DOI: 10.1007/s12204-021-2378-1
    Abstract33)      PDF(pc) (3397KB)(31)       Save
    Stable and fully developed gas flow field is crucial for realizing accurate measurement of gas ultrasonic flow meter. To reduce the flow field distortion, a flow conditioner is usually used. However, the traditional monotype flow conditioner can only improve the flow field distribution partly. The measurement accuracy of the transit time ultrasonic flow meter is still affected because of its serious flow field distortion in the complex pipeline conditions. In this paper, to further improve the flow field distribution, a combined conditioner is investigated. The combined flow conditioner is composed of fan-shaped section, turbulent mixing cavity, and honeycomb-shaped section. The effects of fan blade angle and cavity length on the flow field of the DN50 flow meter are studied using computational fluid dynamics (CFD) simulation. Simulation results indicate that compared with the monotype conditioner, the combined conditioner has better performance on effectively reducing the swirl and turbulence and providing more stable and repetitive velocity profiles. Experiments also validate the effectiveness of the combined conditioner. The flow meter with the combined conditioner has better repeatability of less than 0.2%, which is better than those of the monotype conditioners under the same conditions. This work is very useful for accurate measurement of gas ultrasonic flow meter, especially for the complex pipeline conditions.
    Reference | Related Articles | Metrics | Comments0
    Cross-Modal Entity Resolution for Image and Text Integrating Global and Fine-Grained Joint Attention Mechanism
    ZENG Zhirian(曾志贤),CAO Jianjun*(曹建军),WENG Nianfeng(翁年凤),YUAN Zhen(袁震),YU Xu(余旭)
    J Shanghai Jiaotong Univ Sci    2023, 28 (6): 728-737.   DOI: 10.1007/s12204-022-2465-y
    Abstract57)      PDF(pc) (1951KB)(31)       Save
    In order to solve the problem that the existing cross-modal entity resolution methods easily ignore the high-level semantic informational correlations between cross-modal data, we propose a novel cross-modal entity resolution for image and text integrating global and fine-grained joint attention mechanism method. First, we map the cross-modal data to a common embedding space utilizing a feature extraction network. Then, we integrate global joint attention mechanism and fine-grained joint attention mechanism, making the model have the ability to learn the global semantic characteristics and the local fine-grained semantic characteristics of the cross-modal data, which is used to fully exploit the cross-modal semantic correlation and boost the performance of cross-modal entity resolution. Moreover, experiments on Flickr-30K and MS-COCO datasets show that the overall performance of R@sum outperforms by 4.30% and 4.54% compared with 5 state-of-the-art methods, respectively, which can fully demonstrate the superiority of our proposed method.
    Reference | Related Articles | Metrics | Comments0
    Gas Exchange Mechanism of High Frequency Ventilation: A Brief Narrative Review and Prospect
    YUAN Yueyang1* (袁越阳),CHEN Yuqing2 (陈宇清),ZHOU Lil(周理), LIU Wei3 (刘炜), DAI Zheng3(戴 征)
    J Shanghai Jiaotong Univ Sci    2023, 28 (4): 546-.   DOI: 10.1007/s12204-022-2424-7
    Abstract67)      PDF(pc) (466KB)(29)       Save
    The high frequency ventilation (HFV) can well support the breathing of respiratory patient with 20%—40% of normal tidal volume. Now as a therapy of rescue ventilation when conversional ventilation failed, the HFV has been applied in the treatments of severe patients with acute respiratory failure (ARF), acute respiratory distress syndrome (ARDS), etc. However, the gas exchange mechanism (GEM) of HFV is still not fully understood by researchers. In this paper, the GEM of HFV is reviewed to track the studies in the last decades and prospect for the next likely studies. And inspired by previous studies, the GEM of HFV is suggested to be continually developed with various hypotheses which will be testified in simulation, experiment and clinic trail. One of the significant measures is to study the GEM of HFV under the cross-disciplinary integration of medicine and engineering. Fully understanding the GEM can theoretically support and expand the applications of HFV, and is helpful in investigating the potential indications and contraindications of HFV.
    Reference | Related Articles | Metrics | Comments0
    Measuring Transverse Relaxation Time of Xenon Atoms Based on Single Beam of Laser in Nuclear Magnetic Resonance Gyroscope
    ZHONG Guochen1(钟国宸),LIU Hual*(刘华), GUo Yang1(郭阳),LI Shaoliang2(李绍良),ZHAO Wanliang2(赵万良),CHENG Yuxiang2(成宇翔)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 569-576.   DOI: 10.1007/s12204-022-2436-3
    Abstract59)      PDF(pc) (1758KB)(27)       Save
    Nuclear magnetic resonance gyroscope (NMRG) has the characteristics of high precision and miniaturization, and is one of the main applications of quantum technology in the field of navigation. The transverse relaxation time (T2) of the xenon nuclear spin in the atomic cell of the NMRG directly affects the angular random walk of the gyro. Accurate and rapid measurement of T2 is conducive to further improvement of gyroscope. At present, for the measurement of T2, the schemes of two orthogonal lasers for pumping and detecting are usually used. By applying two fast-switching orthogonal static magnetic fields and a single beam of circularly polarized laser with corresponding wavelength to pump the atomic cell, the xenon nuclear macroscopic magnetic moment Larmor precession is generated. The cesium atoms parametric magnetometer in cell is formed to detect the free induction decay signal generated by nuclear spin precession of xenon atoms. The measurement of T2 by a single laser simplifies the measurement equipment compared with traditional method with two lasers. The experimental results show that the T2 of xenon atoms is more than 10 s, and the effects of temperature are studied, which lay the foundation for the subsequent improvement of gyro performance.
    Reference | Related Articles | Metrics | Comments0
    Dynamic Analysis and Optimal Parameter Design of Flexible Composite Structures via Absolute Nodal Coordinate Formulation
    YANG Dan(杨丹),YU Haidong*(余海东),LIN Zhangpeng(林张鹏)
    J Shanghai Jiaotong Univ Sci    2023, 28 (5): 621-629.   DOI: 10.1007/s12204-022-2419-4
    Abstract45)      PDF(pc) (2376KB)(26)       Save
    The composite structure with the dielectric elastomer and soft materials is the main form of the actuators in soft robots. However, the theoretical model is hard to obtain due to the nonlinear large deformation of materials. In this paper, a new composite element model is established based on the absolute nodal coordinate formulation. The consistent deformation conditions at the contact interface between two thin plates are deduced. The hyperelastic constitutive model and the dielectric elastomer constitutive model are introduced for the two thin plates. Then the dynamic model is established to study the dynamic behaviors of the composite flexible structure with various parameters. The results show that the nonlinear deformation appears obviously when the flexible composite plate structure is driven by various voltages, and the warping deformation becomes more obvious with the increase of the voltage. The width and thickness of the driven thin plate influence the stability of the whole structure. With the decrease of the width or thickness, the deformation of the structure is more consistent with obvious periodicity, and the control performance is improved. Finally, the structural parameters of the composite structures are optimized to improve the control performance based on the dynamic performance. Additionally, smaller width and thickness parameters are preferred to obtain better performance in the design of flexible actuator of soft robot.
    Reference | Related Articles | Metrics | Comments0
    Retinal Vessel Segmentation via Adversarial Learning and Iterative Refinement
    GU Wen (顾闻), XU Yi∗ (徐奕)
    J Shanghai Jiaotong Univ Sci    2024, 29 (1): 73-80.   DOI: 10.1007/s12204-022-2479-5
    Abstract44)      PDF(pc) (914KB)(26)       Save
    Retinal vessel segmentation is a challenging medical task owing to small size of dataset, micro blood vessels and low image contrast. To address these issues, we introduce a novel convolutional neural network in this paper, which takes the advantage of both adversarial learning and recurrent neural network. An iterative design of network with recurrent unit is performed to refine the segmentation results from input retinal image gradually. Recurrent unit preserves high-level semantic information for feature reuse, so as to output a sufficiently refined segmentation map instead of a coarse mask. Moreover, an adversarial loss is imposing the integrity and connectivity constraints on the segmented vessel regions, thus greatly reducing topology errors of segmentation. The experimental results on the DRIVE dataset show that our method achieves area under curve and sensitivity of 98.17% and 80.64%, respectively. Our method achieves superior performance in retinal vessel segmentation compared with other existing state-of-the-art methods.
    Reference | Related Articles | Metrics | Comments0