J Shanghai Jiaotong Univ Sci

Automatic Removal of Multiple Artifacts for Single-Channel Electroencephalography

ZHANG Chenbei (张晨贝), SABOR Nabil, LUO Junwen (罗竣文), PU Yu (蒲宇), WANG Guoxing (王国兴), LIAN Yong∗ (连勇)

2022, 27 (4): 437-451.
doi: 10.1007/s12204-021-2374-5

Abstract ( 295 )

PDF (2934KB) ( 127 )

Removing different types of artifacts from the electroencephalography (EEG) recordings is a critical step in performing EEG signal analysis and diagnosis. Most of the existing algorithms aim for removing single type of artifacts, leading to a complex system if an EEG recording contains different types of artifacts. With the advancement in wearable technologies, it is necessary to develop an energy-efficient algorithm to deal with different types of artifacts for single-channel wearable EEG devices. In this paper, an automatic EEG artifact removal algorithm is proposed that effectively reduces three types of artifacts, i.e., ocular artifact (OA), transmission- line/harmonic-wave artifact (TA/HA), and muscle artifact (MA), from a single-channel EEG recording. The effectiveness of the proposed algorithm is verified on both simulated noisy EEG signals and real EEG from CHB- MIT dataset. The experimental results show that the proposed algorithm effectively suppresses OA, MA and TA/HA from a single-channel EEG recording as well as physical movement artifact.

References | Related Articles | Metrics

Deformable Registration Algorithm via Non-subsampled Contourlet Transform and Saliency Map

CHANG Qing∗ (常青), YANG Wenyou (杨文友), CHEN Lanlan (陈兰岚)

2022, 27 (4): 452-462.
doi: 10.1007/s12204-022-2428-3

Abstract ( 177 )

PDF (4867KB) ( 39 )

Medical image registration is widely used in image-guided therapy and image-guided surgery to esti- mate spatial correspondence between planning and treatment images. However, most methods based on intensity have the problems of matching ambiguity and ignoring the influence of weak correspondence areas on the overall registration. In this study, we propose a novel general-purpose registration algorithm based on free-form defor- mation by non-subsampled contourlet transform and saliency map, which can reduce the matching ambiguities and maintain the topological structure of weak correspondence areas. An optimization method based on Markov random fields is used to optimize the registration process. Experiments on four public datasets from brain, car- diac, and lung have demonstrated the general applicability and the accuracy of our algorithm compared with two state-of-the-art methods.

References | Related Articles | Metrics

Gram Matrix-Based Convolutional Neural Network for Biometric Identification Using Photoplethysmography Signal

WU Caiyu, (吴彩钰), SABOR Nabil, ZHOU Shihong, (周世鸿), WANG Min, (王敏), YING Liang (应亮), WANG Guoxing∗ (王国兴)

2022, 27 (4): 463-472.
doi: 10.1007/s12204-022-2426-5

Abstract ( 240 )

PDF (1049KB) ( 50 )

As a kind of physical signals that could be easily acquired in daily life, photoplethysmography (PPG) signal becomes a promising solution to biometric identification for daily access management system (AMS). State- of-the-art PPG-based identification systems are susceptible to the form of motions and physical conditions of the subjects. In this work, to exploit the advantage of deep learning, we developed an improved deep convolutional neural network (CNN) architecture by using the Gram matrix (GM) technique to convert time-serial PPG signals to two-dimensional images with a temporal dependency to improve accuracy under different forms of motions. To ensure a fair evaluation, we have adopted cross-validation method and “training and testing” dataset splitting method on the TROIKA dataset collected in ambulatory conditions. As a result, the proposed GM-CNN method achieved accuracy improvement from 69.5% to 92.4%, which is the best result in terms of multi-class classification compared with state-of-the-art models. Based on average five-fold cross-validation, we achieved an accuracy of 99.2%, improved the accuracy by 3.3% compared with the best existing method for the binary-class.

References | Related Articles | Metrics

Breast Pathological Image Classification Based on VGG16 Feature Concatenation

LIU Min (刘敏), YI Ming (易鸣), WU Minghu∗ (武明虎), WANG Juan (王娟), HE Yu (何宇)

2022, 27 (4): 473-484.
doi: 10.1007/s12204-021-2398-x

Abstract ( 245 )

PDF (5914KB) ( 41 )

Breast cancer is one of the malignancies that endanger women’s health all over the world. Considering that there is some noise and edge blurring in breast pathological images, it is easier to extract shallow features of noise and redundant information when VGG16 network is used, which is affected by its relative shallow depth and small convolution kernel. To improve the pathological diagnosis of breast cancers, we propose a classification method for benign and malignant tumors in the breast pathological images which is based on feature concatenation of VGG16 network. First, in order to improve the problems of small dataset size and unbalanced data samples, the original BreakHis dataset is processed by data augmentation technologies, such as geometric transformation and color enhancement. Then, to reduce noise and edge blurring in breast pathological images, we perform bilateral filtering and denoising on the original dataset and sharpen the edge features by Sobel operator, which makes the extraction of shallow features by VGG16 model more accurate. Based on transfer learning, the network model trained with the expanded dataset is called VGG16-1, and another model trained with the image denoising and sharpening and mixed with the original dataset is called VGG16-2. The features extracted by VGG16-1 and VGG16-2 are concatenated, and then classified by support vector machine. The final experimental results show that the average accuracy is 98.44%, 98.89%, 98.30% and 97.47%, respectively, when the proposed method is tested with the breast pathological images of 40×, 100×, 200× and 400× on BreakHis dataset.

References | Related Articles | Metrics

USSL Net: Focusing on Structural Similarity with Light U-Structure for Stroke Lesion Segmentation

JIANG Zhiguo (蒋志国), CHANG Qing∗ (常青)

2022, 27 (4): 485-497.
doi: 10.1007/s12204-022-2412-y

Abstract ( 168 )

PDF (1121KB) ( 49 )

Automatic segmentation of ischemic stroke lesions from computed tomography (CT) images is of great significance for identifying and curing this life-threatening condition. However, in addition to the problem of low image contrast, it is also challenged by the complex changes in the appearance of the stroke area and the difficulty in obtaining image data. Considering that it is difficult to obtain stroke data and labels, a data enhancement algorithm for one-shot medical image segmentation based on data augmentation using learned transformation was proposed to increase the number of data sets for more accurate segmentation. A deep convolutional neural network based algorithm for stroke lesion segmentation, called structural similarity with light U-structure (USSL) Net, was proposed. We embedded a convolution module that combines switchable normalization, multi-scale convolution and dilated convolution in the network for better segmentation performance. Besides, considering the strong structural similarity between multi-modal stroke CT images, the USSL Net uses the correlation maximized structural similarity loss (SSL) function as the loss function to learn the varying shapes of the lesions. The experimental results show that our framework has achieved results in the following aspects. First, the data obtained by adding our data enhancement algorithm is better than the data directly segmented from the multi- modal image. Second, the performance of our network model is better than that of other models for stroke segmentation tasks. Third, the way SSL functioned as a loss function is more helpful to the improvement of segmentation accuracy than the cross-entropy loss function.

References | Related Articles | Metrics

Novel Indicators for Adverse Glycemic Events Detection Analysis Based on Continuous Glucose Monitoring Neural Network Predictive Models

LU Guannan1 (卢冠男), WANG Mengling1∗ (王梦灵), FOX Tamara2, JIANG Peng3 (蒋鹏), JIANG Fusong3 (蒋伏松)

2022, 27 (4): 498-504.
doi: 10.1007/s12204-022-2439-0

Abstract ( 147 )

PDF (253KB) ( 30 )

This paper proposes five indicators to evaluate the effectiveness and viability for adverse glycemic events detection based on predicted blood glucose (BG) values. False negative rate (FNR) and false positive rate (FPR) are defined to evaluate whether it can detect adverse glycemic events (AGEs) based on the predicted value. The temporal overlap (TO) and time difference (TD) are proposed to evaluate whether the predicted model can capture the accurate time duration of AGEs. The sum of squared percent (SSP) measures comprehensive similarity between prediction values and true values. We examined 328 patients with type 2 diabetes, containing real continuous glucose monitoring data with 5-minute time intervals. Autoregressive integrated moving average model has lower FNR and FPR. The gated recurrent unit has better temporal behavior where the mean TO with standard deviation is calculated as 0.84±0.18, and the mean TD with standard deviation is (4.39±4.01) min. Neural models have better effects on SSP (for hypoglycemia, long-short tern memory possesses 0.149 and 0.246). These five indicators are able to evaluate whether we can detect abnormal BG levels and reveal the temporal behavior of AGEs effectively. The proposed neural predictive models have more promising application in AGE detection.

References | Related Articles | Metrics

Electrophysiological Characterization of Substantia Nigra Pars Reticulata in Anesthetized Rats

LIU Xinrui 1 (刘信锐), ZHANG Qianwen1 (张骞文), WANG Ying2 (王莹), CHEN Fujun1,3∗ (陈福俊)

2022, 27 (4): 505-511.
doi: 10.1007/s12204-022-2420-y

Abstract ( 158 )

PDF (1435KB) ( 37 )

The substantia nigra pars reticulate (SNr), which plays a pivotal role in motor control, is the key structure in integrating information for cortex, basal ganglia and thalamus. Abnormal gait and posture deficits can be reversed by SNr deep brain stimulation (DBS) in certain Parkinson’s disease cases. However, functional characterization of SNr, which is the key for the optimization of DBS effect, remains elusive. In current study, we recorded extracellular single unit in SNr of urethane anesthetized rats. We have found that urethane can induce slow delta and theta oscillations in SNr local field potential. The high gamma oscillation observed is positively correlated with the occurrence of action potential. The putative GABAergic neurons have a mean firing rate of (20.82 ± 2.04) Hz, of which 65.2% display a regular firing pattern and 34.8% show irregular firing. Our results demonstrated the heterogeneous property of SNr and provided possible theoretical basis for promoting the next generation of DBS electrode design and optimization of clinical DBS parameters.

References | Related Articles | Metrics

Chronic β-Citronellol Inhalation Rescues Parvalbumin Expression Loss in Prefrontal Cortex of Chronic Restraint Stress Mice

ZHUANG Qianqian (庄倩倩), ZHUANG Siyue (庄似玥), GONG Yanling (龚艳玲), LI Shengtian∗ (李胜天)

2022, 27 (4): 512-520.
doi: 10.1007/s12204-022-2455-0

Abstract ( 178 )

PDF (974KB) ( 20 )

Considerable evidence has revealed that essential oils and their main constituents possess anti- depressant and anxiolytic properties. In the current study, we report the effect of β-citronellol, the main component of rose essential oil, on depressive-like and anxiety-like behaviors in chronic restraint stress (CRS) mice. We found that chronic inhalation of β-citronellol for 14 days could increase locomotor activity in the open field test, de- crease the percentage of immobility duration in the forced swimming test, and increase open arms exploration in elevated plus-maze test in CRS mice. Western blot experiment shows that chronic β-citronellol inhalation res- cues parvalbumin (PV) expression loss in the prefrontal cortex (PFC) of CRS mice. Correlation analysis reveals a strong relationship between the PV expression in PFC and the percentage of sucrose preference of the mice. These findings indicate the relationships between the PV gene expression of PFC and the effects of β-citronellol inhalation.

References | Related Articles | Metrics

Personalized Design Method of Bionic Bone Scaffold with Voronoi Spacial Architecture

WU Li ∗ (武力), HUANG Wei (黄伟), LI Xuetao (李学涛)

2022, 27 (4): 521-527.
doi: 10.1007/s12204-022-2410-0

Abstract ( 170 )

PDF (1761KB) ( 39 )

In order to design and manufacture a bionic bone scaffold for personalized bone tissue repairing, we give the performance requirements and indicators, and propose a method of establishing a three-dimensional bone scaffold solid model based on the Voronoi architecture. The modeling parameters, number of seed points and scaling factor are found, which can control the model performance. The finite element analysis model of bone scaffold with different number of seed points and scaling factor is set up. By adopting the method of theory analysis, experiment and simulation calculation, the performance parameters of the porosity, the specific surface area and the elastic modulus of 15 sets of bionic bone scaffold model are obtained. The relation between the bone scaffold model parameters and the porosity, the porosity and the specific surface area, the porosity and the equivalent elastic modulus are established. The individual design method of Voronoi bone scaffold is proposed. In the condition of given equivalent elastic modulus and the defected shape of bone, the bone scaffold model with spatial Voronoi architecture which accords with the performance requirements can be designed and produced rapidly, which provides reference for the personalized treatment of bone defecting.

References | Related Articles | Metrics

Case Study of a Personalized Scoliosis Brace Based on 3D Printing

LU Dezhi 1,2‡ (鲁德志), LI Wentao1‡ (李文韬), WANG Xiaowen2 (王孝文), SONG Yan2 (宋艳), ZHANG Pingping2 (张萍萍), FENG Haiyang2 (冯海洋), WU Yuncheng1 (吴云成), XU Yuanjing3 (许苑晶), LI Tao4 (李涛), MA Zhenjiang1∗ (马振江), WANG Jinwu1,2∗ (王金武)

2022, 27 (4): 528-534.
doi: 10.1007/s12204-022-2461-2

Abstract ( 234 )

PDF (653KB) ( 45 )

We evaluated the effect of a new type of brace (primary material 3300PA) for treating scoliosis, which was produced based on 3D printing technology combined with a non-contact optical mold-taking and computer- aided design. Through the production of a brace for a 13-year-old patient with adolescent idiopathic scoliosis by a multidisciplinary team, the digital design and 3D printing of a personalized scoliosis brace were introduced. Parameters such as the Cobb angle, angle of trunk inclination, spine-coronal plane balance parameters, sagittal vertical axis (SVA), and Scoliosis Research Society-22 score (SRS-22) were measured to evaluate the treatment effect of the brace. The model-taking process of the non-contact optical scanner was successful, data were valid, and personalized scoliosis brace made by the computer-aided design and 3D printing fitted well with the patient. Before wearing, immediate in-brace, and 6 months after wearing, the Cobb angles were 29?, 9?, a n d 1 4?, respectively. The offsets between the C7 vertebra plumb line (C7PL) and central sacral vertical line (CSVL) were 3.2 cm, 2.2 cm, and 2.1 cm, respectively. SVAs were 3.3 cm, 2.9 cm, and 0.3 cm, respectively. Apex vertebral translocations were 4.3 cm, 0.3 cm, and 0.1 cm, respectively. The SRS-22 was 76 before brace application and 91 at the 6-month follow-up. The spine curve returned to normal, and the correction effect was obvious. The scoliosis brace indicates the integration between digital medicine and 3D printing technology, which has personalization and customization as advantages. The brace has good wearing comfort, invisibility, and orthopedic function, follows the psychological needs of teenagers, improves patients’ compliance, and improves the correction of the deformity.

References | Related Articles | Metrics

Design of Real-Time Temperature Monitoring and Control System for Multimodal Ablation

ZOU Ke (邹柯), ZOU Jincheng (邹金成), WANG Yifei (王逸飞), ZHANG Aili ∗ (张爱丽)

2022, 27 (4): 535-542.
doi: 10.1007/s12204-022-2432-7

Abstract ( 190 )

PDF (893KB) ( 31 )

By combining cryogenic cryotherapy and high-temperature radiofrequency therapy, multimodal abla- tion generates a rapidly changing temperature field in tissue by heating after pre-freezing. This method completely breaks tumour cells and releases a large amount of active antigen. Compared with the traditional single modal- ity, the thermal physical ablation method has been shown to have a greater therapeutic effect, but it presents challenges in terms of precise monitoring and rapid control of the temperature during the treatment process. To solve this problem, we propose a temperature control system design utilizing aspects of probe sensing, real- time software and hardware signal interfaces, and dynamic compensation control strategies to accurately monitor the temperature changes during multimodal ablation treatment. The results show that the design system has millisecond-level high-speed control capability, an accuracy of 0.5 ?C, and the dynamic response time is less than 0.1 s. Furthermore, the temperature fluctuation in in vivo e x p e r i m e n t s i s l e s s t h a n 0 . 5 ?C.

References | Related Articles | Metrics

Design of a Low Respiratory Resistance Mask for COVID-19

YANG Pufan (杨朴凡), HUANG Hongxin (黄洪鑫) WEI Siji (卫思霁), YAO Yuan (姚远), ZHANG Zhinan∗ (张执南)

2022, 27 (4): 543-551.
doi: 10.1007/s12204-022-2434-5

Abstract ( 229 )

PDF (1382KB) ( 43 )

Observational evidence suggests that mask-wearing mitigates transmission of COVID-19; at the same time high respiratory resistance leads to an unwillingness to wear masks. This paper proposed a respiratory drive structure to reduce the air resistance of a mask. This structure provides different shapes during expiration and inspiration while focusing on filtering dust, bacteria, or viruses. Meanwhile, the assembled system on the mask can be disassembled and replaced. Then porous media simulation is used to verify the model effect. Experimental results of a new mask show that the ventilation resistance is reduced by 20%, and the bacterial filtration efficiency meets the requirements of YY 0469—2011.

References | Related Articles | Metrics

Biomechanical Analysis of a Radial Expansion Mechanism of Intestinal Robot Coupling with Hyperelastic Intestinal Wall

LIU Dasheng∗ (刘大生), YAN Guozheng (颜国正)

2022, 27 (4): 552-560.
doi: 10.1007/s12204-022-2427-4

Abstract ( 183 )

PDF (2471KB) ( 36 )

This paper proposes a new type of radial expansion mechanism by adopting the scissor type telescopic design for intestinal robot to meet the requirements of the intestinal robot’s movement and residence in the intestinal tract. The robot’s maximum expansion radius is up to 25 mm, which can well adapt to the intestinal tract with different diameters. At first, the mathematical model of the scissors-type telescopic mechanism (STM) is established to further study its dynamics characteristics by theoretical analysis and simulation. Then, in order to study the coupling effect between the STM and intestinal wall, the strain-energy function of Fung-type is used to establish the constitutive model of intestinal wall. Moreover, aimed at solving the non-convergence problem caused by the selection of material parameters in general Fung-type model, the restrictions for selecting material parameters were given by using positive definite matrix theory. Furthermore, the motion coupling characteristics between the mechanism and intestinal wall were analyzed by using the finite element method. The result shows that if the expansion radius of the STM exceeds a certain value, the intestinal wall may reach its deformation limit, which means that the maximum rotating angle of the three-claw butterfly disc of STM can be decided based on the maximum deformation stress of the intestinal wall. Therefore, it provides a design basis for formulating a reasonable expansion radius in mechanism control to avoid damage to the intestinal wall.

References | Related Articles | Metrics

Piezoelectric-Based Smart Bone Plate for Fracture Healing Progress Monitoring

GAO Zihang (高梓航), WANG Xin (王鑫) ZHAO Yifan (赵一帆), JIN Zhehui (金哲慧), WANG Gang (王刚), GAO Shuo∗ (高硕)

2022, 27 (4): 561-569.
doi: 10.1007/s12204-022-2417-6

Abstract ( 168 )

PDF (1304KB) ( 44 )

Fracture healing progress monitoring techniques attract global research attention due to the importance of selecting the timing of removing the fixation device. To this end, in this research, we present a piezoelectric- based smart internal fixation device, in which a piezoelectric sandwich structure is laminated to the surface of a bone plate. In the content, we explain the reasons for utilizing piezoelectric films, elaborate the mechanism of fracture monitoring, and introduce the mechanical parameters of the sensor. The simulation and experimental results show that the electrical output of the device is associated with the elastic modulus of the filler between the tested broken bones when the working load is maintained, indicating that the bone recovery progress could be successfully detected by the developed technique.

References | Related Articles | Metrics

Dynamic Modeling and Performance Evaluation of a Novel Humanoid Ankle Joint

LI Yanbiao∗ (李研彪), CHEN Ke (陈科), SUN Peng (孙鹏), WANG Zesheng (王泽胜)

2022, 27 (4): 570-578.
doi: 10.1007/s12204-022-2422-9

Abstract ( 219 )

PDF (1322KB) ( 31 )

Aimed at the problems of design difficulty and weak kinematic performance caused by spherical joint, a novel PRC+PRCR+RR humanoid ankle joint based on the partially decoupled spherical parallel mechanism is proposed. According to screw theory, the degree of freedom and decoupling characteristics of this mechanism are analyzed. Based on Klein formula and virtual work principle, the kinematic expressions of each link and dynamic model are established. The correctness of the dynamic model is verified by combining the virtual prototype software and the ankle pose function obtained by gait planning and Fourier fitting. The workspace of this mechanism is mapped into a two-dimensional polar coordinate system with the azimuth and elevation angles of the spherical coordinate system as parameters. The motion/force transmission index and constraint index of this mechanism are evaluated and expressed in the workspace, showing this mechanism with excellent kinematic characteristics.

References | Related Articles | Metrics

Analysis of Driving Psychological Load in V-Shaped Subsea Tunnels Considering Driver Skin Electrical Signals

PAN Fuquan1 (潘福全), PAN Haitao1 (泮海涛), ZHANG Lixia1∗ (张丽霞), LU Linjun2 (陆林军), YANG Xiaoxia1 (杨晓霞), WANG Lin1 (王琳)

2022, 27 (4): 579-587.
doi: 10.1007/s12204-022-2437-2

Abstract ( 214 )

PDF (1272KB) ( 26 )

To examine the variation law of the driving psychological load in subsea tunnels under different illumi- nation and longitudinal slope conditions, 22 drivers were recruited to participate in a real vehicle test in off peak hours under similar traffic conditions, and the skin electric signals of the drivers in the free flow state were col- lected. Considering the skin conductance level (SCL) as the load characteristic index, the influence of the different illuminance and slope conditions on the drivers’ skin electrical signals was analyzed, and a measurement model of the relationship between the uphill and downhill slopes, illuminance and driver’s SCL value was constructed. The results indicate that the illuminance change rate and driver’s SCL are positively correlated. A larger illuminance change rate leads to an increase in the SCL and psychological workload of the driver. The influence of the uphill and downhill slopes on the driver’s SCL value in different areas of the subsea tunnel is considerably different. With the increase in the degree of the uphill and downhill slopes, the driver’s SCL value increases, and the maximum SCL appears in a slope range of 3.5%—4%. Moreover, the SCL of the drivers in the downhill section is higher than that in the uphill section, corresponding to a larger driving psychological load.

References | Related Articles | Metrics

Retraction Note to: Estimation of Load-Induced Damage and Repair Cost in Post-Tensioned Concrete Rocking Walls

JAF ARI Abouzar, DUGNANI Roberto∗

2022, 27 (4): 588-588.
doi: 10.1007/s12204-022-2429-2

Abstract ( 167 )

PDF (48KB) ( 24 )

The Editor-in-Chief has retracted this article after an investigation by the University of Michigan - Shanghai Jiao Tong University Joint Institute concluded that it overlaps significantly with Refs. [1-3] and is therefore redundant. Roberto Dugnani agrees with this retraction but not with the wording of the retraction notice. Abouzar Jafari has not responded to correspondence from the Journal about this retraction.

References | Related Articles | Metrics

Table of Content