Home
About Journal
Editorial Board
Instruction
Subscribe
Download
Messages Board
Contact Us
中文
Quick Search
Office Online
Current Issue
Accepted
Current Issue
Archive
Adv Search
Read Articles
Download Articles
Email Alert
Download
More>>
Links
More>>
2024 Vol. 43, No. 6
Published: 2024-12-20
Reviews
Communications
Regular Papers
Indexes
Regular Papers
641
Key Point-Guided Temporal Network Used for Segmentation of Echocardiography
Xiang Zhuo, Chen Weiling, Tian Xiaoyu, Zhao Cheng, Wang Tianfu, Lei Baiying
DOI: 10.3969/j.issn.0258-8021.2024.06.001
Echocardiographic segmentation is an important step in the screening of congenital heart disease. However, the quality of echocardiogram image usually is relatively low, and some key heart structures in the echocardiogram video portion of the frame can blur or disappear due to the beating of the heart. For the target frame whose structure disappears, it is usually necessary to deduce the position of the key structure in the target frame by relying on other frames with clear structure in the echocardiography video. Aiming to address these challenges, this study designed a key point guided timing network to complete the segmentation of echocardiography. Specifically, for the target frame to be segmented, other frames in the ultrasonic video were used as secondary frames. First, a bidirectional temporal network (BTN) was designed to extract the structure information from the auxiliary frame, and in this process, the key points guided the network to extract the key structure information. Then, a transformer temporal attention (TTA) module was proposed to adjust the feature weights of each auxiliary frame and focus on the auxiliary frame with clear structure. In addition, this study proposed an image mapping (IM) module, which mapped the structure information of the auxiliary frame directly to the target frame and completed the supplement of missing structure information in the target frame. In this study, experiments were conducted on the parasternal short axis section data of 98 cases, and the average Dice reached 0.8269. Experimental results showed that the proposed method could be effectively applied to echocardiogram segmentation.
2024 Vol. 43 (6): 641-651 [
Abstract
] (
200
)
HTML
(1 KB)
PDF
(7604 KB) (
140
)
652
Cluster-Based Multiple Instance Learning for Whole Slide Image Classification
Zhong Haiqin, Zhao Cheng, Lei Baiying, Wang Tianfu
DOI: 10.3969/j.issn.0258-8021.2024.06.002
Pathological images are the gold standard for cancer examination. Fast and accurate classification of pathological images, especially whole slide images (WSI), helps medical doctors provide personalized treatment and prognosis assessment for patients. In recent years, multiple instance learning (MIL) has played an increasingly important role in WSI classification. However, due to the limited number of WSIs and the low proportion of positive areas, the existing MIL method based on attention mechanism may lead to overfitting, thus affecting the classification performance. To solve this problem, we proposed a new clustering-based MIL classification method. Specifically, this method divided each bag into multiple pseudo bags to increase the number of packages and let the network pay attention to more positive instances. Then, to solve the problem that a pseudo-bag is easy to be full of negative instances in the pseudo-bag division process, resulting in noise, this paper proposed a new pseudo-bag division method based on clustering. Finally, to obtain more accurate classification results, we conducted secondary learning on the learned pseudo-bag-level features to get the final bag-level features and achieve the final WSI classification. We conducted experiments on the Camelyon16 and TCGA-Lung datasets, which have 399 and 1 038 WSIs, respectively, with classification accuracies of 90.69% and 86.54%, and F1-scores of 90.20% and 86.52%. The experimental results showed that the proposed method could be appled to WSI classification effectively.
2024 Vol. 43 (6): 652-661 [
Abstract
] (
180
)
HTML
(1 KB)
PDF
(4459 KB) (
326
)
662
MSCRHO-Net: A Deep Learning Model for Effective Removal of Hair Occlusion from DermoscopicImages
Du Hongxuan, Liu Qingyi, Ren Yande, Wang Yan, Zhang Yanan, Bai Peirui
DOI: 10.3969/j.issn.0258-8021.2024.06.003
It is of great significance to carry out early detection and diagnosis of skin cancer based on automatic analysis of dermoscopic images. However, hair occlusion poses a challenge to image feature extraction and skin lesion diagnosis. In this paper, a multi-scale cascade deep learning model (MSCRHO-Net) by integrating the Laplace pyramid was proposed. First, Laplace pyramid was employed to extract the key features of different scales in image space. A cascade block was designed for each scale channel to predict the hairless image by a coarse to fine scheme. High precision hair extraction and boundaries details retention were achieved through this operation. Then, a combined loss function including perceptual loss and SSIM loss was constructed, which was helpful to enhance details recovery and obtain more clear hair removal images. The performance of MSCRHO-Net was validated on synthetic dataset and real dataset ISIC2017(4750 training images, 400 synthetic test images, and 223 real test images). The experimental results demonstrated that MSCRHO-Net was able to remove hair effectively without learning of huge parameters.The mean values of SSIM and PSNR reached 0.958 4 and 35.49 respectively, which significantly improved the performance (
P
<0.05) compared with other traditional hair removal methods. MSCRHO-Net shows high adaptability and robustness to complex hair structure, and can deal with complicated scenarios such as damaged lesion texture and blurred image.
2024 Vol. 43 (6): 662-672 [
Abstract
] (
128
)
HTML
(1 KB)
PDF
(11753 KB) (
161
)
673
Research on Lateral Ventricle Surgery Training System Based on Augmented Reality Technology
Zhu Zhaoju, Gao Chuhang, Shi Jiafeng, Chen Liujing, He Bingwei
DOI: 10.3969/j.issn.0258-8021.2024.06.004
To address the problems of lacking image guidance, prone to forgetting, and poor training results during simulated lateral ventricle puncture experienced by medical students, this study proposed an augmented reality (AR) glasses-based lateral ventricle puncture assistance system. Based on the CT and MRI data of a patient with mild lateral ventricle enlargement, the medical modeling software Mimics was used to threshold segment the gray characteristics of various tissue structures in the skull and brain, and a 3D virtual model was constructed. Next, AR construction was performed using Vuforia SDK and Unity 3D, and a method of combining the main recognition image with the opening image was proposed to obtain an average registration error of the system of 1.333 mm. At last, the system was imported into AR glasses and debugged. Twenty clinical medical students were invited to participate and were divided into two groups: one wearing AR glasses and the other not. They performed simulated lateral ventricle frontal horn puncture operations and scored the simulated surgery using a questionnaire. The test results showed that compared with the group not wearing AR glasses, the group wearing them used an average of 4.71 minutes less time during the operation, and the success rate of puncture increased by 30%. Furthermore, the group wearing AR glasses gave significantly better feedback on the questionnaire evaluation than the control group. In conclusion, the AR glasses-based lateral ventricle puncture assistance system built in this study can significantly improve the training effect of the surgery and provide a new idea for improving the quality of lateral ventricle puncture surgery training.
2024 Vol. 43 (6): 673-681 [
Abstract
] (
113
)
HTML
(1 KB)
PDF
(12448 KB) (
79
)
682
An Auxiliary Diagnosis Method for Hierarchical Classification of FUO Based on Multi-Pathand Feature Selection
Du Jianchao, Wang Yanning, Shi Lei, Chen Tianyan, Liang Jingchen, Wang Xin, Lian Jianqi, Zhou Yun
DOI: 10.3969/j.issn.0258-8021.2024.06.005
Many causes of fever of unknown origin (FUO) and high characteristic dimensions lead to difficulty in accurate diagnosis. This paper proposed an auxiliary diagnostic method based on hierarchical classification with multi-path and feature selection. Firstly, according to the structure of FUO causes, this method designed a top-down hierarchical classification model to select a controllable number of candidate categories in each middle layer, constructing a multi-path prediction mode, and finally selecting the optimal classification among multiple paths; secondly, an
L
1,2
paradigm regularization constraint was utilized to eliminate redundant features and preserve the optimal subset of features to reduce interference and improve prediction accuracy. In addition, this paper collected data from the First Affiliated Hospital of Xi'an Jiaotong University regarding patients visiting for FUO from 2011 to 2020 to construct a comprehensive dataset. This dataset included 564 samples and 327 dimensional features, categorized into five coarse-grained categories: bacterial infections, viral infections, other infectious diseases, autoimmune diseases, and other non-infectious diseases, and into 16 subordinate fine-grained categories. The sixteen-classification verification results on the dataset showed that when the proposed method selected 25% of the features with 3 candidate classes in the middle layer, the accuracy,
F
H
and
F
LCA
reached 76.08%, 86.72 % and 85.39 %, respectively, which were 9.42%, 4.69%, and 3.36% higher than the traditional single-path and non-feature selection methods, respectively. The proposed method significantly improved evaluation performance compared to the flat classification algorithms and other existing hierarchical classification algorithms, providing a more effective auxiliary diagnostic method for FUO.
2024 Vol. 43 (6): 682-692 [
Abstract
] (
97
)
HTML
(1 KB)
PDF
(4023 KB) (
157
)
693
Cross-Subject Epileptic Seizure Detection Method Based on Discriminative Manifold Regularizationand Domain Distribution Adaptation
Zhang Yanli, Qiu Wenlong, Zhou Weidong
DOI: 10.3969/j.issn.0258-8021.2024.06.006
Automatic detection of seizure is of great significance for epilepsy diagnosis, monitoring, and intervention treatment. Aiming to address the problems that ictal electroencephalogram (EEG) is limited and data distributions among different patients are significantly different, a cross-subject seizure detection method was proposed based on discriminative manifold regularization and domain distribution adaptation in this work. First, the EEG of patients to be tested andthelabeled EEG of other patients were used as target and source domain data, respectively, and the EEG features such as the mean, variance and sample entropy of the wavelet packet decomposition coefficients were extracted. Then, a manifold regularization containing category information of source samples and an intra-class distance minimization constraint were introduced into domain distribution adaptation. Meanwhile, the relative deviation between conditional distribution distance and marginal distribution distance was adopted to dynamically adjust the distribution weight. Finally, pattern classification and seizure detection of target data were realized using the random forest classifier trained by source domain samples after space projection. The detection performance of the proposed method was validated using scalp EEG data from 24 patients in the CHB-MIT database and compared with existing domain adaptation algorithms. The average detection sensitivity and accuracy achieved by the proposed method were 94.94% and 95.66%, respectively, which were 15.07% and 9.98% higher than the CORAL algorithm using second-order statistic alignment, and 3.90% and 2.52% higher than the BDA algorithm that only performs balanced distribution adaptation. In conclusion, the combination of discriminative manifold regularization and domain distribution adaptation reduced the distribution differences between EEG signals from different patients and effectively utilized the discriminative information in the manifold structure and labels of the source domain data, providing a new idea for the research of cross-subject epileptic seizure detection.
2024 Vol. 43 (6): 693-701 [
Abstract
] (
97
)
HTML
(1 KB)
PDF
(1483 KB) (
151
)
702
Accurate Prediction of Paracetamol and Ascorbic Acid Mixed Solutions Based onPSO-BSA Heuristic Algorithm
Guo Boyu, Bao Hanyang, Rao Zhikang, Liu Zhe, Xu Ying
DOI: 10.3969/j.issn.0258-8021.2024.06.007
In this paper, an electrochemical detection method for simultaneous detection of paracetamol (PA) and ascorbic acid (AA) concentrations was investigated. Experimental data were collected through a three-electrode electrochemical workstation, and PBs/cMWCNTs/GCE electrochemical sensors were prepared using cyclic voltammetry (CV), which resulted in basic separation of the peak potentials of PA and AA. Differential pulse voltammetry (DPV) and chronoamperometry (CA) were used to conduct electrochemical detection of PA and AA mixed solutions with different concentrations, and typical electrochemical features and image features of the detection curves were extracted. Then, the improved particle swarm optimization-backtracking search algorithm (PSO-BSA) network was used to decouple the experimental data and construct an accurate prediction model for the concentration of PA and AA in the mixed sample. Finally, the simultaneous detection of PA and AA dual-chemical drugs was achieved. The R-Square (
R
2
) of the prediction results reached 0.995, the root mean square error (RMSE) reached 0.137 mM, and the mean absolute error (MAE) reached 0.124 mM. The method proposed in this study has the advantages of easy operation, satisfactory goodness of fit, and high cost-effectiveness, which is conducive in achieving synergistic detection results of PA and AA mixed drugs and achieving accurate analysis of the components of a certain drug sample.
2024 Vol. 43 (6): 702-711 [
Abstract
] (
78
)
HTML
(1 KB)
PDF
(8130 KB) (
35
)
712
Hyperthermia Effect of Fe
3
O
4
Nanoparticles Guided by Aptamer on 4T1 Breast Cancer Cells
Chen Enyuan, Jiang Zhengting, Ding Jiayi, Kan Junnan, Zu Hanyu, Yan Peng
DOI: 10.3969/j.issn.0258-8021.2024.06.008
Magnetic nanomaterials have been widely used in tumor imaging diagnosis and magnetic hyperthermia due to their unique magnetic properties. In this study, Fe
3
O
4
nanoparticles were prepared by chemical coprecipitation method, and calreticulin receptor-targeted iron oxide nanoparticles (Fe
3
O
4
@Apt) were formed by coupling aptamer (Apt23) on the surface of the nanoparticles through an amide reaction. The hyperthermia effects of Fe
3
O
4
@Apt nanoparticles on breast cancer tumor cells (4T1) were investigated. The phase and crystal structure of Fe
3
O
4
nanoparticles were analyzed by X-ray diffraction. The morphology, particle size, zeta potential and saturation magnetization of Fe
3
O
4
nanoparticles before and after modification were characterized by transmission electron microscope, nanoparticle size analyzer and vibrating sample magnetometer, respectively. Functional groups on Fe
3
O
4
nanoparticles, both before and post-modification, were analyzed through Fourier transform infrared spectroscopy. The targeting effect of Fe
3
O
4
nanoparticles on 4T1 cells was determined by Prussian blue staining. Cell viability and proliferation of 4T1 cells and mouse embryonic fibroblasts (MEF) were detected by MTT assay. Under an alternating magnetic field (ACMF), the heating performance and in vitro magnetic hyperthermia effects of Fe
3
O
4
nanoparticles before and after modification were measured. There were at least 3 replicates (
n
≥3) in all quantitative tests. Experimental results showed that the crystallization of Fe
3
O
4
nanoparticles was good, and the morphology and magnetic properties of Fe
3
O
4
nanoparticles did not change significantly before and after modification. The aptamer modification enhanced the surface electronegativity and increased the average particle size. The average particle sizes of Fe
3
O
4
and Fe
3
O
4
@Apt were (9±4) and (18±5) nm, respectively. The zeta potentials were (-20.4±0.6) and (-27±0.4) mV, respectively. The absorption peak of Fe
3
O
4
@Apt nanoparticles at1 037 cm
-1
indicated successful coupling with Apt23. Fe
3
O
4
nanoparticles before and after modification could support the normal growth and proliferation of 4T1 and MEF cells. After 24 h incubation, there was no significant difference of the cells viability between Fe
3
O
4
@Apt nanoparticles and Fe
3
O
4
nanoparticles at different concentrations (
P
>0.05). Under the ACMF, Fe
3
O
4
nanoparticles exhibited effective heating capabilities before and after modification, and the cell proliferation rate in the targeted treatment group was lower than that in the control group (
P
<0.05). In conclusion, Fe
3
O
4
@Apt nanoparticles exhibited excellent targeting specificity toward breast cancer cell 4T1, and the magnetic hyperthermia effectively inhibited 4T1 cell proliferation
in vitro
.
2024 Vol. 43 (6): 712-719 [
Abstract
] (
116
)
HTML
(1 KB)
PDF
(6292 KB) (
143
)
Reviews
720
Transcranial Magnetic Stimulation-Electroencephalography and Classified Diagnosis of Consciousness
Cheng Pengfei, Ge Xingyu, Gong Anjuan, Wang Qijun, Feng Zhen, Bai Yang
DOI: 10.3969/j.issn.0258-8021.2024.06.009
In recent years, there has still been a lack of an objective measure for the level of consciousness in specific states such as sleep, anesthesia, and disorders of consciousness. Attention has increasingly focused on non-invasive techniques for examining brain states. Transcranial magnetic stimulation combined with electroencephalography (TMS-EEG) has emerged as one of the most effective techniques in this field. TMS-EEG has become a promising brain assessment tool in contemporary brain science and clinical neuroscience research. This paper presented a systematic review of the development of TMS-EEG technology, focusing on the combination of TMS and EEG, denoising techniques, and characterization methods, as well as summarizing important research advances in the use of TMS-EEG for consciousness grading. The application of the perturbational complexity index (PCI) in differentiating levels of consciousness across wakefulness, different types of anesthesia and sleep, as well as in quantifying the grading diagnosis of patients with disorders of consciousness, has demonstrated that TMS-EEG is poised to become an objective gold standard for consciousness grading and diagnosis. Examples of objective metrics such as TMS-evoked potentials (TEP), global mean field power (GMFP), and the fractional dimensional index of perturbation complexity (FDI) for grading levels of consciousness in patients with disorders of consciousness, as well as for differentiating between healthy controls, were briefly described. The strengths, shortcomings and future directions of TMS-EEG were also summarized.
2024 Vol. 43 (6): 720-729 [
Abstract
] (
125
)
HTML
(1 KB)
PDF
(941 KB) (
158
)
730
Research Progress on Synthesis Methods of Gold Nanomaterials and its Applications in BiomedicalFields
Zheng Zuojing, Lin Yuhong, Zhao Kai
DOI: 10.3969/j.issn.0258-8021.2024.06.010
Nanotechnology has become a trend in scientific fields, and various metal nanomaterials have been widely developed in various biomedical aspects. Among them, gold nanomaterials have received special attentions due to their good biocompatibility, unique optical properties and easy to be functionalized. Nevertheless, preparations of gold nanomaterials by chemical, physical and microbial approaches suffer from environmental, economic and operational drawbacks, which limit the wide applications in industry and medicine. With the development of green synthesis and biomass synthesis, plant-mediated biosynthesis of metal nanomaterials has the advantages of non-pollution and non-toxicity, which can minimize the amount of hazardous chemicals and toxic by-products. This article reviewed the research progress of different preparation methods of gold nanomaterials and their applications in the fields of drug delivery, tumor therapy, antimicrobial, biosensing and imaging, providing references for the development research of more versatile gold nanomaterials.
2024 Vol. 43 (6): 730-740 [
Abstract
] (
219
)
HTML
(1 KB)
PDF
(3111 KB) (
422
)
741
Application of Recombinant Collagen in Tissue Regeneration
Zhang Yue, Jia Yuanyuan, Sun Xiuli, Wang Jianliu
DOI: 10.3969/j.issn.0258-8021.2024.06.011
Recombinant collagen is a novel biomaterial based on the genetic coding of specific functional domains of human collagen, synthesized through structural biology, genetic engineering, and related technologies. Its amino acid sequence is either identical or highly similar to that of human collagen, endowing it with low immunogenicity, excellent biocompatibility, and significant bioactivity. Consequently, recombinant collagen has shown considerable potential in tissue repair and regenerative medicine. This paper summarized the classifications, synthesis processes, and effects of recombinant collagen, while reviewing its advancements in the application of tissue regeneration for skin, cardiovascular tissues, skeletal structures, corneas, and pelvic floor tissues. Additionally, limitations of current research and future directions were discussed, aiming to provide new insights for the research and application of recombinant collagen.
2024 Vol. 43 (6): 741-750 [
Abstract
] (
143
)
HTML
(1 KB)
PDF
(1135 KB) (
263
)
751
Antibacterial Mechanism of Antibacterial Peptide and its Application in Local Anti-InfectionTreatment of Diabetes Foot Ulcer
Zhang Zeyu, Liu Yufei, Zhou Jie, Liu Xiangsheng, Xu Jun, Wang Shufang
DOI: 10.3969/j.issn.0258-8021.2024.06.012
Infectious diseases caused by pathogenic microorganisms can cause inflammation and tissue damage, which slow down the healing process, especially for chronic wounds such as diabetic feet and traumatic ulcers. Patients often suffer from impaired immune function combining with factors such as vascular disease and neuropathy around the wound surface. Therefore, bacteria are prone to multiply in the infected wound site and form biofilms, thus making it difficult for bacteria to be removed, increasing the severity of trauma site infection, and seriously affecting the subsequent life of patients. Antimicrobial peptides, as a class of active small molecule polypeptides, can achieve broad-spectrum antibacterial effect by destroying bacterial cell membrane and other methods. Some antimicrobial peptides can inhibit or destroy biofilms by inhibiting bacterial adhesion, interfering with the formation of biofilm components and interfering with bacterial quorum sensing. Combined with anti-inflammatory, immunomodulatory and other multi-functions, the peptides have been widely used in the treatment of anti-infection and chronic trauma. This review introdced antibacterial and anti-biofilm mechanisms and summarized applications of the antimicrobial peptides in anti-infection treatments of diabetic foot ulcers, and analyzed the advantages and disadvantages of the various applications. Accordingly, development prospects of the antimicrobial peptides were proposed, aiming to provide practical approaches for the anti-infection treatment of diabetic foot ulcers.
2024 Vol. 43 (6): 751-758 [
Abstract
] (
105
)
HTML
(1 KB)
PDF
(2831 KB) (
173
)
Communications
759
Using Approximate Entropy to Modular Process and Analyze EOG Signals of Autism Spectrum Disorder
Lin Dong, Qi Shiyi, Yuan Ding, Zhang Sijia, Zhuang Wanyu, Lin lili, Li Yurong
DOI: 10.3969/j.issn.0258-8021.2024.06.013
2024 Vol. 43 (6): 759-763 [
Abstract
] (
71
)
HTML
(1 KB)
PDF
(2301 KB) (
74
)
764
Study on Ultra-Fine Crystalline Titanium-Based Dopamine/Hydroxyapatite Composite Coating
Jiang Yucong, Chi Yanxia
DOI: 10.3969/j.issn.0258-8021.2024.06.014
2024 Vol. 43 (6): 764-768 [
Abstract
] (
96
)
HTML
(1 KB)
PDF
(3487 KB) (
112
)
Copyright © Editorial Board of Chinese Journal of Biomedical Engineering
Supported by:
Beijing Magtech
