Chinese Journal of Biomedical Engineering

Home

Editorial Board

Download

Quick Search

Office Online

Current Issue

Download Articles

Download

Links

2022 Vol. 41, No. 5 Published: 2022-10-20

Reviews Communications Regular Papers

Regular Papers

	513	Glioma Segmentation Based on Feature Selection of Multi-Modal MR Images
		Cheng Juan#,Zhang Chuya,Liu Yu#*,Li Chang,Zhu Zhiqin,Chen Xun#
		DOI: 10.3969/j.issn.0258-8021.2022.05.001
		Glioma segmentation based on multi-modal MR images plays a positive role for the diagnosis and treatment of tumors. It is known that different modalities of MR images can provide different properties of information representing pathological tissues. Currently, an increasing number of deep-learning-based glioma segmentation methods have been proposed to segment brain gliomas utilizing multi-modal MR images. However, these methods usually stack the original features derived from multi-modal MR images channel by channel, and roughly take the stacked features as inputs, leading to an inadequate feature mining and an unsatisfactory segmentation performance. To solve this problem, this paper proposed to segment three glioma regions with a two-stage segmentation scheme, with each stage having a feature selection module and a segmentation network. The first stage of the segmentation aimed to segment peripheral edema regions, while the second stage tried to segment necrosis/non-enhancing tumor and enhancing tumor regions. Besides, the first-stage segmentation results provided essential location information that would benefit for segmenting the other two tumor regions during the second stage. For each stage, a multi-modal feature selection module was designed to automatically extract effective and cross-modal-fused features from each modality of MR images, and then these features were sent to each following segmentation network. The segmentation network was composed of a V-Net and a variational autoencoder (VAE). Experiments were conducted on three public brain tumor datasets including BraTS2018, BraTS2019 and BraTS2020. Specifically, for dataset BraTS2018, the average Dice scores of the proposed method for segmenting the whole tumor (WT), the tumor core (TC), and the enhanced tumor (ET) regions reached 0.898, 0.854, and 0.818, respectively, while the Hausdorff95 distance of the proposed method for segmenting the aforementioned three regions reached 4.072, 6.179, and 3.763, respectively. As for dataset BraTS2019, the average Dice scores of the proposed method for segmenting the abovementioned three tumor regions reached 0.892, 0.839, and 0.800, respectively, while the corresponding Hausdorff95 distance of the proposed method can reach to 6.168, 7.077 and 3.807, respectively. As for dataset BraTS2020, the average Dice scores of the proposed method for segmenting the same three regions reached 0.896, 0.837, and 0.803, respectively, while the corresponding Hausdorff95 distance of the proposed method reached 6.223, 7.033, and 4.411, respectively. The results of the comparison experiments demonstrated the obvious superior performance of the proposed method in segmenting ET and TC regions, especially that the performance of ET segmentation was the best in BraTS2020. Owing to the proposed two-stage segmentation scheme, with each having a feature selection module followed by a segmentation network, the potentially cross-modality-fused features could be automatically extracted from each modality of MR images, thus the performance of segmenting the three tumor regions was significantly improved.
		2022 Vol. 41 (5): 513-526 [Abstract] ( 287 ) HTML (1 KB) PDF (9389 KB) ( 306 )

	527	Classification of Breast Tumors Based on Dual-Branch Multi-View Transformer
		Liu Yiyao,Yang Yi,Chen Minsi,Wang Tianfu,Jiang Wei,Lei Baiying,
		DOI: 10.3969/j.issn.0258-8021.2022.05.002
		Automatic breast volume scanner (ABVS) system is the primary screening method for breast cancer because of its high efficiency and no radiation. The study of computer-aided breast cancer classification based on ABVS images is helpful for clinicians to diagnose breast cancer accurately and quickly and can even help to improve the diagnostic level of junior doctors. Because of its imaging mode, ABVS system produces a large amount of three-dimensional breast image data, leading to a long training time and huge resources of conventional deep learning. Therefore, we designed a multi-view image extraction method based on ABVS data, which replaced the conventional 3D data and made up for the spatial correlation in 2D deep learning while reducing the number of parameters. Secondly, based on the spatial position relationship of cross view images, we proposed a self-attention encoder (Transformer) to obtain effective feature expression of the images. Our experiment was based on 153 volume images from our own ABVS database. The accuracy of benign and malignant classification was 86.88%, the F1 score was 81.70% and AUC reached 0.831 6. The experimental results indicated that the proposed method could be effectively applied to the benign and malignant screening of breast tumors based on ABVS images.
		2022 Vol. 41 (5): 527-536 [Abstract] ( 227 ) HTML (1 KB) PDF (6899 KB) ( 245 )

	537	Head Pose Estimation of Patients with Monocular Vision for Surgery Robot Based on Deep Learning
		Feng Pengfei&,Li Liang&#,Ding Hui,Wang Guangzhi#*
		DOI: 10.3969/j.issn.0258-8021.2022.05.003
		Patient head pose estimationis one of the key technologies for autonomous and intelligent perception of neurosurgery robots. This paper aimed to use data-driven deep learning method to help the neurosurgery robot to estimate the patient′s head posture, laying the foundation for the intelligence of neurosurgery. This paper firstly established the basic mathematical relationship of the patient head pose estimation task. Next, an efficient and robust head pose labeling method was proposed to solve the problem of 2D head image pose labeling in the absence of facial features. After that, by collecting the neurosurgery scene photos from the perspective of the robot, a patient head pose estimation dataset containing a total of 79 surgical scenes and a total of 4 301 photos was constructed. Finally, the applicability of the HopeNet deep neural network in the patient head pose estimation problem was studied, and methods including cropping, rotation data augmentation, and our newly proposed rotation rate loss function improved the model performance. For the network training and evaluation, on the homologous test set 1 containing 10 surgical scenes and 386 pictures, the pose estimation based on a single perspective could reach an average of ±12.76°in three directions including yaw angle, roll angle, and pitch angle; on the heterogeneous test set 2 of 8 surgical scenes and 229 photos, the average prediction error of ±13.41° could be achieved in the three directions. The results showed that the proposed model could accurately estimate the patient′s head pose, and the proposed optimization methods could effectively improve the accuracy of the algorithm and improve the generalization performance of the model.
		2022 Vol. 41 (5): 537-546 [Abstract] ( 186 ) HTML (1 KB) PDF (7136 KB) ( 166 )

	547	Evaluation of Rehabilitation Status of Patients with Chronic Ischemic Stroke Based onQuantitative EEG
		Li Cenbo,Chen Long*,Gu Bin,Wang Zhongpeng,Zhang Xin,Ming Dong,#
		DOI: 10.3969/j.issn.0258-8021.2022.05.004
		Rehabilitation assessment can help clinicians to understand patients rehabilitation situation, formulate a reasonable rehabilitation training plan, and improve the efficiency of rehabilitation treatment. However, the traditional rehabilitation evaluation methods cannot timely reflect and dynamically evaluate the rehabilitation situation, which would reduce the efficiency of the rehabilitation treatment of stroke patients. In order to solve the above problems, quantitative electroencephalogram (qEEG), an objective rehabilitation evaluation method, was introduced in this study, which has the advantages of convenient collection and rapid feedback, and can quickly reflect the rehabilitation situation of stroke patients and make up for the shortcomings of traditional rehabilitation assessment methods. In this study, 28 patients with chronic ischemic stroke were enrolled. The EEG data of the patients were collected and thepower ratio index, Brain symmetry index andphase synchrony index were calculated, and Spelman correlation coefficient was used to analyze the correlation between the above-mentioned qEEG and the modified barthel index and the upper extremity motor section of the Fugl-Meyer motor assessment (FMA). The results showed that there was a significant positive correlation between the modified barthel index and the IH-PSI in the β band of the global channel (ρ=0.70 P=0.000 2). The correlation between the modified barthel index and the IH-PSI in the β band of the global channel was the strongest compared with the other qEEG, the correlation coefficient was ρ = 0.70, and P=0.000 2<0.01, which was statistically significant. The correlation between the upper extremity motor section of the FMA and the C3FC3C4FC4-PSI in the β band using four channels of C3, C4, FC3, FC4 was the strongest compared with the other qEEG, the correlation coefficient was ρ=0.71, and P=0.000 09<0.01, which was statistically significant. In summary, the use of the IH-PSI in the β band of the global channel combined with the modified barthel index, the use of the C3FC3C4FC4-PSI in the β band using four channels of C3, C4, FC3, FC4 combined with the upper extremity motor section of the FMA was expected to achieve dynamic assessment of patients with chronic ischemic stroke, and improved the efficiency of rehabilitation treatment for stroke patients.
		2022 Vol. 41 (5): 547-557 [Abstract] ( 215 ) HTML (1 KB) PDF (1296 KB) ( 360 )

	558	Diagnosis Model of Chronic Obstructive Pulmonary Disease Based on Deep Learning
		Yu Hui,#,Zhao Jing,Qiu Zhaoyu,Liu Dongyi,Chen Zhen,Gou Chengxiang,Sun Jinglai,Zhao Xiaoyun
		DOI: 10.3969/j.issn.0258-8021.2022.05.005
		Chronic obstructive pulmonary disease (COPD) is a common chronic respiratory disease characterized by continuous airflow restriction, with high morbidity and mortality. At present, clinical diagnosis methods of COPD are very complex, not only time-consuming and invasive or radioactive, and not suitable for daily screening. Therefore, a COPD diagnosis model based on deep learning was designed in this study. Firstly, the lung sound from 42 COPD patients from RespiratoryDatabase@TR multimedia respiratory database were combined with the clinically collected lung sound from 24 COPD patients and 37 healthy subjects from Chest Hospital of Tianjin University, high-pass filter and denoising algorithm based on ensemble empirical mode decomposition (EEMD) and wavelet entropy was used for denoising. Secondly, the pre-processing process was completed through normalization, overlapping shear and data amplification. Thirdly, bispectrum analysis was used to extract the lung sound features. Finally, these features were input into an improved 19-layer convolutional neural network model to achieve the binary classification of healthy subjects and COPD patients. Experimental results showed that the proposed model could effectively diagnose COPD. The accuracy, sensitivity, specificity, F1 score, and Kappa score reached 98.93%, 98.47%, 99.41%, 98.95%, and 97.86%, respectively. Moreover, due to the use of bicentric data and denoising process, the model has higher reliability and is of important clinical significance.
		2022 Vol. 41 (5): 558-566 [Abstract] ( 299 ) HTML (1 KB) PDF (2770 KB) ( 470 )

Reviews

	567	Review on Applications of U-Net and its Variants in Medical Image Segmentation
		Huang Xiaoming,He Fuyun,*,Tang Xiaohu,Wang Xun,Qiu Senhui,Hu Cong
		DOI: 10.3969/j.issn.0258-8021.2022.05.006
		Medical image segmentation can provide a reliable basis for clinical diagnosis and pathology research and assist doctors to make accurate diagnosis. The emergence of medical image segmentation based on deep learning has solved the problems of low robustness and low accuracy in traditional automatic segmentation methods, among which, U-Net stands out among many segmentation networks with its excellent performance. Researchers have successively proposed a variety of improved variants based on U-Net. Taking U-Net and its network variants as the main content, this article first introduced the network structure and common improvement methods of U-Net in detail. Then, divided the U-Net variants into general-purpose networks and specific network according to the different segmentation objects, and discussed the research progress of these networks in the medical image segmentation. At the end, the difficulties and problems existing in this research field were analyzed, and the development directions were prospected.
		2022 Vol. 41 (5): 567-576 [Abstract] ( 565 ) HTML (1 KB) PDF (2487 KB) ( 1410 )

	577	Review of Researches on Common Spatial Pattern and its Extended Algorithms for MovementIntention Decoding
		Pan Lincong,Wang Kun,Xu Minpeng,Ni Guangjian,Ming Dong,
		DOI: 10.3969/j.issn.0258-8021.2022.05.007
		Motor imagery based-brain-computer interfaces (MI-BCIs) are of important research significance and application value in rehabilitation, replacement, and enhancement of human motor function. Common spatial pattern (CSP) algorithm aims to enhance the difference of electroencephalography (EEG) features induced by MI, which is currently one of the most widely used feature extraction algorithms for MI paradigm. However, it does not focus on the time and frequency domain information of EEG, and is sensitive to noise and deviation values, resulting in the limited recognition performance and the low robustness of classifiers. This paper reviewed the development history of CSP and its extended algorithms. We introduced the basic principles and key calculation steps of representative extended algorithms in detail from three aspects: multi-modal information optimization, regularization optimization and other spatial mapping optimization methods. In addition, we discussed the actual challenges and predict the future development trend, aiming to promote the in-depth research and application of relevant BCI technology.
		2022 Vol. 41 (5): 577-588 [Abstract] ( 274 ) HTML (1 KB) PDF (1017 KB) ( 735 )

	589	Research Progress of 3D Printing Organ-on-a-Chip
		Luo Zhiming,Deng Guohao,Wang Zhubing,Zhang Yujie,Zhang Yang,Wang Huanan,
		DOI: 10.3969/j.issn.0258-8021.2022.05.008
		Organ-on-a-chip is a biomimetic microphysiological system that recapitulates certain functions of tissues and organs by seeding live cells in delicately designed microfluidic chips. Organ-on-a-chip has a wide application in disease simulation, toxicity detection, new drug development and screening, personalized medical treatment, etc. The application of 3D printing organ-on-a-chip realizes the simplification and low cost of the chip manufacture and meet the delicate requirements of complex heterogeneous three-dimensional microenvironments. This review updated the recent progress in 3D printing organ-on-a-chip from the aspects of printing methods, printing inks and their biomedical applications. We reviewed the printing methods and ink materials for the construction of structural and functional units of organ chips and discussed the potential to develop more efficient one-step manufacture of organ chip. At last, we summarized the recent progress of using 3D-printed organ chip in bionic heart, liver, lung, kidney, nerve, tumor models, and outlook the future trend of organ-on-a-chip techniques.
		2022 Vol. 41 (5): 589-601 [Abstract] ( 332 ) HTML (1 KB) PDF (6995 KB) ( 248 )

	602	Review of the Portable Fully Integrated Nucleic Acid Analysis System
		Li Nan#,Xu Youchun,,Cheng Jing,#
		DOI: 10.3969/j.issn.0258-8021.2022.05.009
		Nucleic acid testing plays important roles in many fields, such as clinical diagnosis, public health, food safety, molecular breeding, and forensic identification. In standard nucleic acid testing laboratories, physically separated spaces are required for sample preparation, nucleic acid extraction, and amplification, and many manual operations and supporting equipment are involved, which makes the entire detection process tedious, inefficient, and prone to cross-contamination. Therefore, the trend of nucleic acid detection and analysis is to develop fully integrated and automated nucleic acid detection systems. With the development of microfluidic technology, it is now possible to construct a fully integrated portable nucleic acid analysis system to satisfy the needs of point-of-care testing and/or on-site detection. This article reviewed the technical characteristics of the fully integrated portable nucleic acid analysis system from the aspects of biotechnology and microfabrication technology, compared some fully integrated representative systems for nucleic acid analysis, and proposed its technical characteristics of multi-disciplinarity, diversification of technology paths, full integration and functional expansion. At the end of the article, we concluded the challenges of complex molecular diagnostic requirement, high requirement for rapid detection, fully automatic operation and low cost for the portable fully integrated nucleic acid detection system technology, as well as discussed the possible future development.
		2022 Vol. 41 (5): 602-613 [Abstract] ( 371 ) HTML (1 KB) PDF (2651 KB) ( 518 )

	614	Strategies of Modulating Macrophage Behavior to Promote Bone Healing in Bone Tissue Engineering
		Shan Qin,*
		DOI: 10.3969/j.issn.0258-8021.2022.05.010
		In bone tissue engineering, the biomaterial is implanted into the bone defect as a regeneration scaffold to help promote bone healing. With the development of osteoimmunology in recent years, people have realized that the immune response of macrophages determines whether the implantation is successful. Macrophages, which are diverse and plastic, can be polarized into M1-like (pro-inflammatory) and M2-like (anti-inflammatory) phenotypes according to environmental signals and play an important role in different stages of bone healing. Their behavior like migration, proliferation and polarization is sensitive to the properties of the material. This article reviews the current strategies of modulating the behavior of macrophages to promote bone healing by designing the physical properties, surface chemical properties, and biological properties of materials, including increasing the roughness of the material, macropores combining with nanostructures, appropriate electrical or mechanical signals, neutral or anionic surfaces, increasing hydrophilicity, using immunomodulatory materials or delivering immunological active substances to the defect area, etc., which provide ideas for the design of bone tissue engineering materials with good immunomodulatory properties.
		2022 Vol. 41 (5): 614-620 [Abstract] ( 190 ) HTML (1 KB) PDF (822 KB) ( 400 )

Communications

	621	Research on Sliding Mode PD Control of Lower LimbExoskeleton Walking Aid Robot
		Xing Wenqi,Guo Xudong*,Xiao Jianru,Xu Wei,Wei Qiangsheng,Liu Jiannan
		DOI: 10.3969/j.issn.0258-8021.2022.05.011

		2022 Vol. 41 (5): 621-625 [Abstract] ( 182 ) HTML (1 KB) PDF (1662 KB) ( 255 )

	626	Microaneurysm Detection Method Based on Mixed Multi-Features
		Long Shengchun*,Hu Ante,Chen Zhiqing
		DOI: 10.3969/j.issn.0258-8021.2022.05.012

		2022 Vol. 41 (5): 626-630 [Abstract] ( 175 ) HTML (1 KB) PDF (2913 KB) ( 800 )

	631	Energy Analysis and Experimental Verification of Tissue Focal Region under the Action ofHIFU
		Tao Jie,Wang Yuebing*,Tong Shiqi,Zhao Peng,Sheng Yongjie
		DOI: 10.3969/j.issn.0258-8021.2022.05.013

		2022 Vol. 41 (5): 631-635 [Abstract] ( 194 ) HTML (1 KB) PDF (2022 KB) ( 185 )

	636	Screening Study on Biodegradable Materials for Tumor Bed Marker in Breast-Conserving Surgery
		Xie Xin,Zhang Kaixuan,Chen Xinyuan,Dai Jianrong,*
		DOI: 10.3969/j.issn.0258-8021.2022.05.014

		2022 Vol. 41 (5): 636-640 [Abstract] ( 176 ) HTML (1 KB) PDF (1889 KB) ( 441 )

Copyright © Editorial Board of Chinese Journal of Biomedical Engineering
Supported by:Beijing Magtech