Chinese Journal of Biomedical Engineering

Home

Editorial Board

Download

Quick Search

Office Online

Current Issue

Download Articles

Download

Links

2021 Vol. 40, No. 4 Published: 2021-08-20

Reviews Regular Papers

Regular Papers

	385	Prognostic Analysis Model of Renal Clear Cell Carcinoma Based on Multi-Dictionary Learning
		Tu Chao, Ning Zhenyuan, Zhang Yu
		DOI: 10.3969/j.issn.0258-8021.2021.04.01
		Clear cell renal cell carcinoma (ccRCC) is a highly heterogeneous tumor with complex and variable clinical manifestations. Automatic histopathological whole slide image (WSI) analysis is a useful approach for pathologists to make diagnosis. However, feature extraction for the prognostic analysis of ccRCC is a challenging task due to the diversity of tissue structures in the histopathological images. In this work, a novel WSI-based multi-dictionaries learning framework was proposed to adaptively extract the effective features of WSI for prognostic analysis of ccRCC. This framework included multi-dictionaries learning stage based on patch level and survival model construction stage based on patient level. The proposed model was evaluated on 378 hematoxylin-eosin stained WSIs form Cancer Genome Atlas database (TCGA-KIRC). The C-index was 0.681, and AUC was 0.751(P＜0.05). Compared with the traditional Boosted model and Random Survival Trees model, the improvements on C-index were respectively 0.138 and 0.155, and the improvements on AUC was respectively 0.149 and 0.191. Compared with the two deep learning model (DeepSurv and WSISA), the improvements on C-index were respectively 0.046 and 0.035, and the improvements on AUC was respectively 0.096 and 0.090. The results showed that the proposed model achieved superior performance for prognostic analysis of renal clear cell carcinoma.
		2021 Vol. 40 (4): 385-393 [Abstract] ( 435 ) HTML (1 KB) PDF (6900 KB) ( 251 )

	394	Establishment of a Rehabilitation Training Prescription Recommended Model for Stroke Based on Brain Function and Movement Assessment
		Zhang Tengyu, Zhang Jingsha, Xu Gongcheng, Wang Zheng, Zhang Xuemin, Li Zengyong
		DOI: 10.3969/j.issn.0258-8021.2021.04.02
		To establish intelligent recommendation model for rehabilitation training prescriptions based on the evaluation indexes of movement and brain function for stroke patients,120 stroke patients were evaluated by Brunnstrom, Holden walking ability scale, Berg balance scale and improved Ashworth scale, and the cerebral blood oxygen data were collected at rest state by near-infrared brain functional imaging system. The scale evaluation results and the brain function evaluation indexes (activate degrees, cornering, and function connection between different brain regions) were extracted, the support vector machine (SVM), the convolutional neural network (CNN) and the improved CNN-SVM algorithm were used to build the prescription recommended model, and the training modes of eight kinds of training contents were recommended. The model using the scale evaluation results and the functional connectivity index between different brain regions as features had the highest recognition accuracy, and the average recognition accuracy of the three algorithms was above 93%. Among them, the recognition accuracy of the improved CNN-SVM algorithm was highest, which reached 96.43%. This method realized the intelligent recommendation of rehabilitation training prescription based on the movement and brain function evaluation data of stroke patients, which would be beneficial to personalized and accurate rehabilitation.
		2021 Vol. 40 (4): 394-400 [Abstract] ( 457 ) HTML (1 KB) PDF (1226 KB) ( 743 )

	401	DCE-MRI Radiomics Based Non-negative Matrix Factorization for Imputation of Missing Histological Information of Breast Cancer
		Fu Zhenyu, Fan Ming, Li Lihua
		DOI: 10.3969/j.issn.0258-8021.2021.04.03
		Breast cancer pathology report is the main basis for the diagnosis and treatment of breast cancer. However, sometimes there may loss of histological information in the clinical practices. In this study, imaging features of the lesion area of the dynamic enhanced magnetic resonance imaging (DCE-MRI) were combined with the histological information of the corresponding breast cancer patients to establish a non-negative matrix factorization based radiomics model to achieve the imputation of missing molecular subtypes and Cytokeratin 5/6 gene expression. A total of 139 cases of breast cancer patients were collected before surgery or before chemotherapy and were randomly divided into 89 cases as training set and 50 cases as test set. Breast tumor areas were segmented and the morphological and texture features were extracted from the lesion area and statistically analyzed. The cross-validated support vector machine recursive feature elimination (SVM-RFECV) method was used for the feature selection, and the image features were further filtered through a union-based method. Combining the clinical pathological information of breast cancer, a non-negative matrix factorization (NMF) imputation model and a collaborative filtering (CF) imputation model were established, and the AUC was calculated to evaluate the imputation performance of the model. When the clinical pathological information missing rate was different, the AUC value of the NMF model was higher than that of the CF model, the highest AUC was 0.772, and the NMF imputation effect was significantly better (P<0.05) than the CF method when the missing rate was between 20% and 40%. In the case of quantitative image features, the AUC value of the NMF model was higher than that of the CF model, the highest AUC was 0.780, and the difference between the two was statistically significant (P<0.05) when 140 image features were used. These experimental results showed that DCE-MRI radiomics combined with non-negative matrix factorization effectively filled the missing molecular subtypes and CK5/6 clinical indicators.
		2021 Vol. 40 (4): 401-409 [Abstract] ( 310 ) HTML (1 KB) PDF (2153 KB) ( 572 )

	410	Mask R-CNN and Data Augmentation and Transfer Learning
		Wang Congzhi, Xu Zibi, Ma Xiangyuan, Hong Zilan, Fang Qiang, Guo Yanchun
		DOI: 10.3969/j.issn.0258-8021.2021.04.04
		In clinical practices, the segmentation and modeling of brain regions in brain CT images can better observe the relationship between the lesion and the location of each organ. At present, the segmentation is mainly divided by manual outline, which is time-consuming, laborious and susceptible to subjective influence. In this paper, a Mask R-CNN based on augmentation and transfer learning was proposed, aiming to segment several brain regions vulnerable to cerebral hemorrhage from brain CT images more quickly and automatically, the regions included cerebellum, brainstem, basal ganglia region and dorsal thalamus. In this paper, 1 549 brain CT images of 100 cases of healthy people from July 2020 to December 2020 were analyzed. A total of 1 239 brain CT images of 80 cases were selected as the training set, and 310 brain CT images of the remaining 20 cases were selected as the test set. Then, the Mask R-CNN framework was used for training and prediction. Finally, the coordinates, names and masks of each brain region were output. To study the effect of data augmentation and transfer learning on model training, experiments of data augmentation and transfer learning were designed respectively, and the control group of U-NET model was designed. The data augmentation group expanded the training set to 13 629 images by means of rotation. In the transfer learning group, transfer learning was carried out based on the weights trained in MS-COCO. Among them, the transfer learning group had the best effect. In the experiment of transfer learning, the test set mAP was 0.909 7, the average IOU was 0.736 2, and the average DICE values of the test set of brain stem, cerebellum, basal ganglia region and dorsal thalamus were 0.902 5, 0.879 5, 0.781 8 and 0.828 4, respectively. The mAP and average IOU without data augmentation and transfer learning were 0.870 8 and 0.715 9, respectively. Data augmentation group were 0.894 1, 0.729 7; U-NET group were 0.839 0 and 0.671 1. These results showed that the Mask R-CNN convolutional neural network model could be used in the automatic segmentation of the common parts of cerebral hemorrhage, and the transfer learning greatly improved the training effect of the model.
		2021 Vol. 40 (4): 410-418 [Abstract] ( 664 ) HTML (1 KB) PDF (3896 KB) ( 776 )

	419	dMRI Fiber Tracking with Functional MRI of White Matter
		Dong Xiaofeng, Yang Zhipeng, Wu Xi
		DOI: 10.3969/j.issn.0258-8021.2021.04.05
		Diffusion MRI based tractography is the primary tool for mapping white matter structures in the brain. However, existing tractography algorithms are constrained by the diffusion MRI resolution and imaging mechanism, and the accuracy would greatly be reduced when streamlines enter the boundary region between white matter and gray matter. In order to overcome this defect, a new diffusion MRI tractography algorithm combined with functional magnetic resonance imaging is proposed in this work. We introduced the spatial correlation tensor derived from functional magnetic resonance imaging signal anisotropy in the white matter to indirectly describe the geometrical information of the fiber bundle. Then particle filter theory was used to estimate the directional probability distribution and reconstruct the three-dimensional structure of the boundary region of white matter. Finally, the proposed method was used to tracking on functional images of visual stimulation of 8 adults. There were 800 fibers reconstructed in each subject. The average length of the reconstructed fibers reached (18.47±1.82) mm and the coverage of streamline endpoints along the white matter gray matter interface reached 25.15%±1.86%. The results showed that the proposed method effectively reconstructed the white matter fiber path of the brain, especially the area where there were errors in the fiber bundle reconstruction of boundary area between the white matter and gray matter due to the partial volume effect. In conclusion, the proposed method can obtain more accurate results than the traditional tractography methods.
		2021 Vol. 40 (4): 419-428 [Abstract] ( 303 ) HTML (1 KB) PDF (6042 KB) ( 356 )

	429	Improve the Performance of Lower Limb MI-BCI System Based on SSSEP and its Multi-Dimensional EEG Feature Analysis
		Zhang Lixin, Chang Meirong, Wang Zhongpeng, Chen Long, Ming Dong
		DOI: 10.3969/j.issn.0258-8021.2021.04.06
		Motor imagery(MI)-based brain-computer interface (MI-BCI) can decode motor intention of users, providing an additional interactive manner for patients who are unable to exercise autonomously and improving their lifestyle. To solve the key problem of low classification performance of lower limb MI-BCI, we designed a hybrid paradigm, i.e. MI joint somatosensory electrical stimulation (MI+ES) inducing steady state somatosensory evoked potential (SSSEP) for MI-BCI of lower limb. And the performance of MI+ES was compared with the traditional single paradigm (MI). Twenty right-handed healthy subjects were recruited to participate in the experiment, five of them participated in the verification test of optimal induced frequency and fifteen participated in the formal experiment. EEG data of the fifteen subjects were recorded under different conditions. Fast Fourier transform (FFT) and event-related spectral perturbation (ERSP) were used to extract EEG frequency domain response and time-frequency features. The multi-frequency power changes were calculated at alpha (8～14 Hz), low beta (15～24 Hz) and high beta (25～35 Hz) bands. In addition, the performance of lower limb MI-BCI was explored under different conditions of MI/MI+ES and feature extraction methods of CSP/FBCSP. Results showed that the somatosensory electrical stimulation strategy could induce obvious SSSEP features. The classification accuracy of MI+ES condition was significantly improved in reference to the single MI condition (P<0.001). The classification performance based on FBCSP method was significantly better than that of classical CSP method (P <0.01), the classify accuracy of CSP was 70.2% under MI+ES condition, while the accuracy of subject S15 was 84.2%. And the accuracy of FBCSP was 71.7%, the accuracy of subject S15 was 90%. In conclusion, this study preliminarily confirmed that the SSSEP could be evoked by the somatosensory electrical, and the hybrid paradigm could effectively improve the classification performance of lower limb MI-BCI, which could promote the practical development, even provide optimization methods of peripheroneural somatosensory stimulation regulation.
		2021 Vol. 40 (4): 429-437 [Abstract] ( 398 ) HTML (1 KB) PDF (9775 KB) ( 153 )

	438	Recurrence Networks Analysis of Multiple Muscles coordination of Forearm and Hand for the Grip and Pinch Control
		Zhang Na, Li Ke
		DOI: 10.3969/j.issn.0258-8021.2021.04.07
		The aim of this study is to explore the muscle coordination of forearm and hand for the grip and pinch control. There were 24 right-handed healthy subjects participating in the experiment. Subjects were requested to produce precisely 30%, 50% and 70% maximum voluntary contraction (MVC) of grip and pinch, during which the surface electromyography (sEMG) signals were recorded from brachioradialis (BR), flexor carpi ulnaris (FCU), flexor carpi radialis (FCR), extensor digitorum communis (EDC), flexor digitorum superficialis (FDS), abductor pollicis brevis (APB), first dorsal interosseous (FDI) and abductor digitiminimi (ADM). The recurrence networks (RNs) and multiplex recurrence networks (MRNs) based on multivariate sEMG signals were constructed and then analyzed by the average shortest path length (L), the clustering coefficient (C) and the interlayer mutual information (I), the average edge overlap (ω). Results showed that the RNs of BR, FCU and FCR during grip performed significantly higher C values than that during pinch under three force levels, take FCR as an example, grip vs pinch: 0.393 ± 0.040 vs 0.366 ± 0.035, 0.404 ± 0.040 vs 0.372 ± 0.035, 0.412 ± 0.051 vs 0.383 ± 0.040, P<0.05;while for the RNs of FDI, grip vs pinch: 0.443 ± 0.035 vs 0.462 ± 0.046, 0.446 ± 0.032 vs 0.461 ± 0.035, 0.445 ± 0.040 vs 0.465 ± 0.038, P<0.05. However, the L values of BR, FCU and FCR of RNs during pinch were significantly higher than that during grip at three force levels, take FCU as an example, grip vs pinch: 2.870 ± 0.063 vs 2.941 ± 0.124, 2.841 ± 0.065 vs 2.941 ± 0.079, 2.830 ± 0.083 vs 2.901 ± 0.051, P<0.05. The I and ω values of extrinsic MRNs under 50% and 70% MVC during grip were 4.056 ± 0.248 and 4.099 ± 0.232, 0.253 ± 0.015 and 0.257 ± 0.017, which were significantly higher than that during pinch (3.930 ± 0.229 and 3.939 ± 0.195, 0.245 ± 0.011 and 0.246 ± 0.012, P<0.05). In addition, the C of BR, FCU and FCR increased, the L decreased, and the I and ω of the extrinsic muscles MRNs increased with the grip force augmented. These results suggested different muscle coordination pattern between grip and pinch, and theintermuscular similarity and synchronization of extrinsic muscleswouldfuther augmente with the increased force level. These findings revealed the dynamical coordination across muscles with the force outputs and provided novel strategy for evaluating the neuromuscular function and making of the myoelectric prosthesis.
		2021 Vol. 40 (4): 438-445 [Abstract] ( 273 ) HTML (1 KB) PDF (4420 KB) ( 322 )

	446	Effect of Mental Fatigue on Different Working Memory Loads Based on Brain Network
		Yang Shuo, Jiang Wentao, Wang Lei, Xu Guizhi
		DOI: 10.3969/j.issn.0258-8021.2021.04.08
		Working memory is the basis of cognitive behavior. Brain fatigue can lead to the decline in cognitive performance such as memory decline and slow response. In this paper, the brain fatigue experiment was designed by using the character N-back experimental paradigm. The electroencephalogram data of four kinds of visual working memory load were collected from 15 subjects under normal and fatigue conditions and functional brain network was constructed by using mutual information algorithm. Four network characteristic parameters including node degree, node intermediate center degree, node clustering coefficient and node feature path length were calculated and analyzed. The behavior data (reaction time, correct rate) were statistically analyzed. Results showed that under the four working memory loads, compared with the normal state, the brain network node degree in the fatigue state, the centrality of the node and the node clustering coefficient decreased, and the node characteristic path length increased. The nodes with significant differences were mainly distributed in the frontal lobe, parietal lobe and occipital lobe. Results of behavioral data showed that in the normal state, the average response times of the four working memory loads were (578±46), (664±45), (868±44) and (959±42) ms. In the fatigue state, the average response times of the four working memory loads were (656±41), (721±46), (941±50) and (1038±50) ms. The average response time of fatigue state was significantly higher than that of the normal state, and brain fatigue had a negative effect on the working memory under the 4 loads. The research results showed that under the four working memory loads, the brain fatigue had a certain effect on the working memory. In conclusion, the brain fatigue inhibited the increase of brain activity, and brain fatigue had a more obvious effect on the high load than on the low load.
		2021 Vol. 40 (4): 446-452 [Abstract] ( 436 ) HTML (1 KB) PDF (2711 KB) ( 625 )

	453	EEG Stress Emotion Analysis Based on Variable-Scale Symbolic Compensation Transfer Entropy
		Gao Yunyuan, Wang Xiangkun, Tian Yuping, She Qingshan, Dong Hua
		DOI: 10.3969/j.issn.0258-8021.2021.04.09
		Emotion recognition based on EEG signals has important clinical and scientific significance for the diagnosis and treatment of related emotional diseases. How to effectively extract features, improve recognition rate and reduce calculation time is the focus of this paper. From the perspective of studying the directional information interaction between brain channels, this paper combined the compensation algorithm for instantaneous causal effects and proposed an emotional analysis method of Variable-Scale Symbolic Compensation Transfer Entropy. This method was used to construct an emotional causal effect brain network, the network measurement and ReliefF feature optimization selection algorithm were used for channel selection. The results showed that the feature extraction method of VSSCTE improved the accuracy of emotion classification by about 15% to 96.74% over the conventional binary transfer entropy method when using data from the DAEP dataset of 127 stresses and 125 calms. After optimization of EEG channels, when the number of channels was reduced from 32 to 15, the classification accuracy rate only droped by about 2% (the classification accuracy rate was 94.36%), but the calculation time was reduced by about 110%. Overall, the VSSCTE method proposed in this paper was able to effectively analyze the information interaction between brain regions of different emotional states, providing a new method and ideas for emotional analysis.
		2021 Vol. 40 (4): 453-460 [Abstract] ( 291 ) HTML (1 KB) PDF (3382 KB) ( 504 )

	461	Effect of Mental Fatigue on Alpha Oscillation Information Integration of Working Memory
		Yang Shuo, Peng Sen, Wang Lei, Wang Zengxin, Shi Baixue
		DOI: 10.3969/j.issn.0258-8021.2021.04.10
		Mental fatigue refers to a kind of physical and psychological discomfort experienced by people when they carry out cognitive activities of continuous mental attention in a certain period. Cognitive activity depends on the exchange of information between multiple regions of the brain. Phase synchronization, as an important mechanism of information integration between brain regions, builds a bridge of cooperative communication between different regions of the brain. Alpha oscillation is considered to be an active mechanism of neuronal processing inhibition that plays a central role in attention allocation in working memory. In this paper,the adaptive N-back experiment was used to induce mental fatigue. The improved Sternberg paradigm was used as a working memory experiment. EEG signals during working memory before and after mental fatigue were recorded. Phase synchronization analysis and power spectrum analysis were used to analyze the EEG data of α oscillatory working memory before and after mental fatigue, and to study the effect of mental fatigue on brain information integration. Results showed that when the mental fatigue occurred, the phase synchronization of the brain region increased significantly in the whole process of working memory, and there was a significant difference in power spectrum. Among them, the phase synchronization between frontal lobe region and fronto-occipital region increased significantly in all three periods after mental fatigue. During coding period, there were 160 node pairs with significant increase in phase lock value between the frontal lobe regions. During the retention period, there were 222 between frontal lobe and 65 between frontal lobe and occipital parietal lobe. During extraction period, 196 between frontal lobe and 11 between frontal lobe and occipital parietal lobe. During the coding period, there was a significant α oscillationevent—related desynchronization (ERD) (PO8, P=0.048; Oz, P=0.036) in the occipital parietal region, and a tendency of α oscillation event—related synchronization(ERS) appeared in the frontal lobe. During the retention period, there was a significant α oscillatoryERS (Fz, P=0.022) in the frontal lobe and a trend of ERS in the occipital parietal region. During the extraction period, α oscillatory ERS, in frontal region (Fz, P=0.033) and α oscillatory ERD in occipital region (Oz, P=0.045). These studies have shown that the mental fatigue could reduce the ability of information integration in brain regions and reduce the information communication in the brain. The phase synchronization analysis of α oscillation is helpful to explain the mechanism of mental fatigue on information integration in brain regions.
		2021 Vol. 40 (4): 461-468 [Abstract] ( 304 ) HTML (1 KB) PDF (2991 KB) ( 402 )

Reviews

	469	Non-Human Primate Brain Imaging Technology at Ultra-High Field Magnetic Resonance Imaging
		Xu Bin, Gao Yang, Roe Anna Wang, Zhang Xiaotong
		DOI: 10.3969/j.issn.0258-8021.2021.04.11
		Nonhuman primates are important animal models in neuroscience research, which can be compatible with a variety of invasive and non-invasive methods of neural signal detection and neural activity regulation. Combined with the non-invasive method of non-human primate neuroscience research results, it is helpful for the clinical transformation of a large number of basic research results based on animal models. Among them,magnetic resonance (MR) brain imaging technology is the most important non-invasive detection method of brain nerve signal. The research ofmagnetic resonance imaging (MRI) in non-human primates plays an important role in understanding the physiological mechanism ofMRI, the research and development of quantitative physiological detection technology based on MRI, basic research of neuroscience, psychology and clinical pathological mechanism. But non-human primate MRI is facing many technical challenges, including the lack of appropriate MRI scanners and matching imaging hardware, the need for higher imaging resolution, and the compatibility of the needs of diverse animal experiments. Ultra-high field (UHF, field strength>3 T) MRI has the advantages of high signal-to-noise ratio, highblood oxygen level dependent (BOLD) signal detection sensitivity and submillimeter level high-resolution imaging ability, which is widely used in the brain imaging research of non-human primates. In this paper, we reviewed the application of UHF MR brain imaging in non-human primates, the technical challenges faced, and the current technical solutions. Finally, the advantages and limitations of the current methods were summarized and the development trend was proposed.
		2021 Vol. 40 (4): 469-476 [Abstract] ( 352 ) HTML (1 KB) PDF (2703 KB) ( 431 )

	477	Research Progress of Magnetoencephalography in the Functional Mechanism of Bilingual Brain
		Ma Hengfen, Wu Yuntao, Zhao Wen, Jia Liping, Zhou Dandan
		DOI: 10.3969/j.issn.0258-8021.2021.04.12
		It has been proved that bilingual individuals’ regular use of two languages has a broad impact on the linguistic and cognitive functions. However, the impact mechanism of bilingual experience on the brain remains unclear. Magnetoencephalography (MEG) is a noninvasive method to measure the weak brain magnetic field signal, which can more accurately reflect the brain neural activity, therefore is of great significance for the early diagnosis of brain diseases and the frontier research of brain science. This paper reviewed MEG’s application in the study of bilingual brain function mechanism by introducing the development process, analysis methods and software of MEG, mainly including the advantages of bilingualism in brain development, the brain mechanism of switching between two languages, bilingualism and mathematical calculation. At last, a new wearable brain magnetic technology, together with its potential application in the study of bilingual brain function mechanism is introduced.
		2021 Vol. 40 (4): 477-484 [Abstract] ( 325 ) HTML (1 KB) PDF (788 KB) ( 416 )

	485	Polycaprolactone-Based Composite Scaffolds in Bone Tissue Engineering: Research Status and Prospect
		Yang Xiangjun, Chen Junyu, Zhu Zhou, Wan Qianbing
		DOI: 10.3969/j.issn.0258-8021.2021.04.13
		Three-dimensional scaffolds have received considerable attention in the field of bone tissue engineering. Polycaprolactone (PCL) is widely used in the preparation of 3D scaffolds due to its good biocompatibility. However, pure PCL scaffolds have poor hydrophilicity and low biological activity, which limit their application in biomedical field. With the development of bone tissue engineering, a large number of investigations combined PCL with various inorganic substances, metal elements or natural collagen to improve the properties or introduce new properties into the PCL scaffolds. Based on the domestic and foreign literature, in this paper, the selection of materials for PCL bone tissue engineering composite scaffolds was summarized, including inorganic materials, hydrogel materials, metal elements, small molecule drugs and bioactive molecules. The properties and osteogenic effects of the various composite scaffolds were reviewed from five aspects, aiming to provide insights for the research and clinical application of PCL in the bone tissue engineering.
		2021 Vol. 40 (4): 485-492 [Abstract] ( 543 ) HTML (1 KB) PDF (801 KB) ( 832 )

	493	Progress on MRI-Based Molecular Subtyping of Breast Cancer
		Sun Rong, Nie Shengdong, Wei Long
		DOI: 10.3969/j.issn.0258-8021.2021.04.014
		At the level of gene expression, molecular subtyping of breast cancers has important clinical practical value in the application of evaluating the malignant degree of breast cancers and formulating individualized therapeutic programs. In addition to microarray and immunohistochemical (IHC) staining, a new molecular classification method of breast cancers has been provided by radiomics. Herein, the domestic and international developments aboutmagnetic resonance imaging (MRI)-based molecular identification of breast cancers were summarized in this paper. After introducing the principles and characteristics of breast MRI, we reviewed the associated study achievements between molecular subtype information and breast MRI features from the aspect of statistics. Moreover, we highlighted the various prediction algorithms for breast cancer molecular identification from the perspective of machine learning. Finally, the development prospect and existing problems of the technology were pointed out. With aid of higher-order algorithms, it is advisable to further explore MRI imaging features of breast cancers as potential markers, such as morphology, background parenchymal enhancement and statistical texture, dedicating to precision diagnosis and treatment of breast cancers in future.
		2021 Vol. 40 (4): 493-502 [Abstract] ( 399 ) HTML (1 KB) PDF (5890 KB) ( 655 )

	503	Application of Two-Photon Microscopy in the Studies ofCerebral Microcirculation
		LIU Shuangshuang, WANG Jun, YIN Wei, LOU Huifang, SHEN Yi
		DOI: 10.3969/j.issn.0258-8021.2021.04.015
		Two-photon microscopy (TPM) is a new technology with the characteristics of strong penetrability and low photo-toxicity that breaks through the limitations of conventional light microscopy and laser confocal scanning microscopy (LSCM). Based on the two-photon absorption theory and femtosecond pulse technology, TPM provides the special advantages in studying long-term visualization of the microcirculation system in the brain of living animals. It has unique advantages to study the structure of cerebral cortex microcirculation network, blood perfusion and blood oxygen metabolism. In recent years, TPM has been developing iteratively, and its application has been expanding. TPM can be used to detect and photochemical control the blood flow signal in a single sub leptomeningeal microvascular, which opens a valuable microcirculation window for the basic and clinical research of the pathogenesis of ischemic cerebrovascular disease and Alzheimer's disease. Starting with the imaging principle, this review introduced the technological characteristics and progress of TMP, and summarized its application status and prospects from four aspects including quantifying hemodynamic changes, measuring oxygen partial pressure, detecting leukocyte-endothelial cell interaction, and establishing a stroke model, and discussed a few key questions and proposed solutions in the research field as well.
		2021 Vol. 40 (4): 503-512 [Abstract] ( 338 ) HTML (1 KB) PDF (10739 KB) ( 88 )

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