Loading...

Table of Content

    25 July 2024, Volume 44 Issue 4
    Dynamic hybrid multi-attribute group decision making method based on trapezoidal fuzzy number
    LIN Peng, DONG Chunru, LIU Pingping
    2024, 44(4):  337-345.  DOI: 10.3969/j.issn.1000-1565.2024.04.001
    Asbtract ( 180 )   PDF (454KB) ( 83 )  
    References | Related Articles | Metrics
    Based on trapezoidal fuzzy number information environment, the problem of dynamic mixed multi-attribute group decision making is studied. The transformation method of trapezoid fuzzy number and the weighted synthesis method of trapezoid fuzzy number information are given, and the dynamic trapezoid fuzzy weighted geometric operator is defined. The weight of decision-makers is determined by using different expert distance deviation and minimization methods. The attribute weight is determined by using deviation maximization model. The time weight is calculated based on entropy weight method and Orness measure weighting method. On this basis, a dynamic hybrid multi-attribute group decision-making method based on VIKOR method is proposed and the feasibility and effectiveness of the method is verified by a numerical example.
    Algorithm of site selection path integration problem under the background of simultaneous distribution and collection
    CHENG Tao, LI Meixi, LI Jiali
    2024, 44(4):  346-354.  DOI: 10.3969/j.issn.1000-1565.2024.04.002
    Asbtract ( 75 )   PDF (733KB) ( 20 )  
    References | Related Articles | Metrics
    In order to do a good job in the site selection-path planning and design of logistics network under the background of integration of simultaneous distribution and collection, the crossover and mutation processes in the traditional hybrid adaptive genetic algorithm are replaced by the destruction and recombination strategies of the large-scale neighborhood search algorithm, and the optimization design of the algorithm is realized.After analyzing the simulation example, it can be seen that the optimized algorithm can effectively overcome the problems of early maturity and poor stability of the traditional algorithm in the calculation process, improve the probability of obtaining a better solution to a certain extent, and improve customer satisfaction.The effectiveness of the algorithm is tested by using the known benchmark data, and the calculation results show that the indicators of the optimized algorithm perform well, and the calculation results of some data are better than the other three existing algorithms, which is basically consistent with the known optimal solution, and this further verifies the scientificity and effectiveness of the optimization algorithm in this paper.
    Optimizing robust credibility location decision of co-firing power plants considering carbon emission reduction
    CHEN Aixia,CHEN Airu, LIANG Zhiyong
    2024, 44(4):  355-364.  DOI: 10.3969/j.issn.1000-1565.2024.04.003
    Asbtract ( 74 )   PDF (743KB) ( 24 )  
    References | Related Articles | Metrics
    Based on the strategic goal of carbon peak and carbon neutrality, this paper studies the decision-making problem of co-firing power plant location considering carbon emission reduction. Due to the influence of weather conditions, market environments and other external factors, the parameters such as biomass supply capacity and biomass price are uncertain. To address this problem, this paper constructs an ambiguity set of possibility distributions to characterize the uncertain parameters, and then proposes a distributionally robust credibility location optimization model. The original model is reformulated as a computable mixed-integer linear programming model by deriving the equivalent forms of robust credibility objective and robust credibility constraint. Finally, the effectiveness of the proposed method is demonstrated by an example.
    Evaluation the iron supplementation effect of glutamate chelated iron on iron deficiency anemia of mice
    REN Zhiyuan, LIU Xinshuo, LU Shijin, YANG Wenzhi, LI Haiying
    2024, 44(4):  365-372.  DOI: 10.3969/j.issn.1000-1565.2024.04.004
    Asbtract ( 116 )   PDF (1419KB) ( 18 )  
    References | Related Articles | Metrics
    Glutamate chelated ferric iron(Glu-Fe(Ⅲ))and glutamate chelated ferrous iron(Glu-Fe(Ⅱ))were prepared and their iron supplementation effects were evaluated. The iron deficiency anemia(IDA)mouse model was established by combining low iron feed and bloodletting. IDA mice were randomly divided into five groups(n=5): ferrous sulfate group, commercially available iron dextran group, Glu-Fe(Ⅲ)group, Glu-Fe(Ⅱ)group and negative controls group, using normal mice as blank controls. Different iron supplements at equal iron doses were given by intragastric administration for 4 weeks. The changes of hemoglobin(HGB), red blood cells(RBC)and hematocrit(HCT)in blood were monitored by automatic hematology analyzer. The iron status of IDA mice were determined by serum iron(SI), total iron- DOI:10.3969/j.issn.1000-1565.2024.04.004谷氨酸铁螯合物对缺铁性贫血小鼠的补铁效果评价任梽源1,刘新硕1,陆时金2,杨文智1,李海鹰1(1.河北大学 药学院,河北省药物质量分析控制重点实验室,河北 保定 071002;2.中国人民解放军联勤保障部队第九六七医院 药剂科,辽宁 大连 116021)摘 要:为评价自制谷氨酸铁(Glu-Fe(Ⅲ))和谷氨酸亚铁(Glu-Fe(Ⅱ))的补铁效果,采用低铁饲料联合放血法建立缺铁性贫血(IDA)小鼠模型,并将IDA小鼠随机分成硫酸亚铁组、市售右旋糖酐铁组、Glu-Fe(Ⅱ)组和Glu-Fe(Ⅲ)组,分别灌胃给予同等铁剂量的不同补铁剂4周,并以IDA小鼠为阴性对照、正常小鼠为空白对照组进行比较.用全自动血液分析仪监测小鼠血中血红蛋白(HGB)、红细胞(RBC)和血细胞比容(HCT)变化,采用血清铁(SI)、总铁结合力(TIBC)和转铁蛋白(TRF)试剂盒测定小鼠体内铁状况并测定小鼠组织(心、肝、脾和肾)铁含量,利用总超氧化物歧化酶(SOD)、过氧化氢酶(CAT)和丙二醛(MDA)试剂盒检测各种补铁剂对小鼠体内抗氧化的影响.结果显示:IDA小鼠建模成功,Glu-Fe(Ⅲ)和Glu-Fe(Ⅱ)可提高IDA小鼠的HGB、RBC和HCT值,降低TIBC和TRF水平并提升SI含量,而Glu-Fe(Ⅲ)补铁剂效果更佳.此外,相比硫酸亚铁和右旋糖酐铁,饲喂Glu-Fe(Ⅲ)在改善IDA小鼠内脏器官(心、脾和肾)肿大、促进肝恢复、提升肝脾储铁、清除组织活性氧(ROS)且增强抗氧化活性等方面表现更佳.关键词:谷氨酸亚铁螯合物;谷氨酸铁螯合物;缺铁性贫血;补铁剂中图分类号:R973 文献标志码:A 文章编号:1000-1565(2024)04-0365-08Evaluation the iron supplementation effect of glutamate chelated iron on iron deficiency anemia of miceREN Zhiyuan1, LIU Xinshuo1, LU Shijin2, YANG Wenzhi1, LI Haiying1(1. Key Laboratory of Pharmaceutical Quality Control of Hebei Province, College of Pharmaceutical Sciences, Hebei University, Baoding 071002, China;2. Department of Pharmacy, Chinese 967th Hospital of the Joint Logistics Support Force of the Peoples Liberation Army, Dalian 116021, China)Abstract: Glutamate chelated ferric iron(Glu-Fe(Ⅲ))and glutamate chelated ferrous iron(Glu-Fe(Ⅱ))were prepared and their iron supplementation effects were evaluated. The iron deficiency anemia(IDA)mouse model was established by combining low iron feed and bloodletting. IDA mice were randomly divided into five groups(n=5): ferrous sulfate group, commercially available iron dextran group, Glu-Fe(Ⅲ)group, Glu-Fe(Ⅱ)group and negative controls group, using normal mice as blank controls. Different iron supplements at equal iron doses were given by intragastric administration for 4 weeks. The changes of hemoglobin(HGB), red blood cells(RBC)and hematocrit(HCT)in blood were monitored by automatic hematology analyzer. The iron status of IDA mice were determined by serum iron(SI), total iron- 收稿日期:2023-05-04;修回日期:2024-01-19 基金项目:河北省自然科学基金资助项目(C2021201026) 第一作者:任梽源(2000—),男,河北大学在读硕士研究生,主要从事药物制剂方向研究.E-mail:17835420887@163.com 通信作者:李海鹰(1973—),女,河北大学副教授,博士,主要从事药物制剂与质量控制方面研究.E-mail:lihylihy@163.com 第4期任梽源等:谷氨酸铁螯合物对缺铁性贫血小鼠的补铁效果评价河北大学学报(自然科学版) 第44卷binding capacity(TIBC)and transferrin(TRF)kits and the iron contents in mouse tissues(heart, liver, spleen and kidney)were measured. Superoxide dismutase(SOD), catalase(CAT)and malondialdehyde(MDA)kits were used to detect the effects of various iron supplements on antioxidant activity of IDA mice in vivo. The results showed that IDA mouse model was successfully established. Glu-Fe(Ⅲ)and Glu-Fe(Ⅱ)could improve the HGB, RBC and HCT values of IDA mice, meanwhile, reduce TIBC and TRF content and increase SI level for administrated IDA mice, especially Glu-Fe(Ⅲ). Compared with ferrous sulfate and iron dextran, Glu-Fe(Ⅲ)was more effective in reducing the swollen of organs(heart, spleen and kidney), improving liver recovery, enhancing iron storage in liver and spleen, scavenging tissue ROS and enhancing antioxidant activity in IDA mice.
    Quality standard of Qihuang Xingnao Granules
    SHAO Zhenzhen, ZHAO Ziwei, LIU Xiaomeng, LI Lili, ZHANG Yali, HA Jing
    2024, 44(4):  373-381.  DOI: 10.3969/j.issn.1000-1565.2024.04.005
    Asbtract ( 62 )   PDF (923KB) ( 12 )  
    References | Related Articles | Metrics
    The quality standard of Qihuang Xingnao Granule is established to provide reference for its quality control. Thin layer chromatography(TLC)was used to identify Barbary wolfberry, Epimedium, Salvia miltiorrhiza, Rhizoma curculiginis, Angelica sinensis and Polygonum multiflorum in the granules. The content of diphenylvinyl glycosides in the preparation was determined by high performance liquid chromatography(HPLC). The HPLC conditions were: Waters Bridge C18 column(4.6× 250 nm, 5 μL), methanol-water(36∶64), detection wavelength of 320 nm, flow rate of 1.0 mL/min, and a column temperature of 30 ℃. The main TLC spots of Barbary wolfberry, Epimedium, Salvia miltiorrhiza, Rhizoma curculiginis, Angelica sinensis and Polygonum multiflorum were clear, the separation was good, and the negative control had no interference. The stilbenside was linearly related in the range of 0.195 6 to 2.934 0 μg(R2=0.999 5, n=6), the RSD of precision, repeatability and stability was less than 2.0%, the sample recovery rate ranged from 96.38% to 100.19%, and the RSD was 1.48%. The quality standard project is set reasonably, convenient and accurate, and can be used for the quality control of Qihuang Xingnao Granules.
    Design and synthesis of quinoline-triazine derivatives and in vitro butyrylcholinesterase inhibitory activity
    DONG Change, LI Rui, XUE Xuanyi, ZHANG Zhaoyuan, SHI Dahua
    2024, 44(4):  382-389.  DOI: 10.3969/j.issn.1000-1565.2024.04.006
    Asbtract ( 52 )   PDF (808KB) ( 14 )  
    References | Related Articles | Metrics
    Using 2-(chloromethyl)quinoline hydrochloride as raw material, quinoline-triazine derivatives(5a-5f)with novel structures were obtained by oxidation and nucleophilic substitution reaction. The structure and design of all compounds were confirmed to be consistent using characterization methods such as IR, 1H NMR, 13C NMR, and HRMS. Ellman method was used to test the activity of the target compounds against butyrylcholinesterase in vitro. The results showed that all compounds inhibition rate were more than 50% at 80 μmol/L. Compound 5e[IC50=(1.71±0.04)μmol/L] had the best activity and was stronger than the positive control donepezil [IC50=(12.69±0.07)μmol/L]. Further studies on the molecular docking of 5e and BuChE showed that the quinoline ring and sulfur atom in the structure could interact with the residues Ser198, His438, Trp82 and Thr120 of the enzyme active site through hydrogen bonding. Thus, compound 5e is the most active structure screened in this study, which provides a new structural direction for Alzheimers disease(AD)research and is of significance for further research.
    Research development of inflammasome in glioma
    TAN Yanli, LI Xiang, LI Zirui, YU Jia
    2024, 44(4):  390-398.  DOI: 10.3969/j.issn.1000-1565.2024.04.007
    Asbtract ( 110 )   PDF (572KB) ( 40 )  
    References | Related Articles | Metrics
    Inflammasome is composed of apoptosis-associated speckle-like protein(ASC), caspase-1 and pattern recognition receptors(PRR). Inflammasome plays an important role in the process of tumorigenesis, including the regulation of tumor biological behavior, pyroptosis and immunity. Targeting inflammasome may provide new ideas for tumor treatment and prognosis improvement. Gliomas are highly malignant brain tumors with poor prognosis in the central nervous system. This article reviews the composition, activation mechanism and the role of inflammasome in gliomas.
    Inhibition of baicalin on macrophage inflammasome activation induced by Listeria monocytogenes
    LIU Wen, ZHANG Enhua, CHEN Yingying, ZHANG Zonghao, ZHANG Weiwei, LI Wenyan
    2024, 44(4):  399-405.  DOI: 10.3969/j.issn.1000-1565.2024.04.008
    Asbtract ( 64 )   PDF (1007KB) ( 17 )  
    References | Related Articles | Metrics
    In order to study the effect of baicalin on the inflammasome of mouse macrophages activated by Listeria monocytogenes(LM), the optimal concentration gradient of baicalin was determined by methylthiazolyldiphenyl-tetrazolium bromide(MTT)method. Then macrophages were treated with LM and baicalin at concentrations of 25, 50 and 100 μg/mL, respectively, to detect the expression level of inflammasome-related protein or mRNA and lactate dehydrogenase(LDH)release. The results showed that baicalin reduced IL-1β, caspase-1 p10 protein levels, LDH release and mRNA levels of NLRP3, NLRC4, NLRP10, NOD, AIM2, caspase-1, IL-1β and IL-18 after LM treatment in a dose-dependent manner. Therefore, it is speculated that baicalin can inhibit the release of inflammatory factors and cell death in macrophages by inhibiting the activation of different inflammasome in macrophages induced by LM.
    Physiological response and cold resistance evaluation of four species of Quercus robur to low temperature stress
    JIANG Heng, ZHANG Zhigang, YANG Jianjun, LI Bin, WANG Xingsheng, ZHANG Xiaohong
    2024, 44(4):  406-413.  DOI: 10.3969/j.issn.1000-1565.2024.04.009
    Asbtract ( 66 )   PDF (1169KB) ( 15 )  
    References | Related Articles | Metrics
    In order to explore the response of Quercus robur physiological indicators and the strength of cold resistance under low temperature stress, the common Q. robur, fast-growing Q. robur, split-leaf Q. robur and weeping Q. robur were used as experimental materials. The cell membrane permeability, SOD activity, POD activity and soluble protein at different temperatures were measured, and the fitted Logistic- DOI:10.3969/j.issn.1000-1565.2024.04.0094种夏栎对低温胁迫的生理响应及抗寒性评价姜恒1,2,张志刚3,杨建军1,2,李斌4,王兴胜4,张晓红4(1.新疆大学 生态与环境学院,绿洲生态教育部重点实验室,新疆 乌鲁木齐 830017;2.自然资源部 荒漠-绿洲生态监测与修复工程技术创新中心,新疆 乌鲁木齐 830002;3.新疆林业科学院 造林治沙研究所,新疆林木资源与利用国家林草局重点实验室,新疆 乌鲁木齐 830000;4. 伊犁州林木良种繁育试验中心,新疆 察布查尔 835311)摘 要:为探究低温胁迫下夏栎(Quercus robur)生理指标的响应及抗寒性,以普通夏栎、速生夏栎、裂叶夏栎和垂枝夏栎为实验材料,测定在不同温度下细胞膜通透性、超氧化物歧化酶(SOD)、过氧化物酶(POD)、可溶性蛋白等指标,并利用拟合Logistic方程计算半致死温度,结合隶属函数法对4种夏栎综合评价.结果表明:4种夏栎细胞膜通透性均随着温度降低而增加,SOD和POD活性以及可溶性蛋白、脯氨酸和丙二醛(MDA)含量均逐渐下降;4种夏栎半致死温度为-41.9~-38.2 ℃,其中普通夏栎半致死温度最低,速生夏栎半致死温度最高;4种夏栎抗寒性强弱为普通夏栎>垂枝夏栎>裂叶夏栎>速生夏栎.综上可知,普通夏栎抗寒性最强,在不同地区推广和繁育具有显著优势.关键词:夏栎;抗寒性;隶属函数法;生理指标;半致死温度中图分类号:Q945.79 文献标志码:A 文章编号:1000-1565(2024)04-0406-08Physiological response and cold resistance evaluation of four species of Quercus robur to low temperature stressJIANG Heng1,2, ZHANG Zhigang3, YANG Jianjun1,2, LI Bin4, WANG Xingsheng4, ZHANG Xiaohong4(1. Key Laboratory of Oasis Ecology of Education Ministry, College of Ecology and Environment, Xinjiang University,Urumqi 830017,China; 2.Technology Innovation Center for Ecological Monitoring and Restoration of Desert-Oasis, Ministry of Natural Resources(MNR), Urumqi 830002,China; 3. Key Laboratory of Forest Resources and Utilization in Xinjiang of National Forestry and Grassland Administration, Institute of Afforestation and Desertification Control, Xinjiang Academy of Forestry, Urumqi 830000,China; 4. Yili Prefecture Forest Tree Breeding Experiment Center, Chabchal 835311, China)Abstract: In order to explore the response of Quercus robur physiological indicators and the strength of cold resistance under low temperature stress, the common Q. robur, fast-growing Q. robur, split-leaf Q. robur and weeping Q. robur were used as experimental materials. The cell membrane permeability, SOD activity, POD activity and soluble protein at different temperatures were measured, and the fitted Logistic- 收稿日期:2024-01-09;修回日期:2024-05-08 基金项目:新疆三农骨干人才培养项目(2022SNGGNT082);新疆维吾尔自治区2023年度重大科技专项(2023A02008) 第一作者:姜恒(1998—),男,新疆大学在读硕士研究生,主要从事生态修复研究.E-mail:670551763@qq.com 通信作者:杨建军(1978—),男,新疆大学教授,博士生导师,主要从事干旱区水土保持与荒漠化防治研究.E-mail:yjj@xiu.edu.cn张晓红(1972—),女,伊犁州林木良种繁育试验中心高级工程师,主要从事林木良种选育与推广研究. E-mail:1436787731@qq.com 第4期姜恒等:4种夏栎对低温胁迫的生理响应及抗寒性评价河北大学学报(自然科学版) 第44卷equation was used to calculate the half-lethal temperature. And the function method was used for comprehensive evaluation of the four Q. robur species. The results showed that the cell membrane permeability of the four species of Q. robur showed an increasing trend as the temperature decreased, and the SOD activity, POD activity, soluble protein content, proline content and MDA content showed a decreasing trend as a whole, and the cell membrane permeability was calculated by combining the Logistic equation. The semi-lethal temperature of four species of Q. robur ranged from -41.9 ℃ to -38.2 ℃. The common Q. robur had the highest semi-lethal temperature and the fast-growing Q. robur had the lowest semi-lethal temperature. This is consistent with the evaluation by the membership function method. The four species of Q. robur had strong cold resistance. The cold resistance ranking of the four Q. robur species was common Q. robur> weeping Q. robur>split-leaf Q. robur > fast growing Q. robur. The study show that common Q. robur is the most resistant to cold and has significant advantages for promotion and breeding in different regions.
    Automatic segmentation of tumors combining lung prior and synergetic deep supervision
    WANG Bing, JU Mengyi, YANG Ying, ZHANG Xin, ZHAI Junhai
    2024, 44(4):  414-423.  DOI: 10.3969/j.issn.1000-1565.2024.04.010
    Asbtract ( 61 )   PDF (1237KB) ( 15 )  
    References | Related Articles | Metrics
    There are two challenges in automatic segmentation of complex lung tumors(CLT)on computed tomography(CT)images: 1)The class indistinction between tumors and adjacent tissues; 2)Intra-class inconsistencies within tumors. In order to solve these two challenges, the semantic context prior of the relationship between lung tumor and lung is proposed to be incorporated into the segmentation model, so that the model can learn the semantic context features, and the segmentation of CLT can be reconsidered from a macro perspective. The anatomical prior of lung shape is modeled using information entropy. The proposed novel attention module is embedded in the three-classified U-Net network, so as to guide the training process through domain-specific knowledge. In addition, a boundary enhancement auxiliary- DOI:10.3969/j.issn.1000-1565.2024.04.010结合肺先验与协同深监督的肿瘤自动分割王兵1,2,巨梦仪2,杨颖3,张欣4,翟俊海1,2(1.河北省机器学习与计算智能重点实验室,河北 保定 071002;2.河北大学 数学与信息科学学院,河北 保定 071002;3.河北大学附属医院 放射科,河北 保定 071000;4.河北大学 电子信息工程学院,河北 保定 071002)摘 要:计算机断层扫描图像中复杂肺肿瘤(CLT)的自动分割面临2个挑战:1)肿瘤与邻近组织之间的类间不区分;2)肿瘤内的类内不一致性.为了解决这2个问题,提出将肺肿瘤与肺之间关系的语义上下文先验纳入分割模型中,以便于模型学习到语义上下文特征,并从宏观角度重新思考CLT的分割.利用信息熵对肺形状的解剖先验进行建模.在三分类的U-Net网络中嵌入提出的新型注意模块,从而通过特定领域的知识来指导训练过程.另外,设计了一个可以获得肿瘤边界结构图以及保持肿瘤内部特征一致性的边界增强辅助网络.在此基础上,开发了一个协同深度监督网络框架(CLT-ASegNet),该框架利用混合多尺度语义特征融合进一步提高了模型的判别能力和收敛速度.CLT-ASegNet在CLTCTI分割数据集和Lung16数据集上进行了评估.实验结果表明,所提出的CLT-ASegNet可以有效分割肺肿瘤.关键词:注意力机制; 复杂肺肿瘤分割; 语义上下文先验; 协同深度监督中图分类号:TP391.4 文献标志码:A 文章编号:1000-1565(2024)04-0414-10Automatic segmentation of tumors combining lung prior and synergetic deep supervisionWANG Bing1,2, JU Mengyi2, YANG Ying3, ZHANG Xin4, ZHAI Junhai1,2(1. Hebei Key Laboratory of Machine Learning and Computational Intelligence, Baoding 071002, China;2. College of Mathematics and Information Science,Hebei University, Baoding 071002, China;3. Radiology Department, Hebei University Affiliated Hospital, Baoding 071000, China;4. College of Electronic Information Engineering, Hebei University, Baoding 071002, China)Abstract: There are two challenges in automatic segmentation of complex lung tumors(CLT)on computed tomography(CT)images: 1)The class indistinction between tumors and adjacent tissues; 2)Intra-class inconsistencies within tumors. In order to solve these two challenges, the semantic context prior of the relationship between lung tumor and lung is proposed to be incorporated into the segmentation model, so that the model can learn the semantic context features, and the segmentation of CLT can be reconsidered from a macro perspective. The anatomical prior of lung shape is modeled using information entropy. The proposed novel attention module is embedded in the three-classified U-Net network, so as to guide the training process through domain-specific knowledge. In addition, a boundary enhancement auxiliary- 收稿日期:2023-12-06;修回日期:2024-03-19 基金项目:河北省自然科学基金资助项目(F2021201020);河北省自然科学基金青年科学基金资助项目(F2024502006) 第一作者: 王兵(1967—),女,河北大学教授,主要研究方向为机器学习、计算机视觉等.E-mail:wangbing@hbu.edu.cn 通信作者:张欣(1966—),男,河北大学教授,博士,主要研究方向为机器学习、图像处理等.E-mail:zhangxin@hbu.edu.cn第4期王兵等:结合肺先验与协同深监督的肿瘤自动分割河北大学学报(自然科学版) 第44卷network was designed to obtain tumor boundary structure and maintain the consistency of tumor internal features. On this basis, a collaborative deep supervision network framework(CLT-ASegNet)was developed, which further improved the discriminant ability and convergence speed of the model by using hybrid multi-scale semantic feature fusion. CLT-ASegNet was evaluated on CLTCTI segmentation datasets and Lung16 datasets. The experimental results show that the proposed CLT-ASegNet can effectively segment lung tumors.
    A real-time multi-class detection method for colonoscopy polyps based on improved YOLOv5s
    XUE Linyan, LI Xuanang, QI Chaoyi, CAO Jie, ZHANG Ying, AI Shangpu, YANG Kun
    2024, 44(4):  424-432.  DOI: 10.3969/j.issn.1000-1565.2024.04.011
    Asbtract ( 144 )   PDF (1372KB) ( 24 )  
    References | Related Articles | Metrics
    To facilitate rapid identification and detection of colorectal polyps during colonoscopy procedures, a real-time multi-class detection model for colonoscopic polyps based on modified YOLOv5s is proposed. This model utilizes ConvNeXt as thebackbone network and incorporates the SimAM attention mechanism to improve detection performance. Additionally, a slim-neck module based on GSConv is employed in the neck network to reduce network parameters. For model training and testing, a colorectal polyp dataset containing 1 676 images annotated by professional doctors was constructed. The proposed model achieves a mean Average Precision(mAP@0.5)of 83.0% on the test set, which is an improvement- DOI:10.3969/j.issn.1000-1565.2024.04.011基于改进YOLOv5s的肠镜息肉多分类实时检测方法薛林雁1,2,3,李轩昂1,齐晁仪1,曹杰1,张颖1,艾尚璞1,杨昆1,2,3(1.河北大学 质量技术监督学院,河北 保定 071002;2.计量仪器与系统国家地方联合工程研究中心,河北 保定 071002;3.河北省新能源汽车动力系统轻量化技术创新中心,河北 保定 071002)摘 要:为了在肠镜检查过程中对结直肠息肉进行快速鉴别检测,提出一种基于改进YOLOv5s的肠镜息肉多分类实时检测模型.该模型采用ConvNeXt作为主干网络, 融入SimAM注意力机制提升检测性能,同时在颈部网络中使用基于GSConv的slim-neck模块减少网络参数.为了对模型进行训练和测试,构建了包含1 676张息肉图像并由专业医生标注的结直肠息肉数据集.提出的模型在测试集上的平均精度均值(mAP@0.5)为83.0%,相较于改进前提升8.4%,检测速度达到120帧/s. 此外,模型在边缘侧部署检测速度超过25帧/s.结果表明,改进的YOLOv5s满足临床结肠镜检查对实时性与准确性的要求.关键词:息肉;腺瘤;检测;YOLOv5s;实时性中图分类号:TP391.7 文献标志码:A 文章编号:1000-1565(2024)04-0424-09A real-time multi-class detection method for colonoscopy polyps based on improved YOLOv5sXUE Linyan1,2,3, LI Xuanang1, QI Chaoyi1, CAO Jie1, ZHANG Ying1, AI Shangpu1, YANG Kun1,2,3(1. College of Quality and Technical Supervision, Hebei University, Baoding 071002, China; 2. National & Local Joint Engineering Research Center of Metrology Instrument and System, Baoding 071002, China; 3. New Energy Vehicle Power System Lightweight Technology Innovation Center of Hebei Province, Baoding 071002, China)Abstract: To facilitate rapid identification and detection of colorectal polyps during colonoscopy procedures, a real-time multi-class detection model for colonoscopic polyps based on modified YOLOv5s is proposed. This model utilizes ConvNeXt as thebackbone network and incorporates the SimAM attention mechanism to improve detection performance. Additionally, a slim-neck module based on GSConv is employed in the neck network to reduce network parameters. For model training and testing, a colorectal polyp dataset containing 1 676 images annotated by professional doctors was constructed. The proposed model achieves a mean Average Precision(mAP@0.5)of 83.0% on the test set, which is an improvement- 收稿日期:2024-03-16;修回日期:2024-05-06 基金项目:河北省自然科学基金资助项目(F2023201069);保定市创新能力提升专项项目(2394G027);河北大学研究生创新项目(HBU2024BS021;HBU2024SS011);河北大学科研创新团队项目(IT2023B07);大学生创新创业训练计划创新训练项目(DC2024376;DC2024381) 第一作者:薛林雁(1981—),女,河北大学副教授,主要从事生物医学图像处理方向研究.E-mail: lyxue@hbu.edu.cn 通信作者:杨昆(1976—),男,河北大学教授,博士生导师,主要从事生物医学图像处理方向研究.E-mail: yangkun@hbu.edu.cn第4期薛林雁等:基于改进YOLOv5s的肠镜息肉多分类实时检测方法河北大学学报(自然科学版) 第44卷of 8.4% compared to the model before modification, with a detection speed of 120 frames per second. Moreover,the model exhibits a detection speed exceeding 25 frames per second when deployed on edge devices. The results demonstrate that the improved YOLOv5s meets the clinical requirementsfor real-time and accurate colonoscopy examinations.
    Gastric cancer pathological image diagnosis system based on ResNet and UNet
    ZHANG Wenyue, JIA Ziyan, LI Qing, ZHANG Dachuan, PAN Lingjiao, SHEN Dawei
    2024, 44(4):  433-440.  DOI: 10.3969/j.issn.1000-1565.2024.04.012
    Asbtract ( 94 )   PDF (1234KB) ( 17 )  
    References | Related Articles | Metrics
    Considering that manual identification and diagnosis of gastric cancer pathological images may cause missed detection and in order to make diagnosis more accurate, a pathological image diagnosis system based on ResNet and UNet is proposed, aiming to classify, segment and output the diagnosis results of pathological images. The ResNet model is used to classify gastric cancer pathological images with and without cancer. The UNet model is improved, and the improved model adds a convolutional block attention module before each down-sampling and up-sampling to enhance the models attention to cancerous areas. The residual module is used to replace the two convolutions in the encoding part to improve feature utilization; and the Inception module is used to replace the two convolutions in the up-sampling- DOI:10.3969/j.issn.1000-1565.2024.04.012基于ResNet和UNet的胃癌病理图像诊断系统张文悦1,贾子彦1,李青2,张大川2,潘玲佼1,沈大伟1(1.江苏理工学院 电气信息工程学院,江苏 常州 213001;2.常州市第一人民医院 病理科,江苏 常州 213004)摘 要:考虑到人工对胃癌病理图像的判别和诊断可能存在漏检的问题,为使诊断更加准确,提出一种基于ResNet和UNet的病理图像诊断系统,旨在实现对病理图像的分类、分割以及输出诊断结果.采用ResNet模型对胃癌病理图像进行有癌和无癌的分类.对UNet模型进行改进,改进后的模型在每个下采样和上采样之前加入卷积注意力模块,以增强模型对癌变区域的关注.使用残差模块替代编码部分的2次卷积,来提高特征的利用率;利用Inception模块来替代解码部分上采样中的2个卷积,从而扩充其宽度并获取不同尺度的特征.将分类与分割结果综合考虑,获取最终的胃癌病理图像的诊断结果.实验结果表明,该系统可以有效地诊断胃癌病理图像中是否存在癌变.关键词:病理图像;图像分类;UNet;图像分割;胃癌诊断中图分类号:TP391.4 文献标志码:A 文章编号:1000-1565(2024)04-0433-08Gastric cancer pathological image diagnosis system based on ResNet and UNetZHANG Wenyue1, JIA Ziyan1, LI Qing2, ZHANG Dachuan2, PAN Lingjiao1, SHEN Dawei1(1. School of Electrical and Information Engineering, Jiangsu University of Technology, Changzhou 213001, China; 2. Department of Pathology, Changzhou First Peoples Hospital, Changzhou 213004, China)Abstract: Considering that manual identification and diagnosis of gastric cancer pathological images may cause missed detection and in order to make diagnosis more accurate, a pathological image diagnosis system based on ResNet and UNet is proposed, aiming to classify, segment and output the diagnosis results of pathological images. The ResNet model is used to classify gastric cancer pathological images with and without cancer. The UNet model is improved, and the improved model adds a convolutional block attention module before each down-sampling and up-sampling to enhance the models attention to cancerous areas. The residual module is used to replace the two convolutions in the encoding part to improve feature utilization; and the Inception module is used to replace the two convolutions in the up-sampling- 收稿日期:2023-10-27;修回日期:2024-04-25 基金项目:国家自然科学基金资助项目(62001196); 江苏省“333高层次人才培养工程”项目(2022-3-4-107); 常州市科技计划项目(CM20223015); 常州应用基础研究项目(CJ20220064;CJ20220059) 第一作者:张文悦(1997—),女,江苏理工学院在读硕士研究生,主要从事计算机医学图像处理方向研究.E-mail:zhangwenyue97wren@163.com 通信作者:贾子彦(1981—),男,江苏理工学院副教授,主要从事可见光通信、机器视觉、5G方向研究.E-mail:jiaziyan@jsut.edu.cn第4期张文悦等:基于ResNet和UNet的胃癌病理图像诊断系统河北大学学报(自然科学版) 第44卷of the decoding part, thereby expanding its width to obtain features of different scales. The classification and segmentation results are comprehensively considered to obtain the final diagnostic results of gastric cancer pathological images. Experimental results show that this system can effectively diagnose the presence of cancer in gastric cancer pathological images.
    Semi-supervised skin cancer diagnosis based on self-feedback threshold learning
    HAN Shuo, YUAN Weicheng, DU Zeyu
    2024, 44(4):  441-448.  DOI: 10.3969/j.issn.1000-1565.2024.04.013
    Asbtract ( 110 )   PDF (963KB) ( 24 )  
    References | Related Articles | Metrics
    To address the challenges associated with the need for a large amount of annotated data in supervised skin cancer diagnosis models, such as the high cost, time consumption, and fatigue experienced by medical experts during annotation, this study proposes a semi-supervised skin cancer diagnosis method based on Self-Feedback Threshold Learning(SFTL). Building upon the ResNet network pre-trained with labeled data, a global and local class pseudo-label self-feedback threshold learning mechanism is introduced to dynamically select unlabeled samples with ResNet prediction probabilities exceeding the self-feedback threshold. Unsupervised threshold learning loss and classification cross-entropy loss are incorporated for model training, thereby deeply mining the diagnostic information from unlabeled data when labeled samples are scarce and significantly reducing the misdiagnosis rate in unlabeled skin lesion images. Experimental validation was conducted using the publicly available HAM10000 skin lesion dataset, achieving an accuracy of 0.8229 and an F1 score of 0.7651 with only 50% of the data labeled. The results demonstrate that the proposed SFTL model effectively addresses the skin cancer diagnosis task in semi-supervised scenarios and outperforms other compared methods in terms of classification performance.- DOI:10.3969/j.issn.1000-1565.2024.04.013基于自反馈阈值学习的半监督皮肤癌诊断模型韩硕1,袁伟珵1,杜泽宇2(1.河北医科大学 基础医学院,河北 石家庄 050017;2.曼彻斯特大学 健康科学学院,英格兰 曼彻斯特 M139PL)摘 要:为解决监督学习皮肤癌诊断模型的训练需要大量数据标注,且医学专家标注工作成本高、耗时长、易疲劳等问题,提出了一种基于自反馈阈值学习(Self-Feedback Threshold Learning, SFTL)的半监督皮肤癌诊断方法.在标注数据预训练的ResNet网络基础上,引入全局和局部类别间伪标签自反馈阈值学习机制动态筛选ResNet预测概率大于自反馈阈值的无标记样本,引入无监督阈值学习损失和分类交叉熵损失进行模型训练,在标记样本稀缺的情况下深入挖掘无标记数据的鉴别诊断信息,显著降低模型在无标记皮肤病变图像中的误判率.选取公开数据集HAM10000的皮肤病变图像展开实验验证,在仅需50%标记数据下实现了0.822 9的准确率和0.765 1的F1分数,证明所提出的SFTL模型在半监督场景下可有效解决皮肤癌诊断任务,相比其他同类方法具有更好的分类性能.关键词:半监督皮肤癌诊断;自反馈阈值学习;卷积神经网络;半监督学习中图分类号:U492.2;TP301.6 文献标志码:A 文章编号:1000-1565(2024)04-0441-08Semi-supervised skin cancer diagnosis based on self-feedback threshold learning HAN Shuo1, YUAN Weicheng1, DU Zeyu2(1.College of Basic Medicine, Hebei Medical University, Shijiazhuang 050017, China;2.School of Health Science, University of Manchester, Manchester M139PL, UK)Abstract: To address the challenges associated with the need for a large amount of annotated data in supervised skin cancer diagnosis models, such as the high cost, time consumption, and fatigue experienced by medical experts during annotation, this study proposes a semi-supervised skin cancer diagnosis method based on Self-Feedback Threshold Learning(SFTL). Building upon the ResNet network pre-trained with labeled data, a global and local class pseudo-label self-feedback threshold learning mechanism is introduced to dynamically select unlabeled samples with ResNet prediction probabilities exceeding the self-feedback threshold. Unsupervised threshold learning loss and classification cross-entropy loss are incorporated for model training, thereby deeply mining the diagnostic information from unlabeled data when labeled samples are scarce and significantly reducing the misdiagnosis rate in unlabeled skin lesion images. Experimental validation was conducted using the publicly available HAM10000 skin lesion dataset, achieving an accuracy of 0.8229 and an F1 score of 0.7651 with only 50% of the data labeled. The results demonstrate that the proposed SFTL model effectively addresses the skin cancer diagnosis task in semi-supervised scenarios and outperforms other compared methods in terms of classification performance.- 收稿日期:2024-01-08;修回日期:2024-05-23 基金项目:河北省自然科学基金资助项目(H2019206316) 第一作者:韩硕(1974—),男,河北医科大学讲师,博士,主要从事肿瘤及神经退行性疾病相关的基础研究及人工智能医疗应用.E-mail:hanshuo@hebmu.edu.cn第4期韩硕等:基于自反馈阈值学习的半监督皮肤癌诊断模型河北大学学报(自然科学版) 第44卷