Force/Moment Isotropy of 8/4-4 Parallel Six-Axis Force Sensor Based on Performance Atlases

A six-axis force sensor with parallel 8/4-4 structure is introduced and its measurement principle is analyzed.Based on condition numbers of Jacobian matrix spectral norm of the sensor,the relationship between the force and moment isotropy and some structural parameters is deduced.Orthogonal test methods are used to determine the degree of primary and secondary factors that have significant effect on sensor characteristics.Furthermore,the relationship between each performance index and the structural parameters of the sensor is analyzed by the method of the atlas,which lays a foundation for structural optimization design of the force sensor.

Electronic geothermal atlases of Asian Russia

Generalized geothermal data was used to produce two electronic atlases for Asian Russia,Geothermal Atlas of Siberia（GAS）（1995-2000） and Geothermal Atlas for Siberia and Russian Far East（GASRFE）（2009-2012）.The atlases include heat flow maps,temperatures at depths of 0.5,1,2,3,5 km and lower boundary of permafrost.Quantitative values of parameters are presented as isolines（GAS） and symbols（GASRFE）.GAS website is located at the Trofimuk Institute（www.ipgg.sbras.ru/ru/institute/structure/geophysics/natural-fields）.GASRFE provides the most complete geothermal data on Asian Russia,which has been growing for the last 50 years,and is published on the Internet at http：//maps.nrcgit.ru/geoterm.In this atlas,data about the depth of permafrost lower boundary（ ＂zero＂ isotherm） are presented for the first time.

Optimal Design of a Main Driving Mechanism for Servo Punch Press Based on Performance Atlases

The servomotor drive turret punch press is attracting more attentions and being developed more intensively due to the advantages of high speed,high accuracy,high flexibility,high productivity,low noise,cleaning and energy saving.To effectively improve the performance and lower the cost,it is necessary to develop new mechanisms and establish corresponding optimal design method with uniform performance indices.A new patented main driving mechanism and a new optimal design method are proposed.In the optimal design,the performance indices,i.e.,the local motion/force transmission indices ITI,OTI,good transmission workspace good transmission workspace(GTW) and the global transmission indices GTIs are defined.The non-dimensional normalization method is used to get all feasible solutions in dimensional synthesis.Thereafter,the performance atlases,which can present all possible design solutions,are depicted.As a result,the feasible solution of the mechanism with good motion/force transmission performance is obtained.And the solution can be flexibly adjusted by designer according to the practical design requirements.The proposed mechanism is original,and the presented design method provides a feasible solution to the optimal design of the main driving mechanism for servo punch press.

基于ICESat-2和Sentinel-2A数据的森林蓄积量反演

森林蓄积量是林业调查的重要指标,在衡量森林健康状况和评价森林固碳能力等方面发挥重要作用,协同主被动遥感是当前反演大区域森林蓄积量的主要手段。以云南香格里拉森林为研究区,分别提取ICESat-2/ATLAS和Sentinel-2A影像的特征变量,并通过相关性分析和共线性诊断方法筛选特征变量,构建Sentinel-2A变量集和ICESat-2/ATLAS变量集,以及二者联合的变量集,然后结合样地实测数据与3个特征变量集,采用逐步线性回归和随机森林方法分别建立线性和非线性回归模型,反演森林蓄积量,并对结果进行精度验证及对比分析。研究结果表明:对3个变量集,随机森林方法精度均优于逐步线性回归;ICESat-2/ATLAS变量集在2种回归方法下的反演精度均高于Sentinel-2A变量集;联合Sentinel-2A和ICESat-2/ATLAS变量集,随机森林方法的反演精度最高,其R 2,RMSE和rRMSE分别为0.7034,84.78 m^(3)/hm^(2)和36.46%。整体来说,与Sentinel-2A数据相比,基于ICESat-2/ATLAS数据及其与多源数据联合的反演模型均可以提高森林蓄积量反演精度和模型稳定性。

基于星载光子计数雷达数据的森林郁闭度估测模型优化

【目的】旨在评估星载光子计数雷达数据估测森林郁闭度(Forest Canopy Closure,FCC)的潜力,为优化森林管理规划提出一种高效率、低成本估测区域尺度FCC的新技术方法。【方法】研究以星载激光雷达ICESat-2/ATLAS光子点云数据为信息源,以滇西北生态脆弱区香格里拉为研究区,结合54块地面实测样地数据,在前期对点云数据进行去噪、分类等预处理的基础上,对研究区74808个有林地光斑冠层参数进行提取(共计59个),使用支持向量机的递归特征消除算法(SVM-RFE)优选特征变量,采用普通克里格(OK)插值出区域尺度特征变量的空间分布,基于贝叶斯优化(BO)算法改进后的随机森林(RF)、K-最近邻值法(KNN)、梯度回归(GBRT)模型建模,以决定系数(R^(2))、均方根误差(RMSE)、总体预测精度(P)、残差平方和(R_(SS))和相对均方根误差(R_(RMSE))作为模型评价指标,以此构建研究区FCC估测模型。【结果】1)由ICESat-2/ATLAS提取的光斑冠层参数经SVM-RFE优选后,6个(asr、n_toc_photons、n_ca_photons、h_min_canopy、toc_roughness、photon_rate_can)冠层参数的平均交叉验证精度高为0.60,可作为OK插值变量;2)以优选的冠层参数作为OK插值变量拟合最佳半方差函数,所有变量的块金效应(SR<25%)较弱,具有强烈的空间自相关性,除asr变量的最佳拟合模型为球状模型(R^(2)=0.689,R_(SS)=2.05×10^(-6),R_(RMSE)=0.1602)外,其余5个变量的最佳拟合模型均为指数模型(R^(2),0.71~0.93;R_(SS),2.34×10^(-9)~1.54×10^(-4);R_(RMSE),0.0239~0.1886);3)在BO-RF、BO-GBRT、BO-KNN郁闭度估测模型中,以BO-RF模型综合建模精度最高(R^(2)=0.73,RMSE=0.09、P=80.13%),可作为研究区FCC遥感估测模型;4)使用BO-RF模型估测的研究区FCC进行空间制图,均值为0.53,主要分布在0.3~0.6之间,占比77.44%。FCC高值区域总体呈现出由东南向北延伸分布的趋势,与研究区森林资源分布情况基本一致。【结论】该方法可为优化森林资源管理提供一种技术与方法上的参考。

Mapping lineaments using Landsat 8 OLI and SRTM data;a case study of the eastern part of the Ouarzazate Basin,Morocco

The integration of remote sensing and geographic information system(GIS)was employed in this study to delineate the structural lineaments within the eastern section of the Ouarzazate Basin,situated between the southern front of the Central High Atlas and the northern slopes of the Eastern Anti-Atlas(also known as the Saghro Massif).To achieve this objective,Landsat 8 Operational Land Imager(OLI)and Shuttle Radar Topography Mission(SRTM)data were used.Principal Component Analysis(PCA)was computed and a directional filter was applied to the first PCA and the panchromatic band(B8).Additionally,shading was applied to the SRTM data in four directions;N0°,N45°,N90°,N135°.After removing of the non-geological linear structures,the results obtained,using the automatic extraction method,allowed us to produce a synthetic map that included 1251 lineaments with an average length of 1331 m and was dominated by NE-SW,ENE-WSW and E-W directions,respectively.However,the high lineament density is clearly noted in the Anti-Atlas(Saghro Massif)and at the level of the northern part,extending from the Ait Ibrirne to Arg-Ali Oubourk villages.High lineament density can always be found around the major faults affecting this area.The data collected during the field investigations and from geological maps show that the major direction of the faults and structural accidents range mostly between N45°,N70°and N75°.The correlation of remote sensing results with those collected in the field shows a similarity and coincidence with each other.From these results,it is possible to consider the automatic extraction method as a supplementary kind that can serve classical geology by quickly enriching it with additional data.As shown in this work,this method provides more information when applied in arid areas where the fields are well outcropped.

Comprehensive integration of single-cell transcriptomic data illuminates the regulatory network architecture of plant cell fate specification

Plant morphogenesis relies on precise gene expression programs at the proper time and position which is orchestrated by transcription factors(TFs)in intricate regulatory networks in a cell-type specific manner.Here we introduced a comprehensive single-cell transcriptomic atlas of Arabidopsis seedlings.This atlas is the result of meticulous integration of 63 previously published scRNA-seq datasets,addressing batch effects and conserving biological variance.This integration spans a broad spectrum of tissues,including both below-and above-ground parts.Utilizing a rigorous approach for cell type annotation,we identified 47 distinct cell types or states,largely expanding our current view of plant cell compositions.We systematically constructed cell-type specific gene regulatory networks and uncovered key regulators that act in a coordinated manner to control cell-type specific gene expression.Taken together,our study not only offers extensive plant cell atlas exploration that serves as a valuable resource,but also provides molecular insights into gene-regulatory programs that varies from different cell types.

Identification and validation of a new prognostic signature based on cancer-associated fibroblast-driven genes in breast cancer

BACKGROUND Breast cancer(BC),a leading malignant disease,affects women all over the world.Cancer associated fibroblasts(CAFs)stimulate epithelial-mesenchymal transition,and induce chemoresistance and immunosuppression.AIM To establish a CAFs-associated prognostic signature to improve BC patient out-come estimation.METHODS We retrieved the transcript profile and clinical data of 1072 BC samples from The Cancer Genome Atlas(TCGA)databases,and 3661 BC samples from the The Gene Expression Omnibus.CAFs and immune cell infiltrations were quantified using CIBERSORT algorithm.CAF-associated gene identification was done by weighted gene co-expression network analysis.A CAF risk signature was established via univariate,least absolute shrinkage and selection operator regression,and mul-tivariate Cox regression analyses.The receiver operating characteristic(ROC)and Kaplan-Meier curves were employed to evaluate the predictability of the model.Subsequently,a nomogram was developed with the risk score and patient clinical signature.Using Spearman's correlations analysis,the relationship between CAF risk score and gene set enrichment scores were examined.Patient samples were collected to validate gene expression by quantitative real-time polymerase chain reaction(qRT-PCR).RESULTS Employing an 8-gene(IL18,MYD88,GLIPR1,TNN,BHLHE41,DNAJB5,FKBP14,and XG)signature,we attemp-ted to estimate BC patient prognosis.Based on our analysis,high-risk patients exhibited worse outcomes than low-risk patients.Multivariate analysis revealed the risk score as an independent indicator of BC patient prognosis.ROC analysis exhibited satisfactory nomogram predictability.The area under the curve showed 0.805 at 3 years,and 0.801 at 5 years in the TCGA cohort.We also demonstrated that a reduced CAF risk score was strongly associated with enhanced chemotherapeutic outcomes.CAF risk score was significantly correlated with most hallmark gene sets.Finally,the prognostic signature were further validated by qRT-PCR.CONCLUSION We introduced a newly-discovered CAFs-associated gene signature,which can be employed to estimate BC patient outcomes conveniently and accurately.

Systematic analysis of DNA polymerases as therapeutic targets in pan-cancers

Introduction:DNA polymerases are crucial for maintaining genome stability and influencing tumorigenesis.However,the clinical implications of DNA polymerases in tumorigenesis and their potential as anti-cancer therapy targets are not well understood.Methods:We conducted a systematic analysis using TCGA Pan-Cancer Atlas data and Gene Set Cancer Analysis results to examine the expression profiles of 15 DNA polymerases(POLYs)and their clinical correlations.We also evaluated the prognostic value of POLYs by analyzing their expression levels in relation to overall survival time(OS)using Kaplan-Meier survival curves.Additionally,we investigated the correlations between POLY expression and immune cells,DNA damage repair(DDR)pathways,and ubiquitination.Drug sensitivity analysis was performed to assess the relationship between POLY expression and drug response.Results:Our analysis revealed that 14 out of 15 POLYs exhibited significantly distinct expression patterns between tumor and normal samples across most cancer types,except for DNA nucleotidylexotransferase(DNTT).Specifically,POLD1 and POLE showed elevated expression in almost all cancers,while POLQ exhibited high expression levels in all cancer types.Some POLYs showed heightened expression in specific cancer subtypes,while others exhibited low expression.Kaplan-Meier survival curves demonstrated significant prognostic value of POLYs in multiple cancers,including PAAD,KIRC,and ACC.Cox analysis further validated these findings.Alteration patterns of POLYs varied significantly among different cancer types and were associated with poorer survival outcomes.Significant correlations were observed between the expression of POLY members and immune cells,DDR pathways,and ubiquitination.Drug sensitivity analysis indicated an inverse relationship between POLY expression and drug response.Conclusion:Our comprehensive study highlights the significant role of POLYs in cancer development and identifies them as promising prognostic and immunological biomarkers for various cancer types.Additionally,targeting POLYs therapeutically holds promise for tumor immunotherapy.

基于生物信息技术分析GINS1基因在肺癌中的表达及意义

背景与目的 肺癌是全球最常见的恶性肿瘤,大多数患者初次诊断时已发生转移,因此寻找肺癌新的诊断标志物并探索其在肺癌发生中的作用具有重要价值。基于生物信息数据探讨分析肺癌中GINS1基因与肺癌预后的相关性具有重要意义。方法 检索Oncomine、EGPIA、TCGA、Human protein atlas等基因数据库,分析GINS1基因在肺癌中的表达差异。应用蛋白质免疫印记法检测肺癌及癌旁组织中GINS1蛋白的表达水平,应用Kaplan-Meier进行患者生存分析,并利用String-DB数据库分析GINS1蛋白相互作用网络。结果 对Oncomine、GEPIA和The Human Protein Atlas数据库中肺癌与癌旁组织蛋白表达数据进行差异性分析,发现GINS1蛋白在肺癌组织中显著高表达,特别以肺腺癌、肺鳞癌中表达明显增加;通过生存分析发现高表达GINS1肺腺癌患者生存期明显低于低表达者,高表达患者预后更差;利用String-DB数据库发现与GINS1蛋白关联最为密切的为GINS蛋白家族(GINS2、GINS3、GINS4),其次为CDC45、MCM蛋白家族(MCM2、MCM3、MCM4、MCM5、MCM6、MCM7),富集分析发现互作蛋白主要参与DNA复制和细胞周期。结论GINS1基因在肺癌组织中显著高表达,与患者预后相关,其有可能成为肺癌诊断及药物治疗的新靶点。

磁共振图像引导下前列腺癌在线自适应放疗自动勾画研究

目的:探讨深度学习(deep learning,DL)和图谱库(Atlas)联合的MRI引导下在线自适应放疗自动勾画方案的勾画效果。方法:选取2020年1月至2021年9月在某院进行MRI引导下在线自适应放疗的15例前列腺癌患者,以随机抽样法分成训练集(12例)和测试集(3例)。基于训练集分别建立DL勾画和Atlas勾画的临床靶区(clinical target volume,CTV)和危及器官(organs at risk,OAR)模型,基于测试集分别对2种勾画模型的结果进行修改并记录修改时长。对比2种勾画方法的勾画精度[戴斯相似性系数(Dice similarity coefficient,DSC)、豪斯多夫距离(Hausdorff distance,HD)和平均一致性距离(mean distance to agreement,MDA)]和勾画效率。结合2种方法勾画的优势及特点,建立DL+Atlas联合自动勾画方案,对比DL+Atlas联合自动勾画方案与单独勾画方案的勾画用时。结果:勾画精度对比结果显示,与DL勾画的CTV和OAR模型相比,Atlas勾画的CTV各项指标均优于DL勾画,差异有统计学意义(P<0.05);在膀胱和直肠方面,Atlas勾画的DSC和MDA劣于DL勾画,差异有统计学意义(P<0.05)。勾画效率对比结果显示,医生以DL勾画为基准修改CTV和OAR的平均时间为9.4 min,以Atlas勾画为基准修改CTV和OAR的平均时间为12 min。DL+Atlas联合自动勾画方案所需平均时间为8 min,优于DL及Atlas单独勾画方案。结论:DL+Atlas联合的MRI引导下在线自适应放疗自动勾画方案勾画用时少,具有较高的准确性,可提升勾画效率。

《Atlas of Uveitis:Diagnosis and Treatment》一书出版

由杨培增教授独自完成的大型英文葡萄膜炎专著《Atlas of Uveitis:Diagnosis and Treatment》已于2020年12月由Springer出版社和人民卫生出版社联合出版。此书是目前国际上单人完成的最大葡萄膜炎专著,共860页、3000余张患者图片,约合中文170万字。此书分为Overview、Diagnosis of Uveitis、Treatment of Uveitis.Uveitisentities,Scleritis and Episcleritis四大部分,详细介绍了葡萄膜炎类疾病的疾病谱,常见类型的临床特征、进展和致盲规律,疾病诊疗的思想、原则、策略,并使用大量精美图片展现了常见葡萄膜炎类型的表现谱,即从发病到疾病后期,从最细微的异常到最严重的改变,从罕见的表现到典型的特征,展示出葡萄膜炎的全貌和细节,既可使临床医师快速掌握疾病要点,又不至于遗漏细微之处。这也是中国眼科医师首次向国际介绍葡萄膜炎诊疗的中国标准、方案和经验。

基于ICESat-2雷达数据的邢台北澧河局部水深测量

水深是水文水资源研究的重要参数之一,对径流量估算、水利工程建设、旱涝灾害防御具有重要意义。为提升水深测量时效性,文章通过对星载雷达反演了河北的北澧河局部水深。结果表明:经滤波处理后的ICESat-2光斑群准确区分了水面与水下地形特征,通过差分处理能有效提取水深信息。研究构建的基于异源遥感数据反演水深方法,可为类似问题的解决提供借鉴。

Atlas的谢幕

正当全球人工智能的研究和应用如火如荼之时,4月16日,波士顿动力发布了一段约3分钟的视频,宣布研发了十余年的人形机器人Atlas正式退役。在这段告别视频中,Atlas完成了最后一次精彩表演,并回顾了艰辛历程中的种种失败,向人类躹躬致谢,无数观众为之泪目。在过去的十年中,Atlas一直是人形机器人技术领域的标志性产品,它的幽默的舞蹈、谨慎的搬运、灵敏的三级跳、华丽的后空翻、灵活的雪地行走、顽强的抗击打能力,曾经让人们惊艳不已,被视作人形机器人的巅峰之作。

新款Tinius Olsen 20 kN气动夹具全球发售

Tinius Olsen最近发布了一款新型20 kN气动夹具,并于上个月在中国上海举办的ITMA ASIA+CITME 2023展会期间在SDL Atlas展位上首次公开展示。该种力值更大的气动夹具由一个内置阀门操作,即使试样在测试过程中厚度减小,也能保持正压力。

ICESat-2/ATLAS雷达数据在林业中的应用

ICESat-2/ATLAS作为新一代星载光子计数雷达,较ICESat-1/GLAS在载荷设备、模型算法及测距精度等方面都有了极大的提升。本文介绍了ICESat-2/ATLAS的物理参数、数据产品,并列出了当前ATLAS数据产品去噪分类的各类算法,总结了ICESat-2/ATLAS数据在林业中的应用,包括制作高分辨率的森林高度制图、森林地上生物量的反演、林下地形的反演等,并对比分析了各类研究结果的观测精度及可靠性。最后,对ICESat-2/ATLAS在多源遥感协同不同数据源的应用上进行了展望。本研究能为星载光子计数雷达在林业中的应用提供一定的参考。

Neuroform Atlas支架辅助弹簧圈栓塞未破裂性颅内宽颈动脉瘤

目的:探讨Neuroform Atlas支架辅助弹簧圈栓塞治疗未破裂性颅内宽颈动脉瘤的安全性和有效性。方法:回顾性分析2020年8月至2021年9月于北京大学第三医院介入血管外科采用Neuroform Atlas支架辅助弹簧圈栓塞治疗的62例未破裂性颅内宽颈动脉瘤患者的临床资料。62例患者共64个颅内动脉瘤,其中25个位于大脑中动脉M1分叉处,16个位于前交通动脉,10个位于颈内动脉C7段,5个位于颈内动脉C6段,4个位于基底动脉顶端,3个位于大脑前动脉A3段,1个位于大脑中动脉M2段。有49个动脉瘤采取单支架辅助栓塞,其余15个采用双支架技术辅助栓塞(“Y”型14个和“X”型1个)。所有患者术后即刻行数字减影血管造影(digital subtraction angiography,DSA),评估动脉瘤的即刻栓塞效果(Raymond评分)和载瘤动脉血流情况。出院后对患者进行临床随访及影像学随访,临床随访为术后3个月采用改良Rankin评分量表(modified Rankin Scale,mRS)评估患者的临床预后;影像学随访为术后6个月复查脑血管DSA评估患者动脉瘤闭塞情况(Raymond评分)和载瘤动脉通畅情况。结果:62例患者手术均获成功,技术成功率为100%。术后即刻DSA显示:57个动脉瘤(89.1%,57/64)完全闭塞(RaymondⅠ级),6个动脉瘤(9.3%,6/64)瘤颈残留(RaymondⅡ级),1个动脉瘤(1.6%,1/64)瘤体残留(RaymondⅢ级)。3例患者(4.8%,3/62)出现围手术期并发症:2例为术中血栓形成,予动脉内替罗非班治疗后血流恢复,1例为术后局限性蛛网膜下腔出血,经保守治疗出院时均无致残性功能障碍。术后3个月时55例患者获临床随访,均预后良好(mRS≤2分),7例患者失访。50例患者(52个动脉瘤)获术后6个月DSA影像随访:RaymondⅠ级45个(86.5%,45/52),RaymondⅡ级4个(7.7%,4/52),RaymondⅢ级3个(5.8%,3/52),12例患者失访。结论:Neuroform Atlas支架辅助弹簧圈栓塞未破裂性颅内宽颈动脉瘤有很高的技术成功率,并具有较低的围手术期并发症发生率和较高的动脉瘤闭塞率,安全性和有效性均较好,但其远期疗效还需进一步随访观察。

Neuroform Atlas在双支架技术治疗复杂颅内分叉部宽颈动脉瘤中的应用

目的 探讨Neuroform Atlas支架在双支架技术辅助弹簧圈栓塞治疗复杂颅内分叉部宽颈动脉瘤(wide-neck bifurcation aneurysm,WNBA)的安全性和有效性。方法 回顾性分析我院2020年12月到2022年5月采用Neuroform Atlas支架以双支架辅助栓塞(dual stent-assisted coiling,DSAC)治疗的19例复杂WNBA患者临床资料。19例患者共19个WNBA,其中5个为破裂性动脉瘤。分布部位:10个位于大脑中动脉M1分叉处、7个位于前交通动脉、2个位于基底动脉尖。其中18个动脉瘤采取“Y”型DSAC、1个采取“X”型DSAC技术治疗。栓塞后复查脑血管数字减影血管造影(digital subtraction angiography,DSA),采取Raymond量表评估即刻栓塞效果和载瘤动脉情况。临床随访采用改良Rankin评分量表(modified Rankin Scale,mRS)评估患者术后3个月临床预后,影像学随访采取复查全脑血管DSA,评估患者术后6个月时动脉瘤闭塞情况和载瘤动脉通畅情况。结果 19例患者手术步骤均成功,技术成功率为100%。DSAC后复查DSA显示动脉瘤即刻闭塞情况:16个为RaymondⅠ级(完全闭塞)、2个为RaymondⅡ级(瘤颈残留)、1个为RaymondⅢ级(瘤体残留)。1例患者出现围手术期并发症:为分支动脉内血栓形成,予替罗非班治疗后血流恢复,苏醒后无明显神经功能缺失。术后3个月时18例患者获临床随访,均预后良好(m RS≤2分)。术后6个月时16例患者获DSA影像随访:RaymondⅠ级14个、RaymondⅡ级1个、RaymondⅢ级1个。结论 Neuroform Atlas支架用于DSAC治疗WNBA技术成功率较高,有较低的围手术期并发症发生率和较高的动脉瘤闭塞率,较传统支架有一定优势,但其远期疗效还需进一步随访观察。

嵌入式智能计算机计算能力评测方法

计算能力评测方法是嵌入式智能计算机领域的研究热点之一.嵌入式智能计算机存在多种神经形态计算的优化方案和机器学习加速器,导致其评测难度大于通用计算机.基准测试是目前普遍采用的评测方法,但是在资源受限的嵌入式设备上,基准测试集和评价指标的复用能力有限,难以适应配置多样化的嵌入式智能系统;测试集中神经网络模型的计算强度存在一定的随机性,无法充分挖掘待测设备的计算潜力;评价指标不统一,难以对不同嵌入式智能计算机的计算能力进行对比分析.本文提出了一种基于神经进化算法的嵌入式智能计算机计算能力评测方法.首先,基于Roofline理论模型,融合计算潜力挖掘、资源适配和评价指标统一等优势,提出了一种适配各种嵌入式智能计算机的计算能力评测框架,并对其合理性进行分析;其次,提出了一种评测计算能力的神经网络模型生成算法,利用神经进化算法,使生成模型的计算强度逼近嵌入式智能计算机的计算强度上限,充分挖掘待测设备的计算潜力,使评测结果更客观;然后,采用环境固定的上位机作为对照,分别在待测设备和上位机交叉运行生成的神经网络模型,并以两次执行推断任务时的每秒浮点运算次数作为计算因子,给出计算能力评测的通用公式,可以实现不同嵌入式智能计算机计算能力的对比分析;最后,在Mindspore-cpu、Tensorflow-cpu和Mindspore-ascend310框架下评测华为Atlas200,相比基准测试中常用的5种神经网络模型,采用本文生成的神经网络模型的测评结果更合理,证实两个DaVinci核心的智能计算能力是八个Cortex-A55核心的42.37倍.

Visualizing the neuroanatomical changes in Han Chinese adulthood: A pseudo-longitudinal study based on age-related large-scale statistical Chinese brain atlases

Objective:Understanding how brain changes over lifetime provides the basis for new insights into neurophysiology and neuropathology.In this study,we carried out a pseudo-longitudinal study based on age-related Chinese brain atlases(i.e.,Chinese2020)constructed from large-scale volumetric brain MRI data collected in normal Han Chinese adults at varying ages.Methods:In order to quantify the deformation and displacement of brains for each voxel as age increases,optical flow algorithm was employed to compute motion vectors between every two consecutive brain templates of the age-related brain atlas,i.e.,Chinese2020.Results:Dynamic age-related neuroanatomical changes in a standardized brain space were shown.Overall,our results demonstrate that brain inward deformation(mainly due to atrophy)can appear in adulthood and this trend generally accelerates as age increases,affecting multiple regions including frontal cortex,temporal cortex,parietal cortex,and cerebellum,whereas occipital cortex is least affected by aging,and even showed some degree of outward deformation in the midlife.Conclusion:Our findings indicated more complicated age-related changes instead of a simple trend of brain volume decrease,which may be in line with the recently increasing interests in the age-related cortical complexity with other morphometry measures.