简介:Interactivepickingisafundamentaltaskforprocessingdigitalgraphicsandimages.Combiningtwo-dimensional(2D)andthree-dimensional(3D)viewsisanimportanttrendinprocessingmedicalsliced-geometries.Withthegoalofaccuratelypickingpointsinsliced-geometries,wepresentamethodtopickpointsinteractivelybycombining2Dand3Dviewsofsliced-geometries.Fourviewwindowsareconstructedtorendersliced-geometriesusingmedicalimaginginteractiontoolkit(MITK):threefor2DslicesparalleltotheXY,YZ,andXZplanesandoneforsurfacerendering.Pointspickedinaviewwindowcanberenderedandmodifiedinanywindow.Afterapointispicked,itisimmediatelyconvertedintoaworldcoordinateandstored.Pointsareonlyrenderedandmodifiedbasedontheirworldcoordinates.Thispickingmethodcanbeusedforfitting,observing,andmeasuringlocalregions.Thismethodcanalsobeusedforinteractivesegments.Ourexperimentshowsthatthepickingmethodisaccurateandhasareal-timepickingeffect.
简介:Totesttheaccuracyofanewfingerprintscanner,researcherstypicallyrunmillionsofknownfingerprintimagesthroughthesystem'smatchingsoftware.Butthistestingprocedurecan'tquitemimicrealoperatingconditions,asa2-Dimagefedintoaprogramisfundamentallydifferentthana3-Dfingerpressedtoasensor.
简介:
简介:Comparedwithtraditionalmagneticinductiontomography(MIT)system,theinnovationofthispaperisthatthefunctionofvoicebroadcastbasedonWT588Dmoduleisaddedtothesystem.ThenewMITsystemavoidsfrequentlylookingintotheinstrumentanddecreasessartificiallyreadingresidualfromthehardwaredebuggingtothetestprocess.Itmakestheprocessofdataacquisitioneasierandenhancestheaccuracyandefficiencyofdetectioninthesystem.Italsogivesadetaileddescriptiontooptimizethedesignofsoftwareandhardware.Thevoicebroadcastsubsystemhastheadvantagesofsimpleintegralstructure,highqualitysound,lowcost,convenientfunctionexpansion,whichcanbeusedasatemplatebasedontheWT588Dmodularappliedinotherrelatedfields.
简介:Inrecentyears,thevolumetricapproachof3Dreconstructionhasemergedasapowerfulclinicmethodforbiomedicalapplications.Thisapproachisbasedonavoxelrepresentedobjectwhichmaintainsmanydetailsofthesurfaceandtheinternalcon-structuresoftheoriginalbody.Becauseofahugeamountofdata,thevolumedatasettakesalargesizeofmemory,andahighprocessingspeedisrequiredinorderto
简介:Usingofactivecardiacmedicaldevicesincreasessteadily.InEurope,therewere183implantsofICDand944implantsofPM,119ofbiventricularICDand41ofbiventricularPM,allpermillionin-habitantsin2014.Healthcareenvironments,includingradiotherapytreatmentrooms,areconsideredchallengingfortheseimplantabledevices.Exposuretoradiationmaycausethedevicetoexperienceprematureelectivereplacementindicator,decreasedpacingamplitudeorpacinginhibition,inappropri-ateshocksorinhibitionoftachyarrhythmiatherapyandlossofdevicefunction.
简介:Cellaffinityisanimportantfactortobeconcernedwhenbiodegradablepolymericmaterialsareutilizedascellscaffoldintissueengineering.Manystudieshaveprovedthathydrophilicity,surfaceenergy,chargeandroughnessofthematerialsurfacegreatlyinfluencethecellattachmentandcellgrowthonthematerial.
简介:Inordertoeliminatedisplacementandelasticdeformationbetweenimagesofadjacentframesincourseof3Dultrasonicimagereconstruction,elasticregistrationbasedonskeletonfeaturewasadoptinthispaper.Anewautomaticallyskeletontrackingextractalgorithmispresented,whichcanextractconnectedskeletontoexpressfigurefeature.Featurepointsofconnectedskeletonareextractedautomaticallybyaccountingtopicalcurvatureextremepointsseveraltimes.Initialregistrationisprocessedaccordingtobarycenterofskeleton.Whereafter,elasticregistrationbasedonradialbasisfunctionareprocessedaccordingtofeaturepointsofskeleton.Resultofexampledemonstratethataccordingtotraditionalrigidregistration,elasticregistrationbasedonskeletonfeatureretainnaturaldifferenceinshapefororgan'sdifferentpart,andeliminateslightelasticdeformationbetweenframescausedbyimageobtainedprocesssimultaneously.Thisalgorithmhasahighpracticalvalueforimageregistrationincourseof3Dultrasoundimagereconstruction.
简介:在生物医学的图象数据的显示,操作和分析的进程,他们通常需要通过插值的进程被变换成各向同性的discretization的数据,当立方的卷绕旋转插值广泛地由于它在计算费用和精确性之间的好折衷被使用时。在这份报纸,我们在场为3D的一个整个概念医药图象插值基于立方的卷绕旋转,和六个方法与不同锋利的控制参数,它在细节被提出。而且,我们也与不同的片间距用数据集合为这些方法给客观比较。在这些数据集合的每片被每个插值方法并且与原来的片相比用三项措施估计:摆平平均数的差别,争论的地点的数字,和最大的差别。根据试验性的结果,我们在场为3D的一个建议在不同状况下面的医药图象最后。
简介:据JohnsonBN2015年9月18日(AdvancedFunctionalMaterials,2015.doi:10.1002/adfm.2015017.)报道,美国明尼苏达大学、普林斯顿大学研究人员发现3D打印技术或许有助于神经系统修复。神经系统疾病一直是困扰着医学工作者的一个难题。其中神经元的损伤是导致神经系统疾病的重要诱因。因此,如果要治疗这些疾病,如何修复受损神经元就成为了摆在科学家面前的头号难题。一直以来,人们都在寻找有效方法来促使受损神经元再生。研究人员在小鼠实验中证实了这一想法。利用3D扫描获得了小鼠坐骨神经的轮廓,然后再利用3D打印技术打造出了一个内含能够促进神经元再生化学成分的硅酮类支架。
简介:以氨基酸组成为特征对膜蛋白的分类,忽略了序列残基之间的相关性信息,而采用传统支持向量机算法作为分类算法,在解决多类问题时会出现分类盲区问题。针对这两种情况,计算蛋白质序列的氨基酸组成、二肽组成以及6种氨基酸相关系数,将三类特征结合,作为膜蛋白序列的特征向量;同时采用模糊支持向量机作为分类器,解决了传统支持向量机在多类数据识别中的盲区问题。测试结果表明,在相同特征输入下,模糊支持向量机分类性能优于传统支持向量机;在相同分类器的情况下,氨基酸组成、二肽组成和相关系数组合的特征选择方法的分类性能优于只使用其中一类或两类特征的方法;而采取组合特征和模糊支持向量机相结合的分类策略,在独立性数据集测试中的整体预测精度达到97%,优于现有的多种分类策略,是目前最有效的膜蛋白分类方法之一。
简介:Segmentingwholeheartfromcardiaccomputedtomography(CTimagescanprovideanimportantbasisfortheevaluationofcardiacfunctionandhelpimprovetheaccuracyofclinicaldiagnosis.Manualsegmentationisthemostaccuratemethodforcardiacsegmentation.Butitistimeconsumingandnotsufficientlyreproducible.However,cliniciansstillrelyonthismethodinpracticalapplications.Soafullyautomaticmethodisneededtoimprovethesegmentationefficiency.Thispapeproposesaregistration-basedautomaticapproachforthree-dimensional(3DsegmentationofcardiacCTimages.TheproposedmethodutilizesthesimilarityocardiacCTimagesbetweendifferentindividuals,andusesregistrationtoachievethesegmentation.Affinetransformationisfirstlyimplementedtoachieveglobalcoarseregistration.Then,cubicB-splinesareusedtorefinethelocaldetailsinlocallyaccurateregistration.Mutualinformation(Ml)isusedasthesimilaritymeasure,andadaptivestochasticgradientdescent(ASGD)astheoptimizationalgorithm.Oumethodisappliedtothedual-sourcecardiacCTimagestosegmentwholeheartExperimentalresultsshowthattheproposedmethodcanautomaticallysegmentwholeheartfromcardiacCTimages.
简介:Thispaperproposesapracticalalgorithmsofplanecutting,stereoclippingandarbitrarycuttingfor3Dsurfacemodelreconstructedfrommedicalimages.Inplanecuttingandstereoclippingalgorithms,the3Dmodeliscutbyplaneorpolyhedron.Listsofedgeandvertexineverycutplaneareestablished.Fromtheseliststheboundarycontoursarecreatedandtheirrelationshipofembraceisascertained.TheregionclosedbythecontoursistriangulatedusingDelaunaytriangulationalgorithm.Arbitrarycuttingoperationcreatescuttingcurveinteractively.Thecutmodelstillmaintainsitscorrecttopologystructure.Withtheseoperations,tissuesinsidecanbeobservedeasilyanditcanaiddoctorstodiagnose.Themethodscanalsobeusedinsurgeryplanningofradiotherapy.
简介:目的探讨3D打印辅助改良Stoppa入路治疗骨盆骨折疗效分析。方法2016年2月至2016年12月我科应用3D打印结合改良的Stoppa入路治疗骨盆骨折12例,其中平均年龄38.5岁,男性8例,女性4例,根据Tile分型:B1型4例,B2型4例,B3型1例,Cl型2例,C2型1例。结果平均手术时间75min(50~125)min,平均出血量350mL(150~500)mL。术后根据Matta标准评分优10例,良2例,3月后随访所有患者骨折完全愈合,髋关节功能良好,无内固定损伤断裂,无大腿麻木等神经损伤症状,无腹股沟疝,下肢不等长,深静脉血栓形成等并发症。结论3D打印辅助改良Stoppa入路治疗骨盆骨折可以缩短手术时间,降低出血量,减少并发症等优点,但在一定程度上增加患者经济负担。
简介:目的通过手术对照,探讨急诊三维CT血管造影(3D-CTA)在前循环动脉瘤的临床应用及其价值,同时为了检验3D-CTA检查的诊断符合情况。方法59例自发性蛛网膜下腔出血(SAH)患者,其中男性31例,女性28例;年龄35~70岁,平均年龄51岁。对普通CT高度怀疑为动脉瘤性SAH患者行急诊Philips256iCT扫描,以最大密度投影(MIP)和容积重建(VR)进行三维重建,多角度显示动脉瘤病灶。对3D-CTA检查未发现动脉瘤患者再进行数字减影血管造影(DSA)检查核对。对诊断为前循环动脉瘤的48例患者在72h内进行开颅显微手术夹闭术。术后复查CTA检查夹闭效果。结果前循环动脉瘤患者48例,其中男性25例,女性23例;年龄39~70岁,平均年龄52岁。急诊3D-CTA从59例SAH患者中共检出前循环动脉瘤49个,手术证实前循环动脉瘤50个,漏诊1个。手术证实动脉瘤形态、大小、指向、瘤颈宽度与3D-CTA所见基本相同,准确率98%。48例接受手术治疗的患者术后0~18个月进行随访,预后良好(GOS4、5级)42例(87.5%),严重残疾(GOS2、3级)4例(8.4%),死亡2例(4.1%)。结论急诊3D-CTA检查具有无创、检查迅速、便捷、相对经济等优点,可作为颅内动脉瘤筛查的最佳方法。并可在早期或超早期指导显微神经外科手术治疗。
简介:目的探讨3D打印个体化手术治疗方案在骨盆骨折中的临床疗效。方法收集南京医科大学第二附属医院骨盆骨折的患者22例,随机分为3D组和对照组,3D组病例采用3D打印技术建模,术前模拟手术,定制个体化解剖钢板后进行手术治疗,对照组常规手术治疗,所有病例记录手术所用时间、出血量和内固定物的长度、弧度、螺钉数量和方向,术后应用Matta评分评估疗效,比较两组病例手术时间、术中出血量的差异。结果所有病例术后随访2~8月,Mata评估标准均为优,3D组手术时间为(60±12)min,较对照组明显缩短,术中出血量为(218±63)mL,较对照组明显减少,手术入路与术前设计一致,内固定位置与术前手术方案设计完全一致。结论3D打印个体化治疗方案治疗骨盆骨折,可以给每一位骨盆骨折患者设计个体化解剖形内固定物,制定个性化手术方案,并且具有减少手术创伤、缩短手术时间、降低手术难度等优势,临床疗效满意,值得推广。