楊志永,房培娜,劉琬鈺,姜?杉
基于代價(jià)地圖的活檢路徑規(guī)劃研究
楊志永,房培娜,劉琬鈺,姜?杉
(天津大學(xué)機(jī)械工程學(xué)院,天津 300350)
穿刺活檢手術(shù)作為多種疾病早期診斷最常用的技術(shù)之一,需要將一根細(xì)長(zhǎng)的穿刺針刺入組織以獲取組織標(biāo)本,因此,穿刺路徑的安全性和精確性對(duì)活檢手術(shù)準(zhǔn)確率和術(shù)后恢復(fù)至關(guān)重要.針對(duì)機(jī)器人輔助下的穿刺活檢手術(shù),本文提出了一種基于代價(jià)地圖的穿刺路徑規(guī)劃方法.首先,將包含三維規(guī)劃空間大小、空間中障礙物位置、形狀等信息的原始地圖作為輸入,根據(jù)人工勢(shì)能場(chǎng)理論構(gòu)建代價(jià)地圖,以定量評(píng)估障礙物以外空間的風(fēng)險(xiǎn)程度.其次,考慮帶有斜尖的穿刺針插入組織中產(chǎn)生的實(shí)際偏轉(zhuǎn),采用非完整運(yùn)動(dòng)學(xué)模型作為穿刺針的偏轉(zhuǎn)預(yù)測(cè)模型.在模擬解剖結(jié)構(gòu)特征的障礙物環(huán)境下,使用代價(jià)地圖計(jì)算可行入刺區(qū)域中不碰撞穿刺路徑的代價(jià)值,并采用模擬退火算法進(jìn)行優(yōu)化,獲得安全可行的穿刺路徑以指導(dǎo)機(jī)器人執(zhí)行的自動(dòng)穿刺過程.最后,在自制的仿生模型中設(shè)計(jì)模體實(shí)驗(yàn),分別按照傳統(tǒng)直線穿刺路徑和本文規(guī)劃方法得出的安全穿刺路徑執(zhí)行自動(dòng)穿刺操作.實(shí)驗(yàn)表明,按照規(guī)劃路徑進(jìn)行穿刺的實(shí)際穿刺路徑與規(guī)劃路徑較為相符,靶點(diǎn)穿刺誤差在1.5mm以內(nèi),較傳統(tǒng)直線路徑穿刺精度提高80%以上,且較臨床的靶點(diǎn)穿刺誤差(5.5~5.6mm)有明顯降低.實(shí)驗(yàn)結(jié)果表明該運(yùn)動(dòng)規(guī)劃方法能夠?yàn)闄C(jī)器人輔助下的自動(dòng)穿刺手術(shù)規(guī)劃出安全合理的穿刺路徑,有效提高自動(dòng)穿刺手術(shù)的安全性和精確性.
活檢;路徑規(guī)劃;人工勢(shì)能場(chǎng);代價(jià)地圖
癌癥已經(jīng)成為威脅中國(guó)人群健康的主要公共衛(wèi)生問題之一[1],早期診斷及術(shù)后的有效干預(yù)能夠大大提高患者生存率.近年來,活檢手術(shù)已經(jīng)成為臨床多種疾病診斷的重要標(biāo)準(zhǔn).組織活檢作為經(jīng)皮穿刺手術(shù)的一種,合理安全穿刺路徑的選擇和高穿刺精度能夠有效提高手術(shù)成功率,減輕患者痛苦.然而,在術(shù)前確定安全有效的穿刺路徑仍然是一個(gè)重要挑戰(zhàn),即使在CT、MRI等醫(yī)學(xué)圖像輔助的基礎(chǔ)下,外科醫(yī)生仍必須依據(jù)經(jīng)驗(yàn)評(píng)估解剖結(jié)構(gòu)內(nèi)部可能的軌跡來選擇穿刺路徑以盡量避開危及器官.此外,臨床中通常規(guī)劃直線穿刺路徑[2-3],忽略針于穿刺過程中的變形.對(duì)于具有不對(duì)稱斜尖的標(biāo)準(zhǔn)規(guī)格活檢針,實(shí)際穿刺路徑往往會(huì)偏離所規(guī)劃的直線路徑.實(shí)踐中,外科醫(yī)生借助影像系統(tǒng)確定穿刺路徑并在穿刺過程中手動(dòng)調(diào)整針的偏轉(zhuǎn),調(diào)查顯示有經(jīng)驗(yàn)的醫(yī)生穿刺手術(shù)的靶區(qū)誤差為5.5~5.6mm[4-5].
為了提高手術(shù)精度和穿刺安全性,隨著機(jī)器人自動(dòng)穿刺系統(tǒng)的逐步發(fā)展,用于指導(dǎo)機(jī)器人穿刺的可操縱針路徑規(guī)劃得到了廣泛研究.國(guó)內(nèi)外學(xué)者分別在二維、三維環(huán)境下,研究穿刺針的變形預(yù)測(cè)[6-7]、操控方法[8-9]、規(guī)劃算法[10-14]等內(nèi)容,取得了大量成果,但也存在一些問題.例如,現(xiàn)有的可操縱針路徑規(guī)劃方法,規(guī)劃環(huán)境的選擇多以規(guī)則的幾何空間為主,不能夠模擬復(fù)雜的解剖結(jié)構(gòu)特點(diǎn).此外,穿刺操作需借助實(shí)時(shí)傳感技術(shù)和術(shù)中成像技術(shù),更加大了臨床手術(shù)的實(shí)施難度.
基于上述原因,本文以活檢穿刺手術(shù)為例,從臨床需求出發(fā),考慮三維復(fù)雜規(guī)劃環(huán)境和針的實(shí)際偏轉(zhuǎn),研究機(jī)器人操控下斜尖針的穿刺機(jī)理,提出了一種基于代價(jià)地圖的斜尖針穿刺路徑規(guī)劃方法,可有效避開危及器官并對(duì)靶區(qū)實(shí)現(xiàn)精準(zhǔn)穿刺,大大降低手術(shù)風(fēng)險(xiǎn),提高手術(shù)成功率.
本文提出的路徑規(guī)劃方法為了滿足穿刺的安全性和準(zhǔn)確性,利用預(yù)先構(gòu)建的代價(jià)地圖來描述穿刺路徑與危及器官之間發(fā)生碰撞的可能性,采用非完整運(yùn)動(dòng)學(xué)模型作為針的變形預(yù)測(cè)模型,基于模擬退火算法獲得安全可行的穿刺路徑,確定有效的入刺位置和方向以提高穿刺精度,最大限度減輕組織損傷.
代價(jià)地圖最初被應(yīng)用于移動(dòng)機(jī)器人領(lǐng)域,即用地圖像素灰度值描述障礙物存在的可能性[15].默認(rèn)的地圖是一張黑白灰三色圖像,即地圖中只標(biāo)出有、無障礙物和未知3種區(qū)域,在這樣地圖中進(jìn)行的路徑規(guī)劃將會(huì)是不安全的.代價(jià)地圖則是在原始地圖上進(jìn)行膨脹(向外擴(kuò)張),以原始地圖作為輸入,選用特定函數(shù)定義受到障礙物影響區(qū)域的代價(jià)值,該函數(shù)為空間點(diǎn)與障礙物距離的遞減函數(shù),可以將規(guī)劃空間坐標(biāo)系下的距離值轉(zhuǎn)換為代價(jià)地圖中的代價(jià)值.以描述路徑與障礙物之間發(fā)生碰撞的可能性,通過地圖中的代價(jià)而不是簡(jiǎn)單的布爾值判斷來指導(dǎo)路徑規(guī)劃,從而提供更加合理的路徑規(guī)劃依據(jù),對(duì)穿刺路徑的危險(xiǎn)程度進(jìn)行定量評(píng)估,實(shí)現(xiàn)更加安全可靠的規(guī)劃.
圖2 障礙物二值圖像編碼代價(jià)地圖流程二維示意
本文方法在可行入刺區(qū)域內(nèi)隨機(jī)采樣入刺點(diǎn)以生成多種路徑解,基于構(gòu)建的代價(jià)地圖進(jìn)行路徑碰撞檢測(cè)并計(jì)算各路徑的代價(jià)值,采用模擬退火算法優(yōu)化穿刺路徑.
非完整運(yùn)動(dòng)學(xué)模型[6]假設(shè)組織相對(duì)于穿刺針具有足夠的剛度,穿刺針在組織中以一個(gè)恒定曲率(曲率數(shù)據(jù)通過穿刺實(shí)驗(yàn)獲得)的圓弧軌跡運(yùn)動(dòng),Wilson等[17]對(duì)該運(yùn)動(dòng)學(xué)模型的準(zhǔn)確性做出評(píng)估,結(jié)果顯示單彎曲剖面的模擬路徑與實(shí)際路徑的平均誤差為0.8mm,雙彎曲剖面的模擬路徑與實(shí)際穿刺路徑的平均誤差為0.9mm.為了提高穿刺精度、簡(jiǎn)化操作要求、提高臨床可行性,本文以單彎曲非完整運(yùn)動(dòng)學(xué)模型作為斜尖穿刺針的變形預(yù)測(cè)模型,指導(dǎo)穿刺路徑規(guī)劃器.盡管一段圓弧軌跡被限制在二維平面內(nèi),本文將穿刺路徑求解方法拓展到三維空間,如圖3所示,當(dāng)給定穿刺點(diǎn)和方向,通過在穿刺前改變針斜尖朝向,針的可達(dá)區(qū)間為一個(gè)空間錐形的表面.
運(yùn)動(dòng)規(guī)劃問題表述如下,參數(shù)如圖4所示.
圖3?穿刺針的可達(dá)域
圖4?運(yùn)動(dòng)規(guī)劃方法圖解
輸出:一個(gè)能夠確定穿刺路徑的解集合,包括插入點(diǎn)位置、插入方向等信息,或者報(bào)告無可行解.
式中:通過和世界坐標(biāo)系z(mì)軸單位向量做向量積歸一化得到;表示與夾角,由點(diǎn)乘的定義求得;.給定、和,計(jì)算出穿刺深度和坐標(biāo)系下可行入刺點(diǎn)集合.
在實(shí)際操作中,醫(yī)生可以根據(jù)臨床需求來執(zhí)行運(yùn)動(dòng)規(guī)劃器,給定不同的插入方向,獲得可行的解決?方案.
本節(jié)設(shè)計(jì)并進(jìn)行了穿刺實(shí)驗(yàn),以獲得針軌跡的曲率半徑,并在自制的模擬解剖特征的障礙物環(huán)境中構(gòu)建代價(jià)地圖,規(guī)劃得出安全路徑并進(jìn)行模體實(shí)驗(yàn),驗(yàn)證了本文方法的可行性和準(zhǔn)確性.
本文使用20G斜尖穿刺針,制作仿生組織(食用明膠和水以質(zhì)量比1∶10混合,冷藏8h以上獲得)模擬穿刺環(huán)境.
自動(dòng)穿刺及數(shù)據(jù)采集實(shí)驗(yàn)平臺(tái)如圖6(a)所示,分為針操控裝置、數(shù)據(jù)采集裝置,通過線性模組實(shí)現(xiàn)穿刺動(dòng)作,三自由度直角坐標(biāo)機(jī)器人夾持邁瑞公司生產(chǎn)的中心頻率為7.5MHz的線陣超聲探頭采集穿刺后組織中的針截面二維圖像,識(shí)別針截面位置以獲得針軌跡數(shù)據(jù).采用標(biāo)準(zhǔn)UVC-USB采集卡將超聲圖像傳輸?shù)絇C端,采集圖片像素為640×546.使用該設(shè)備在明膠模型中執(zhí)行多次插入,重建針形狀以獲得針在穿刺過程中的實(shí)際偏轉(zhuǎn)曲率參數(shù).
胸腹部自動(dòng)經(jīng)皮穿刺機(jī)器人及仿生組織模型如圖6(b)所示,穿刺機(jī)器人具有、、方向平移和沿軸轉(zhuǎn)動(dòng)的定位自由度,穿刺模組具有直線穿刺和調(diào)整針斜尖朝向自由度.仿生模型使用明膠制作,具有150mm×120mm×55mm邊界大小,內(nèi)部嵌有不規(guī)則分叉狀結(jié)構(gòu)用來模擬危及器官(大血管).使用該機(jī)器人操控針在模型中按照規(guī)劃路徑執(zhí)行穿刺操作,驗(yàn)證規(guī)劃結(jié)果的準(zhǔn)確性.
圖6?實(shí)驗(yàn)設(shè)備及材料
為了獲得針插入仿生組織明膠中路徑的曲率半徑,將實(shí)驗(yàn)所用穿刺針以5mm/s的速度(臨床的針穿刺速度在0.4~10mm/s之間[18])進(jìn)行穿刺,圖7為針在明膠模型中的軌跡及通過數(shù)據(jù)采集系統(tǒng)獲得的針實(shí)驗(yàn)軌跡平均值及誤差.實(shí)驗(yàn)獲得針軌跡在本實(shí)驗(yàn)仿生組織中的曲率半徑為605mm,其中10次插入實(shí)驗(yàn)的平均誤差為0.9mm.
圖7?自動(dòng)穿刺實(shí)驗(yàn)結(jié)果
圖8?值對(duì)代價(jià)地圖構(gòu)建時(shí)間及規(guī)劃效果的影響
圖9?代價(jià)地圖的體繪制顯示
基于建立的代價(jià)地圖,在三維模擬環(huán)境下進(jìn)行路徑規(guī)劃.在臨床穿刺手術(shù)中,醫(yī)生根據(jù)解剖結(jié)構(gòu)特點(diǎn),目標(biāo)靶區(qū)位置,選擇最佳穿刺層面,根據(jù)最短進(jìn)針原則(盡可能垂直皮膚表面),避開重要結(jié)構(gòu)原則在病人體表確定穿刺進(jìn)針點(diǎn)、穿刺深度等信息[19].本文依據(jù)臨床術(shù)前穿刺計(jì)劃規(guī)范,通過設(shè)定的目標(biāo)靶點(diǎn)及規(guī)劃環(huán)境特點(diǎn),確定入刺區(qū)域中心點(diǎn),并執(zhí)行下一步的穿刺路徑優(yōu)化以指導(dǎo)機(jī)器人自動(dòng)穿刺操作.圖10展示了本文方法依據(jù)代價(jià)地圖計(jì)算出的優(yōu)化路徑結(jié)果,分別為規(guī)劃路徑在模擬環(huán)境下和代價(jià)地圖中的結(jié)果,黑色直線為傳統(tǒng)方式的針插入路徑,即從插入點(diǎn)指向靶點(diǎn)的直線路徑.本文方法規(guī)劃出的針軌跡(藍(lán)色曲線)從計(jì)算的插入點(diǎn)(藍(lán)色點(diǎn))開始,到達(dá)選定的靶點(diǎn)(紅色點(diǎn)),同時(shí)盡量遠(yuǎn)離危及器官.
圖10?模擬解剖空間路徑規(guī)劃結(jié)果
使用課題組自主研發(fā)的胸腹部穿刺機(jī)器人進(jìn)行模型的穿刺實(shí)驗(yàn),將規(guī)劃入刺點(diǎn)轉(zhuǎn)化到機(jī)器人坐標(biāo)系下進(jìn)行穿刺,通過平面和平面的圖像獲取組織中的實(shí)際穿刺路徑,擬合結(jié)果如圖11所示. 從規(guī)劃路徑和實(shí)驗(yàn)結(jié)果可以得出,傳統(tǒng)方法實(shí)際穿刺路徑和規(guī)劃路徑的最大誤差為7.62mm,均方根誤差2.32mm,靶點(diǎn)處誤差7.62mm,本文方法穿刺路徑和規(guī)劃路徑最大誤差1.46mm,均方根誤差0.49mm,靶點(diǎn)誤差1.46mm.可以看出,本文方法的實(shí)際穿刺路徑能夠較好跟隨規(guī)劃路徑,在保證穿刺安全性的前提下,準(zhǔn)確到達(dá)靶點(diǎn).按照文本方法進(jìn)行穿刺的靶點(diǎn)穿刺精度相較于按照直線路徑進(jìn)行穿刺的靶點(diǎn)穿刺精度提升80%以上.靶點(diǎn)穿刺誤差較實(shí)際醫(yī)生手術(shù)中5.5~5.6mm的誤差[4-5]有較大降低,同時(shí),該誤差小于文獻(xiàn)[14]自動(dòng)穿刺過程中在明膠組織中產(chǎn)生的1.74mm的靶點(diǎn)誤差.從該結(jié)果可以看出,對(duì)于臨床廣泛應(yīng)用的帶有斜尖的穿刺針來說,本文提出的路徑規(guī)劃方法,考慮實(shí)際穿刺中針的偏轉(zhuǎn),基于代價(jià)地圖確定安全合理的術(shù)前入刺點(diǎn),盡可能減少穿刺過程中的調(diào)整,能夠減少手術(shù)過程中的擾動(dòng),有效提升穿刺精度,對(duì)機(jī)器人輔助的自動(dòng)穿刺手術(shù)具有指導(dǎo)意義.
圖11?規(guī)劃與實(shí)驗(yàn)結(jié)果對(duì)比
按照本文方法規(guī)劃的路徑進(jìn)行穿刺,靶點(diǎn)誤差產(chǎn)生的原因主要有兩方面,分別是針曲率識(shí)別過程中引入的誤差,致使實(shí)驗(yàn)獲得的針曲率與實(shí)際穿刺的曲率存在偏差.機(jī)器人穿刺操作過程中產(chǎn)生的誤差.從上述兩個(gè)原因來看,未來工作可通過改進(jìn)超聲圖像中針截面識(shí)別算法,提高曲率辨識(shí)精度,考慮包含組織和針力學(xué)特性參數(shù)的模型,提高模型普適性,提高穿刺機(jī)器人操作的穩(wěn)定性等方面來進(jìn)一步提升穿刺精度.
本文提出了一種基于代價(jià)地圖的斜尖針穿刺路徑規(guī)劃方法,用于指導(dǎo)機(jī)器人操控的穿刺操作.該方法基于編碼三維規(guī)劃環(huán)境風(fēng)險(xiǎn)信息的代價(jià)地圖,考慮穿刺針在組織中的實(shí)際偏轉(zhuǎn),在以斥力場(chǎng)公式建立的代價(jià)地圖指導(dǎo)下進(jìn)行規(guī)劃,以模擬退火算法求得安全可行的穿刺路徑.并按照規(guī)劃路徑和傳統(tǒng)直線路徑分別進(jìn)行模體實(shí)驗(yàn),結(jié)果顯示本文方法能夠有效指導(dǎo)機(jī)器人控制的自動(dòng)穿刺,提高穿刺精度,為后續(xù)研究自動(dòng)穿刺機(jī)器人提供指導(dǎo).
[1] 鄭榮壽,孫可欣,張思維,等. 2015年中國(guó)惡性腫瘤流行情況分析[J]. 中華腫瘤雜志,2019,41(1):19-28.
Zheng Rongshou,Sun Kexin,Zhang Siwei,et al. Report of cancer epidemiology in China,2015[J]. China Oncology,2019,41(1):19-28(in Chinese).
[2] Cynamon J,Shabrang C,Golowa Y,et al. Transfemoral transcaval core-needle liver biopsy:An alternative to transjugular liver biopsy[J]. Journal of Vascular and Interventional Radiology,2015,27(3):370-375.
[3] 鮑?潮,姜?杉,張國(guó)彬,等. 基于肺部CT 影像的針道識(shí)別關(guān)鍵技術(shù)研究[J]. 天津大學(xué)學(xué)報(bào)(自然科學(xué)與工程技術(shù)版),2019,52(4):346-352.
Bao Chao,Jiang Shan,Zhang Guobin,et al. Research on key techniques of needle detection based on lung CT images[J]. Journal of Tianjin University(Science and Technology),2019,52(4):346-352(in Chinese).
[4] Blumenfeld P,Hata N,Dimaio S P,et al. Transperineal prostate biopsy under magnetic resonance image guidance:A needle placement accuracy study[J]. Journal of Magnetic Resonance Imaging,2007,26(3):688-694.
[5] Schouten M G,Bomers J G,Yakar D,et al. Evaluation of a robotic technique for transrectal MRI-guided prostate biopsies[J]. European Radiology,2012,22(2):476-483.
[6] Webster R J,Kim J S,Cowan N J,et al. Nonholonomic modeling of needle steering[J]. The International Journal of Robotics Research,2006,25(5):509-525.
[7] Misra S,Reed K B,Schafer B W,et al. Mechanics of flexible needles robotically steered through soft tissue[J]. The International Journal of Robotics Research,2010,29(13):1640-1660.
[8] Minhas D S,Engh J A,F(xiàn)enske M M,et al. Modeling of needle steering via duty-cycled spinning[C]//29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Lyon,F(xiàn)rance,2007:2756-2759.
[9] Rucker D C,Das J,Gilbert H B,et al. Sliding mode control of steerable needles[J]. IEEE Transactions on Robotics,2013,29(5):1289-1299.
[10] Park W,Wang Y,Chirikjian G S,et al. The path-of-probability algorithm for steering and feedback control of flexible needles[J]. The International Journal of Robotics Research,2010,29(7):813-830.
[11] Bernardes M C,Adorno B V,Poignet P,et al. Adaptive path planning for steerable needles using duty-cycling[C]// Intelligent Robots and Systems. San Francisco,USA,2011:2545-2550.
[12] Khadem M,Rossa C,Usmani N,et al. Semi-automated needle steering in biological tissue using an ultrasound-based deflection predictor[J]. Annals of Biomedical Engineering,2017,45(4):924-938.
[13] 楊志永,肖洪旭,李雨澤,等. 基于胸部CT影像的肺血管樹分割關(guān)鍵技術(shù)研究[J]. 天津大學(xué)學(xué)報(bào)(自然科學(xué)與工程技術(shù)版),2018,51(2):175-180.
Yang Zhiyong,Xiao Hongxu,Li Yuze,et al. Research on key technologies of pulmonary vascular Trees segmentation based on thoracic CT images[J]. Journal of Tianjin University(Science and Technology),2018,51(2):175-180(in Chinese).
[14] Moreira P,Boskma K J,Misra S,et al. Towards MRI-guided flexible needle steering using fiber bragg grating-based tip tracking[C]// 2017 IEEE International Conference on Robotics and Automation. Singapore,2017:4849-4854.
[15] Jaillet L,Cortés J,Siméon T,et al. Sampling-based path planning on configuration-space costmaps[J]. IEEE Transactions on Robotics,2010,26(4):635-646.
[16] Khatib O. Real-time obstacle avoidance for manipulators and mobile robots[C]// 1985 IEEE International Conference on Robotics and Automation. St Louis,USA,1985,2:500-505.
[17] Wilson E,Ding J,Carignan C,et al. Evaluation of nonholonomic needle steering using a robotic needle driver[C]// Medical Imaging 2010:Visualization,Image-Guided Procedures,and Modeling. San Diego,USA,2010:762523.
[18] DiMaio S P,Salcudean S E. Needle insertion modeling and simulation [J]. IEEE Transactions on Robotics and Automation,2003,19(5):864-875.
[19] 陳超民. 肺癌近距離放療手術(shù)機(jī)器人遠(yuǎn)程控制平臺(tái)的關(guān)鍵技術(shù)研究[D]. 天津:天津大學(xué),2017.
Chen Chaomin. Research on Key Technologies of Remote Control Platform of Surgical Robot for Lung Cancer Brachytherapy[D]. Tianjin:Tianjin University,2017(in Chinese).
Path Planning for Biopsy Based on Cost Map
Yang Zhiyong,F(xiàn)ang Peina,Liu Wanyu,Jiang Shan
(School of Mechanical Engineering,Tianjin University,Tianjin 300350,China)
Needle biopsy is one of the most commonly used techniques in the diagnosis of several diseases. A slender puncture needle is inserted into the tissue to get a tissue specimen. Therefore,the safety of the needle path and the accuracy of puncture are crucial to the accuracy of biopsy surgery and postoperative recovery. For robot-assisted biopsy surgery,this study proposed a path planning method based on the cost map. First,the original map containing the size of the three-dimensional planning space,location,and shape of obstacles in the space was taken as input. A cost map was constructed based on an artificial potential field to assess the degree of risk in the space beyond the obstacle. The actual deflection of the needle with a beveled tip during its insertion into the tissue was then considered. A nonholonomic kinematics model was used as the needle deflection prediction model. The cost value of the non-collision puncture path in a feasible insertion area,which was used to optimize the safe path based on a simulated annealing algorithm to guide the automatic puncture process performed by the robot,was calculated based on the cost map in an obstacle environment that simulates anatomical features. Finally,a phantom experiment was designed in the self-made bionic model. The automatic puncture operations were executed according to the traditional linear puncture path and the safe puncture path obtained by the planning method in this paper. The experiment showed that the path obtained by this method coincided well with the planning path. The target puncture error was within 1.5mm. The punc-ture accuracy was improved by more than 80% compared with the straight-line path. The target puncture error was significantly reduced compared with it in the clinical process(5.5—5.6mm). Results indicated that the motion planner could calculate a safe and reasonable path for the robot-controlled puncture surgery while enhancing the position precision and patient safety.
biopsy;path planning;artificial potential field;cost map
10.11784/tdxbz202008061
TK448.21
A
0493-2137(2022)02-0151-07
2020-08-24;
2020-09-18.
楊志永(1972—??),男,博士,教授. Email:m_bigm@tju.edu.cn
楊志永,meyang@tju.edu.cn.
國(guó)家自然科學(xué)基金資助項(xiàng)目(51775368,81871457,51811530310);天津市科技支撐計(jì)劃資助項(xiàng)目(18YFZCSY01300).
the National Natural Science Foundation of China(No. 51775368,No. 81871457,No. 51811530310),the Science and Technology Supportment Program of Tianjin,China(No. 18YFZCSY01300).
(責(zé)任編輯:王曉燕)
天津大學(xué)學(xué)報(bào)(自然科學(xué)與工程技術(shù)版)2022年2期