馬 珂 MA Ke岳云龍 YUE Yunlong周賾辰 ZHOU Zechen孫愛(ài)琦 SUN Aiqi陳 碩 CHEN Shuo李 睿 LI Rui
?
利用4D-Flow技術(shù)定量分析肺動(dòng)脈血流和相對(duì)壓力的可重復(fù)性研究
馬 珂 MA Ke
岳云龍 YUE Yunlong
周賾辰 ZHOU Zechen
孫愛(ài)琦 SUN Aiqi
陳 碩 CHEN Shuo
李 睿 LI Rui
Department of Biomedical Engineering, Tsinghua University, Beijing 100084, China
Address Correspondence to: LI Rui E-mail: leerui@tsinghua.edu.cn
修回日期:2015-09-06
中國(guó)醫(yī)學(xué)影像學(xué)雜志
2016年 第24卷3期:215-217,222
Chinese Journal of Medical Imaging 2016 Volume 24(3): 215-217, 222
【摘要】目的 肺動(dòng)脈血流動(dòng)力學(xué)參數(shù)的無(wú)創(chuàng)測(cè)量,對(duì)于肺動(dòng)脈高壓的診斷、預(yù)后評(píng)估和跟蹤研究有重要意義。本文主要基于4D-Flow技術(shù)對(duì)肺動(dòng)脈血流定量測(cè)量和相對(duì)壓力計(jì)算的可重復(fù)性進(jìn)行研究。資料與方法 對(duì)8例健康志愿者1周內(nèi)進(jìn)行2次獨(dú)立的肺動(dòng)脈4D-Flow成像,在肺動(dòng)脈干、左肺動(dòng)脈和右肺動(dòng)脈取3個(gè)截面,定量分析每個(gè)截面的最大流速、每搏輸出量、最大流量和平均相對(duì)壓力等參數(shù),分析各參數(shù)的相關(guān)性和可重復(fù)性。結(jié)果 利用2次掃描結(jié)果成功測(cè)量了肺動(dòng)脈最大流速、每搏輸出量、最大流量和平均相對(duì)壓力,2次掃描結(jié)果相關(guān)性較好(r=0.880 0.988、0.958、0.953,P<0.001)。Bland-Altman分析結(jié)果表明,最大流速、最大流量、每搏輸出量和峰值壓力結(jié)果有較好的可重復(fù)性(d=1.30、4.60、1.00、0.04 σd=±8.90、±29.30、±9.20、±0.35)。結(jié)論 利用4D-Flow技術(shù)可以對(duì)肺動(dòng)脈血流及相對(duì)壓力進(jìn)行定量分析,并且有較好的可重復(fù)性。
【關(guān)鍵詞】高血壓,肺性;磁共振成像;肺動(dòng)脈;血流動(dòng)力學(xué);圖像處理,計(jì)算機(jī)輔助
肺動(dòng)脈高壓是一種死亡率較高的惡性疾病,其特點(diǎn)是肺動(dòng)脈壓力升高和近端肺動(dòng)脈硬化,并伴隨血流動(dòng)力學(xué)變化,最終導(dǎo)致右心衰竭[1-2]。目前,右心導(dǎo)管術(shù)是診斷肺動(dòng)脈高壓的“金標(biāo)準(zhǔn)”,盡管具有較高的可靠性和較低的風(fēng)險(xiǎn),但仍為有創(chuàng)檢查。近年,隨時(shí)間變化的三維三方向相位對(duì)比MRI,即4D-Flow技術(shù)[3]得到迅速發(fā)展,利用該方法可進(jìn)行大范圍感興趣區(qū)的血流定量分析、復(fù)雜血流的三維可視化和高級(jí)血流動(dòng)力學(xué)參數(shù)計(jì)算[4-5]。4D-Flow技術(shù)使肺動(dòng)脈血流動(dòng)力學(xué)參數(shù)的無(wú)創(chuàng)測(cè)量成為可能,但在臨床應(yīng)用之前需要對(duì)這種測(cè)量方法的可重復(fù)性進(jìn)行驗(yàn)證,以保證參數(shù)測(cè)量的穩(wěn)定性。因此,本研究基于4D-Flow技術(shù)對(duì)肺動(dòng)脈流速、流量、每搏輸出量和相對(duì)壓力進(jìn)行測(cè)量,并對(duì)其進(jìn)行可重復(fù)性研究。
1.1 研究對(duì)象 對(duì)8例健康志愿者1周內(nèi)進(jìn)行2 次獨(dú)立的肺動(dòng)脈4D-Flow成像,其中男6例,女2例;年齡23~30歲,平均(25.0±2.5)歲。本研究經(jīng)本院醫(yī)學(xué)倫理委員會(huì)批準(zhǔn),所有受檢者均簽署知情同意書。
1.2 儀器與方法 采用 Philips Achieva TX 3.0T MRI儀和32通道心臟線圈分別進(jìn)行2次肺動(dòng)脈成像,采用斜橫斷面掃描,并使用回顧式心電門控,每心動(dòng)周期采集20個(gè)時(shí)相,掃描參數(shù):采集視野200 mm×200 mm ×80 mm,采集體素2 mm×2 mm×4 mm,翻轉(zhuǎn)角10°,3個(gè)梯度方向上的流速編碼范圍均設(shè)定為150 cm/s,TR 5 ms,TE 3 ms,時(shí)間分辨率40~50 ms。
1.3 數(shù)據(jù)分析 4D-Flow數(shù)據(jù)經(jīng)過(guò)渦流矯正和噪聲去除后[6],生成幅值加權(quán)的時(shí)間平均3D相位對(duì)比MR血管造影數(shù)據(jù),即3D PC-MRA數(shù)據(jù)[7],并將此數(shù)據(jù)輸入三維可視化軟件(EnSight,CEI,USA),在肺動(dòng)脈干、右肺動(dòng)脈、左肺動(dòng)脈分別截取截面,對(duì)于每個(gè)截面使用Matlab(Matlab,the Mathworks,USA)程序手工繪制所有時(shí)間幀的肺動(dòng)脈管腔輪廓,進(jìn)行最大流速、每搏輸出量、最大流量的定量分析[8],截面具體位置見(jiàn)圖1A。
圖1 男,27歲。A.不同時(shí)間肺動(dòng)脈跡線(pathline)可視化及分析血流的截面位置;B.血流流量分析結(jié)果
相對(duì)壓力的計(jì)算流程見(jiàn)圖2,使用3D PC-MRA數(shù)據(jù)可得到三維的動(dòng)脈結(jié)構(gòu)信息,并利用三維區(qū)域生長(zhǎng)算法對(duì)肺動(dòng)脈進(jìn)行半自動(dòng)分割,可獲得肺動(dòng)脈三維結(jié)構(gòu)模型。將血液假設(shè)為不可壓縮的黏性液體,壓力的梯度可通過(guò)納維葉-斯托克斯公式(Navier-Stokes)計(jì)算,見(jiàn)公式(1)。
其中,P為相對(duì)壓力,μ為血液的黏滯系數(shù),ρ為血液的密度,g為重力加速度。在實(shí)際掃描中,志愿者位于水平位,其血液重力可忽略。
對(duì)于每個(gè)給定的體素,其速度的時(shí)間和空間一階導(dǎo)數(shù)以及空間的二階導(dǎo)數(shù)可由其相鄰的時(shí)間上和空間上的速度值中心差分法求出[9]?;谒玫膲毫μ荻冉Y(jié)果,將區(qū)域生長(zhǎng)種子點(diǎn)相對(duì)壓力設(shè)為0,壓力的初始值為區(qū)域生長(zhǎng)路徑上每點(diǎn)的壓力梯度總和,相對(duì)壓力可通過(guò)迭代壓力泊松方程法[10]得出,其迭代過(guò)程見(jiàn)公式(2)。
圖2 相對(duì)壓力計(jì)算流程。A. PC-MRA數(shù)據(jù)的最大密度投影圖;B. PC-MRA數(shù)據(jù)二值化后的結(jié)果;C.分割后的肺動(dòng)脈;D.相對(duì)壓力的計(jì)算結(jié)果
其中,Pk為第k次迭代結(jié)果,i確定了6個(gè)相鄰的體素,Pi為第i個(gè)相鄰體素的壓力梯度,α值取0.5,當(dāng)平均壓力值變化小于0.1%時(shí)迭代結(jié)束,可以得到每個(gè)體素相對(duì)于參考點(diǎn)的壓力值。在肺動(dòng)脈干、右肺動(dòng)脈、左肺動(dòng)脈分別截取與分析肺動(dòng)脈血流相同的截面(圖3A),并計(jì)算相對(duì)壓力在截面上的均值,可得收縮期和舒張期平均相對(duì)壓力的峰值。
圖3 男,27歲。A.計(jì)算平均相對(duì)壓力截面位置;B~D分別為右肺動(dòng)脈、左肺動(dòng)脈、肺動(dòng)脈干各截面平均相對(duì)壓力結(jié)果。MPA:肺動(dòng)脈干;RPA:右肺動(dòng)脈;LPA:左肺動(dòng)脈
1.4 統(tǒng)計(jì)學(xué)方法 采用SPSS 16.0軟件,2次掃描測(cè)得的最大流速、最大流量、每搏輸出量和峰值壓力值的相關(guān)性采用Pearson相關(guān)分析,采用Bland-Altman分析法評(píng)價(jià)2次掃描獲得結(jié)果的差異和一致性,P<0.05表示差異有統(tǒng)計(jì)學(xué)意義。
本組8例受試者均成功完成2次掃描,并成功測(cè)定了肺動(dòng)脈流量及相對(duì)壓力,圖1B和圖3B~D分別給出了其中1例受試者1個(gè)心動(dòng)周期內(nèi)右肺動(dòng)脈、左肺動(dòng)脈、肺動(dòng)脈干的血流流量和平均相對(duì)壓力的變化曲線。2次掃描所得最大流速、最大流量、每搏輸出量和峰值壓力均有較好的相關(guān)性(r=0.880、0.988、0.958、0.953,P<0.001),見(jiàn)表1。Bland-Altman分析結(jié)果見(jiàn)表1及圖4,結(jié)果顯示2次掃描所得最大流速、最大流量、每搏輸出量和峰值壓力的結(jié)果具有較好的可重復(fù)性。
表1 2次掃描所得各參數(shù)的相關(guān)性和Bland-Altman分析結(jié)果
圖4 2次掃描所得各參數(shù)的Bland-Altman分析。A~D分別為最大流速、最大流量、每搏輸出量、峰值壓力的Bland-Altman分析結(jié)果
本研究通過(guò)健康志愿者肺動(dòng)脈4D-Flow成像獲得了流速、流量、每搏輸出量、相對(duì)壓力等血流動(dòng)力學(xué)參數(shù),并具有較高的穩(wěn)定性和可重復(fù)性。由最大流速、最大流量、每搏輸出量的相關(guān)性分析結(jié)果可見(jiàn),最大流量和每搏輸出量的相關(guān)性高于最大流速,其原因在于最大流速為分析截面中速度的最大值,相對(duì)于流量和每搏輸出量對(duì)于截面內(nèi)流速面積上和流量時(shí)間上的積分,容易產(chǎn)生誤差,但總體上3個(gè)參數(shù)均有較好的相關(guān)性和可重復(fù)性,可以應(yīng)用于相關(guān)臨床研究。
在相對(duì)壓力計(jì)算方面,本研究采用納維葉-斯托克斯(Navier-Stokes)方程計(jì)算壓力梯度,與簡(jiǎn)化的伯努利方程通過(guò)速度峰值計(jì)算相對(duì)壓力[11]相比,納維葉斯托克斯方程可提供血管內(nèi)相對(duì)壓力關(guān)于時(shí)間或空間的變化信息,而通過(guò)迭代壓力泊松方程法可以避免相對(duì)壓力計(jì)算結(jié)果受到積分路徑的影響[12],并且其準(zhǔn)確性已經(jīng)在水模上得到驗(yàn)證[13]。在計(jì)算壓力的過(guò)程中本研究采用了3D PC-MRA技術(shù)進(jìn)行血管的分割,只能提供血管內(nèi)流動(dòng)的區(qū)域范圍,并不能提供血管壁的確切位置,并且圖像噪聲及血液流動(dòng)的部分容積效應(yīng)使得速度場(chǎng)并非完全符合散度為0的條件,因此在計(jì)算血壓時(shí)并未考慮流體的邊界條件和散度為0的條件后續(xù)研究中,基于不同時(shí)間幀的分割以及利用散度為0條件對(duì)速度場(chǎng)進(jìn)行噪聲去除可能會(huì)解決上述問(wèn)題。
目前對(duì)于肺動(dòng)脈高壓相關(guān)血流動(dòng)力學(xué)參數(shù)的無(wú)創(chuàng)測(cè)量研究多集中在利用超聲[14]及二維相位對(duì)比技術(shù)盡管4D-Flow技術(shù)采集時(shí)間較長(zhǎng),但其定位較為簡(jiǎn)單可以在一次掃描中提供所掃描的三維容積內(nèi)任意位置及截面的血流速度信息,非常適合多截面或血管空間內(nèi)血流動(dòng)力學(xué)參數(shù)的計(jì)算和分析,并且可以提供包括向量、流線、跡線的三維可視化信息[15]。在后續(xù)研究中,可以利用4D-Flow技術(shù)對(duì)肺動(dòng)脈高壓患者進(jìn)行肺動(dòng)脈血流、相對(duì)壓力、三維可視化等多種血流動(dòng)力學(xué)參數(shù)的綜合評(píng)價(jià),并與右心導(dǎo)管術(shù)數(shù)據(jù)及健康志愿者結(jié)果進(jìn)行對(duì)比,進(jìn)行肺動(dòng)脈高壓的無(wú)創(chuàng)評(píng)估。4D-Flow技術(shù)也為主動(dòng)脈、顱內(nèi)動(dòng)脈、頸動(dòng)脈等相關(guān)血管疾病研究提供了新的研究手段[16],如先天性心臟?。?7]、主動(dòng)脈夾層[18]、顱內(nèi)動(dòng)脈瘤[19]等。
總之,本研究采用4D-Flow技術(shù),完成了對(duì)肺動(dòng)脈血流和相對(duì)壓力的測(cè)量和定量分析,并驗(yàn)證了測(cè)量的可重復(fù)性,有望應(yīng)用于多血流動(dòng)力學(xué)參數(shù)的肺動(dòng)脈高壓的評(píng)估。
參考文獻(xiàn)
[1] Shah SJ. Pulmonary hypertension. JAMA, 2012, 308(13): 1366 1374.
[2] Thenappan T, Shah SJ, Rich S, et al. Survival in pulmonary arterial hypertension: a reappraisal of the NIH risk stratification equation. Eur Respir J, 2010, 35(5): 1079-1087.
[3] Markl M, Harloff A, Bley TA, et al. Time-resolved 3D MR velocity mapping at 3T: improved navigator-gated assessment o vascular anatomy and blood flow. J Magn Reson Imaging, 2007
25(4): 824-831.
[4] Markl M, Chan FP, Alley MT, et al. Time-resolved threedimensional phase-contrast MRI. J Magn Reson Imaging, 2003, 17(4): 499-506.
[5] Frydrychowicz A, Stalder AF, Russe MF, et al. Threedimensional analysis of segmental wall shear stress in the aorta by flow-sensitive four-dimensional-MRI. J Magn Reson Imaging, 2009, 30(1): 77-84.
[6] Bock J, Kreher BW, Hennig J, et al. Optimized pre-processing of time-resolved 2D and 3D phase contrast MRI data//Proceedings of the 15th Annual Meeting of ISMRM, Berlin, 2007: 3138.
[7] Bock J, Frydrychowicz A, Stalder AF, et al. 4D phase contrast MRI at 3 T: effect of standard and blood-pool contrast agents on Snr, PC-MRA, and blood flow visualization. Magn Reson Med, 2010, 63(2): 330-338.
[8] Stalder AF, Russe MF, Frydrychowicz A, et al. Quantitative 2D and 3D phase contrast MRI: optimized analysis of blood flow and vessel wall parameters. Magn Reson Med, 2008, 60(5):1218-1231.
[9] Balleux-Buyens F, Jolivet O, Bittoun J, et al. Velocity encoding versus acceleration encoding for pressure gradient estimation in MR haemodynamic studies. Phys Med Biol, 2006, 51(19): 4747-4758.
[10] Yang GZ, Kilner PJ, Wood NB, et al. Computation of flow pressure fields from magnetic resonance velocity mapping. Magn Reson Med, 1996, 36(4): 520-526.
[11] Nogami M, Ohno Y, Koyama H, et al. Utility of phase contrast MR
imaging for assessment of pulmonary flow and pressure estimation in patients with pulmonary hypertension: comparison with right heart catheterization and echocardiography. J Magn Reson Imaging, 2009, 30(5): 973-980.
[12] Nagao T, Yoshida K, Okada K, et al. Development of a noninvasive method to measure intravascular and intracardiac pressure differences using magnetic resonance imaging. Magn Reson Med Sci, 2008, 7(3): 113-122.
[13] Tyszka JM, Laidlaw DH, Asa JW, et al. Three-dimensional, timeresolved (4D) relative pressure mapping using magnetic resonance imaging. J Magn Reson Imaging, 2000, 12(2): 321-329.
[14] 魏麗群, 李越, 王廣義, 等. 超聲心動(dòng)圖無(wú)創(chuàng)性評(píng)估肺血管阻力. 中國(guó)醫(yī)學(xué)影像學(xué)雜志, 2014, 22(4): 272-277.
[15] Markl M, Draney MT, Hope MD, et al. Time-resolved 3-dimensional velocity mapping in the thoracic aorta: visualization of 3-directional blood flow patterns in healthy volunteers and patients. J Comput Assist Tomogr, 2004, 28(4): 459-468.
[16] Pineda Zapata JA, Delgado de Bedout JA, Rascovsky Ramírez S, et al. A practical introduction to the hemodynamic analysis of the cardiovascular system with 4D Flow MRI. Radiologia, 2014, 56(6): 485-495.
[17] Fran?ois CJ, Srinivasan S, Schiebler ML, et al. 4D cardiovascular magnetic resonance velocity mapping of alterations of right heart flow patterns and main pulmonary artery hemodynamics in tetralogy of Fallot. J Cardiovasc Magn Reson, 2012, 14(3): 16.
[18] 郭子義, 陳晶. 動(dòng)態(tài)三維血管建模法分析Stanford B型胸主動(dòng)脈夾層四維相位對(duì)比MRI. 第二軍醫(yī)大學(xué)學(xué)報(bào), 2014, 35(6): 651-656.
[19] Xiang J, Natarajan SK, Tremmel M, et al. Hemodynamicmorphologic discriminants for intracranial aneurysm rupture. Stroke, 2011, 42(1): 144-152.
(本文編輯 張春輝)
作者單位清華大學(xué)生物醫(yī)學(xué)工程系 北京 100084
Doi:10.3969/j.issn.1005-5185.2016.03.018
通訊作者李 睿
基金項(xiàng)目
北京市科技計(jì)劃課題(Z131100005213001)。
中圖分類號(hào)R445.2
收稿日期:2015-07-10
Reproducibility of Pulmonary Artery Flow and Pressure Difference Measurements Using 4D-Flow MRI
【Abstract】Purpose to investigate the reproducibility of blood flow and pressure differenc measurements for pulmonary artery using 4D-Flow MRI.Materials and Methods Eigh healthy volunteers were scanned twice within a week using 4D-Flow MRI for pulmonary artery segment. Three 2D cross-sectional views including main pulmonary artery right pulmonary artery and left pulmonary artery were manually chosen to evaluat local hemodynamic parameters such as peak velocity, stroke volume, peak flow and mean pressure difference as well as the reproducibility between two measurements.Results Th evaluation of peak velocity, stroke volume, peak flow and mean pressure differenc were successfully implemented in all volunteers and a significant correlation between two scans was also found for every measured parameter with correlation coefficient of 0.880, 0.988, 0.958 and 0.953 (P<0.001) respectively. The Bland-Altman results furthe demonstrated good reproducibility in the measurements for peak velocity, stroke volume peak flow and mean pressure difference (d=1.30, 4.60, 1.00 and 0.04, σd=±8.90, ±29.30 ±9.20 and ±0.35).Conclusion Blood flow and PD can be quantitatively measured in pulmonary artery using 4D-Flow MRI with good reproducibility.
【Key words】Hypertension, pulmonary; Magnetic resonance imaging; Pulmonary artery Hemodynamics; Image processing, computer-assisted