小視野高分辨DWI對胰腺實性病變的鑒別診斷價值

李晶 馬超 邊云 王馨蕊 史張 王莉 邵成偉 陳士躍 陸建平

小視野高分辨DWI對胰腺實性病變的鑒別診斷價值

李晶 馬超 邊云 王馨蕊 史張 王莉 邵成偉 陳士躍 陸建平

目的探討小視野高分辨擴散加權成像(rFOV DWI)對胰腺實性占位的鑒別診斷價值。方法收集139例胰腺實性占位患者，其中胰腺導管腺癌(PDAC)105例，神經內分泌腫瘤(NET)16例，腫塊型慢性胰腺炎(MFCP)7例，實性假乳頭狀瘤(SPT)11例。招募38名健康成年志愿者作為對照組。行包括單次激發平面回波成像(ss-EPI)DWI、rFOV DWI(b值為0和600 s/mm2)的MRI檢查。采用四分法從解剖結構的可視性、胰腺病灶對比度、運動及磁敏感偽影3方面評估rFOV DWI及ss-EPI DWI圖像質量，通過工作站自帶軟件測量感興趣區(ROI)的表觀擴散系數(ADC)值。比較兩種DWI的圖像質量及ADC值在各胰腺疾病及正常胰腺間的差異。繪制ADC 值的受試者工作特征(ROC)曲線，評價PDAC與其他胰腺良性腫塊及正常胰腺的差異。結果b值為0和600 s/mm2的rFOV DWI在顯示胰腺解剖結構、病灶對比度、偽影評分均優于ss-EPI DWI(b=0 s/mm2時為2.99±0.51比2.79±0.64、2.37±0.48比1.81±0.63、3.17±0.56比2.91±0.60；b=600 s/mm2時為3.63±0.50比3.32±0.56、3.45±0.50比3.01±0.49、3.74±0.44比3.12±0.37)，差異均有統計學意義(P值均<0.001)。PDAC、NET、MFCP、SPT、正常胰腺rFOV DWI獲得的ADC值分別為(1.38±0.17)×10-3、(1.22±0.35)×10-3、(1.29±0.13)×10-3、(1.04±0.38)×10-3、(1.86±0.15)×10-3mm2/s；ss-EPI DWI的ADC值分別為(1.73±0.24)×10-3、(1.63±0.39)×10-3、(1.58±0.19)×10-3、(1.25±0.26)×10-3、(2.04±0.20)×10-3mm2/s，各組間的差異及同組內兩ADC值間的差異均有統計學意義(P值均<0.001)。MFCP與PDAC、NET與MFCP、MFCP與SPT間rFOW DWI的ADC值的差異及MFCP與PDAC、PDAC與NET、SPT與MFCP間ss-EPI DWI的ADC值差異無統計學意義，而其他兩兩組間差異均有統計學意義(P值均<0.05)。PADC與正常胰腺rFOV DWI和ss-EPI DWI的ROC曲線下面積(AUC)分別為0.983(95%CI0.944～0.998)和0.889(95%CI0.822～0.936)，差異有統計學意義(P=0.0004)，而PDAC與所有良實性病變的rFOV DWI和ss-EPI DWI的ADC值的AUC分別為0.799(95%CI0.719～0.864)和0.755(95%CI0.672～0.827)，差異無統計學意義。結論rFOV DWI顯著提高DWI圖像質量，并且對胰腺導管腺癌的診斷效能更佳。

胰腺腫瘤； 磁共振成像； 診斷，鑒別

磁共振擴散加權成像(diffusion-weighted imaging, DWI)通過定量計算出表觀擴散系數(apparent diffusion coefficient, ADC)從而反映組織內水分子擴散快慢。臨床DWI應用中使用最為廣泛的是單次激發平面回波成像(single-shot echo-planar-imaging, ss-EPI)序列，其具有成像速度快、對運動不敏感等優點，但與常規MRI結構圖像(如T1、T2圖像)相比，DWI因受到偏共振和磁敏感效應的影響而導致圖像分辨率偏低。小視野(reduced field-of-view, rFOV)技術可以實現高分辨DWI，最初應用于脊髓成像[1]。有研究報道，rFOV在胰腺成像中能提高DWI圖像質量并減少偽影，但研究的樣本量較小且未見rFOV DWI對不同類型胰腺實性占位診斷的應用價值報道[2-4]。本研究通過rFOV DWI與ss-EPI DWI圖像質量及ADC值的對比研究，明確rFOV DWI 在胰腺實性占位診斷中的應用價值。

資料與方法

一、病例資料

研究為前瞻性設計，經長海醫院倫理委員會批準，所有參與者均簽署知情同意書。收集2014年2月至2016年1月間經CT或者B超檢查懷疑胰腺實性占位的住院患者139例。所有患者檢查前均未行放療或化療，并在長海醫院行MR檢查后1周內手術治療，有明確的病理檢查結果。同期招募38名健康志愿者作為對照組。

139例胰腺實性占位患者中男性86例，女性53例，平均年齡(57±12)歲。其中胰腺導管腺癌(pancreatic ductal adenocarcinomas，PDAC)105例，男性69例，女性36例，平均年齡(61±9)歲；神經內分泌腫瘤(neuroendocrine tumor，NET)16例，男、女各8例，平均年齡(51±8)歲；腫塊型慢性胰腺炎(mass-forming chronic pancreatitis，MFCP)7例，均為男性，平均年齡(51±9)歲；實性假乳頭狀瘤(solid pseudopapillary tumor，SPT)11例，男性2例，女性9例，平均年齡(32±9)歲。38名健康志愿者中男性26名，女性12名，年齡(49±11)歲。

二、MRI掃描方法

MRI檢查均在同一臺3-T磁共振儀上完成。所有受檢者均于檢查前禁食6 h以上并接受呼吸訓練。被檢者采取仰臥位、腳先進體位，行包括胰腺橫斷T1加權像(T1WI)、T2加權像(T2WI)、磁共振胰膽管造影(MRCP)、動態增強、ss-EPI DWI、rFOV DWI在內的胰腺MRI掃描。橫斷面呼吸觸發T2WI序列：重復時間(TR)6 316 ms，回波時間(TE)72 ms，層厚5 mm，視野(FOV)380 mm×400 mm，矩陣320×192；肝臟容積加速采集(liver acquisition with volume acceleration，LAVA)序列：TR 2.5 ms，TE 1.1 ms，層厚5 mm，FOV 440 mm×418 mm，矩陣256×180；厚層塊2D-MRCP序列：TR 7 000 ms，TE 1 221 ms，層厚54 mm，FOV 300 mm×300 mm，矩陣288×288。ss-EPI WDI序列：擴散梯度因子(b)值為0、600 s/mm2，TR 6 000 ms，TE 58.6 ms，層厚 5 mm，FOV 380 mm×400 mm，矩陣128×96；rFOV DWI序列：b值為0、600 s/mm2，TR 5 000 ms，TE 58.6 ms，層厚5 mm，FOV 160 mm×80 mm，矩陣128×64。rFOV DWI掃描層數為6層?；颊呓M最后還要經靜脈注射釓噴酸葡胺注射液0.2～0.3 ml/kg體重行動態增強掃描。

三、圖像分析及數據測量

1．DWI圖像質量評估：由一名腹部疾病診斷經驗豐富的放射科醫師對b值為0和600 s/mm2的DWI圖像質量進行評分。四分法標準如下：(1)解剖結構的可視性：1分，肉眼無法觀察解剖結構致圖像無法用于診斷；2分，大體顯示解剖結構但胰腺邊緣模糊；3分，較清晰顯示解剖結構且胰腺邊緣清晰；4分，胰管結構清晰可見。(2) 胰腺病灶對比度：1分，病灶未見顯示；2分，病灶與周圍組織有微弱對比；3分，病灶與周圍組織有中等對比度或有強對比但病灶邊緣模糊；4分，病灶與周圍組織有明顯的對比度且病灶邊緣清晰。(3)運動及磁敏感偽影：1分，偽影嚴重致圖像無法用于診斷；2分，中度偽影；3分，輕度偽影；4分，無偽影。3個分值之和為DWI圖像質量總分[5]。

2．ADC值：ADC值的測量在工作站自帶軟件(Function 6.3.1e, GE AW VolumeShare 2,GE Healthcare, Milwaukee,WI, USA)上進行。數據測量時避開胰膽管、血管及胰腺邊緣來選取圓形或橢圓形感興趣區(ROI)測量ADC平均值。ROI的面積分別為20～287 mm2(ss-EPI DWI)和20～238 mm2(rFOV DWI)。

四、統計學分析

結 果

一、圖像質量分析

b值為0和600 s/mm2的rFOV DWI在顯示胰腺解剖結構、病灶對比度均優于ss-EPI DWI，且偽影也明顯少于后者，差異具有統計學意義(表1，圖1、圖2)。

二、ADC定量分析

PDAC、NET、MFCP、SPT、正常胰腺rFOV DWI獲得的ADC值分別為(1.38±0.17)×10-3、(1.22±0.35)×10-3、(1.29±0.13)×10-3、(1.04±0.38)×10-3、(1.86±0.15)×10-3mm2/s；ss-EPI DWI的ADC值分別為(1.73±0.24)×10-3、(1.63±0.39)×10-3、(1.58±0.19)×10-3、(1.25±0.26)×10-3、(2.04±0.20)×10-3mm2/s，各組間的差異均有統計學意義(P值均<0.001)。此外，同組內rFOV DWI獲得的ADC值均顯著小于ss-EPI DWI的ADC值，差異均有統計學意義(P值均<0.001)。兩兩組比較時， rFOV DWI的ADC值在MFCP與PDAC(P=0.078)、NET與MFCP(P=0.937)、SPT與MFCP(P=0.09)的組間差異以及ss-EPI DWI的ADC值在MFCP與PDAC(P=0.104)、PDAC與NET(P=0.133)、SPT與MFCP(P=0.285)的組間差異無統計學意義，而其他兩兩組間差異均有統計學意義(P值均<0.05)。

表1 rFOV DWI與ss-EPI DWI圖像質量評分(分，

圖1 胰頭部中分化導管腺癌。1A為ss-EPI DWI(b=600 s/mm2)，1B為rFOV DWI(b=600 s/mm2)，1C為軸位T2WI，1D為軸位T1WI。rFOV DWI對病灶的顯示邊緣結構清晰，信號顯示較ss-EPIDWI均勻

圖2 胰尾部慢性腫塊型胰腺炎。2A為ss-EPI DWI(b=600 s/mm2)，2B為rFOV DWI(b=600 s/mm2)，2C為軸位T2WI，2D為軸位T1WI。rFOV DWI對病灶的顯示邊緣結構清晰，信號顯示較ss-EPIDWI均勻

PDAC與正常胰腺的rFOV DWI和ss-EPI DWI的ADC值的ROC曲線下面積(AUC)分別為0.983(95%CI0.944～0.998)和0.889(95%CI0.822～0.936)，差異有統計學意義(P=0.0004，圖3A)。

將NET、MFCP、SPT合并為良實性病變組，PDAC與良實性病變組的rFOV DWI和ss-EPI DWI的ADC值的AUC 分別為0.799(95%CI0.719～0.864)和0.755(95%CI0.672～0.827)，差異無統計學意義(P=0.4355，圖3B)。

討 論

影響DWI分辨率的因素有兩個，一個是FOV，另一個是采集矩陣。ss-EPI DWI使用的ss-EPI序列TE時間長且在相位編碼方向的帶寬窄，這有可能會產生磁敏感偽影和化學位移偽影，從而導致圖像質量欠佳、信噪比較低以及空間分辨率較差。對于ss-EPI DWI序列來說，采集矩陣的大小直接決定了整體讀出窗口的長度，采集矩陣越大，整體讀出窗口的長度越長，ss-EPI DWI序列的偽影也越重，因此通過增加矩陣提高ss-EPI DWI分辨率是行不通的。此外，可以通過減小FOV實現圖像的分辨率的提高，因為FOV小于成像物體的面積時，圖像在相位編碼的方向會產生卷褶偽影，所以ss-EPI DWI的FOV不能隨意縮小。為保證FOV之外的物體不被激發或激發之后的信號對圖像沒有影響，本研究使用的rFOV DWI利用一種2D選擇性激勵的射頻脈沖來激發小范圍的感興趣區以避免卷褶偽影的干擾，且隨著FOV的增大，b=0 s/mm2時磁場的不均勻性也會越明顯，所以rFOV DWI脂肪壓制的效果優于ss-EPI DWI，有助于提高小病灶的檢出率[1,6-8]。本研究采用的rFOV DWI的空間分辨率是ss-EPI DWI的4倍以上，可同時明顯改善DWI圖像的變形和偽影。

圖3 ROC曲線分析。3A為rFOV DWI和ss-EPI DWI對PDAC與正常胰腺的鑒別診斷ROC分析；3B為rFOV DWI和ss-EPI DWI對PDAC與胰腺良性病變的鑒別診斷ROC分析

腹部DWI檢查中b值常取值在500～1 000 s/mm2[9-13]。本研究在胰腺DWI成像使用b值為600 s/mm2可以保持較好的信噪比和圖像質量[9]?；趕s-EPI DWI的研究可以有效地鑒別胰腺良惡性病變，因胰腺癌的ADC值顯著低于良性病變[10-18]。本研究結果也顯示胰腺癌的ADC值顯著低于胰腺良性病變，其主要原因是胰腺癌基質纖維化含量、增多的腫瘤細胞限制了細胞外間隙水分子的運動，而腫瘤細胞內核質比的增加也限制了細胞內水分子的擴散運動[1]。但本研究的ss-EPI DWI及rFOV DWI的ADC值對MFCP與PDAC的無明顯鑒別診斷價值，與Philipp等[19]報道結果一致。其可能原因一方面是MFCP入組例數少，另一方面是MFCP富含纖維化基質，且其有可能進展為胰腺癌。此外，PDAC的腫瘤組織內及其周圍可因為主胰管或分支胰管的阻塞而伴發胰腺炎也可導致MFCP與PDAC的ADC值存在一定程度的重疊[20]。Fattahi等[15]的研究(b為600 s/mm2)則認為ADC值可以鑒別這兩種疾病。也有研究認為高b值的DWI可以有效地鑒別診斷MFCP與PDAC[21]。目前ADC值對兩者的鑒別診斷價值尚存在爭議。

rFOV DWI較ss-EPI DWI對PADC的診斷準確性高的可能原因是rFOV DWI的體素體積(7.8 mm3)小于ss-EPI DWI的體素體積(61.8 mm3)，從而有效減少腫瘤組織內各成分之間的部分容積效應，其ADC值更能反映病變的本質。

本研究尚存在一些不足。首先，本研究選取了相對較低的b值(600 s/mm2)來減少運動偽影、提高信噪比，但更高的b值才有可能更準確地反映組織的真實擴散[22-23]。此外，兩個b值(0和600 s/mm2)的DWI雖然可減少檢查時間，但多b值DWI成像可能得到更加準確的ADC值[24-25]。

總之，rFOV DWI對胰腺解剖結構的顯示、對病灶的顯示對比度明顯優于ss-EPI DWI，其鑒別診斷胰腺導管腺癌與正常胰腺的效能更高。

[1] Zaharchuk G, Saritas EU, Andre JB, et al. Reduced field-of-view diffusion imaging of the human spinal cord: comparison with conventional single-shot echo-planar imaging[J]. AJNR Am J Neuroradiol, 2011, 32(5): 813-820. DOI: 10.3174/ajnr.A2418.

[2] Riffel P, Michaely HJ, Morelli JN, et al. Zoomed EPI-DWI of the pancreas using two-dimensional spatially-selective radiofrequency excitation pulses[J]. PLoS One, 2014, 9(3): e89468. DOI: 10.1371/journal.pone.0089468.

[3] Ma C, Li YJ, Pan CS, et al. High resolution diffusion weighted magnetic resonance imaging of the pancreas using reduced field of view single-shot echo-planar imaging at 3T[J]. Magn Reson Imaging, 2014, 32(2): 125-131. DOI: 10.1016/j.mri.2013.10.005.

[4] Kim H, Lee JM, Yoon JH, et al. Reduced field-of-view diffusion-weighted magnetic resonance imaging of the pancreas: comparison with conventional single-shot echo-planar imaging[J]. Korean J Radiol, 2015,16(6):1216-1225. DOI: 10.3348/kjr.2015.16.6.1216.

[5] Kartalis N, Loizou L, Edsborg N, et al. Optimising diffusion-weighted MR imaging for demonstrating pancreatic cancer: a comparison of respiratory-triggered, free-breathing and breath-hold techniques[J]. Eur Radiol, 2012, 22(10): 2186-2192. DOI: 10.1007/s00330-012-2469-3.

[6] Wilm BJ, Svensson J, Henning A, et al. Reduced field-of-view MRI using outer volume suppression for spinal cord diffusion imaging[J]. Magn Reson Med, 2007, 57(3): 625-630. DOI: 10.1002/mrm.21167.

[7] Jeong EK, Kim SE, Guo J, et al. High-resolution DTI with 2D interleaved multislice reduced FOV single-shot diffusion-weighted EPI (2D ss-rFOV-DWEPI)[J]. Magn Reson Med, 2005. 54(6): 1575-1579. DOI: 10.1002/mrm.20711.

[8] Karampinos DC, Van AT, Olivero WC, et al. High resolution reduced-FOV diffusion tensor imaging of the human pons with multi-shot variable density spiral at 3T[J]. Conf Proc IEEE Eng Med Biol Soc, 2008, 2008: 5761-5764. DOI: 10.1109/IEMBS.2008.4650523.

[9] Koc Z, Erbay G. Optimal b value in diffusion-weighted imaging for differentiation of abdominal lesions[J]. J Magn Reson Imaging, 2014, 40(3): 559-5566. DOI: 10.1002/jmri.24403.

[10] Giannarini G, Petralia G, Thoeny HC. Potential and limitations of diffusion-weighted magnetic resonance imaging in kidney, prostate, and bladder cancer including pelvic lymph node staging: a critical analysis of the literature[J]. Eur Urol, 2012, 61(2): 326-340. DOI: 10.1016/j.eururo.2011.09.019.

[11] Kartalis N, Lindholm TL, Aspelin P, et al. Diffusion-weighted magnetic resonance imaging of pancreas tumours[J]. Eur Radiol, 2009, 19(8): 1981-1990. DOI: 10.1007/s00330-009-1384-8.

[12] Matsuki M, Inada Y, Nakai G, et al. Diffusion-weighed MR imaging of pancreatic carcinoma[J]. Abdom Imaging, 2007, 32(4): 481-483. DOI: 10.1007/s00261-007-9192-6.

[13] Lee SS, Byun JH, Park BJ, et al. Quantitative analysis of diffusion-weighted magnetic resonance imaging of the pancreas: usefulness in characterizing solid pancreatic masses[J]. J Magn Reson Imaging, 2008, 28(4): 928-936. DOI: 10.1007/s00330-012-2687-8.

[14] Muraoka N, Uematsu H, Kimura H, et al. Apparent diffusion coefficient in pancreatic cancer: characterization and histopathological correlations[J]. J Magn Reson Imaging, 2008, 27(6): 1302-1308. DOI: 10.1002/jmri.21340.

[15] Fattahi R, Balci NC, Perman WH, et al. Pancreatic diffusion-weighted imaging (DWI): comparison between mass-forming focal pancreatitis (FP), pancreatic cancer (PC), and normal pancreas[J]. J Magn Reson Imaging, 2009, 29(2): 350-356. DOI: 10.1002/jmri.21651.

[16] Fukukura Y, Takumi K, Kamimura K, et al. Pancreatic adenocarcinoma: variability of diffusion-weighted MR imaging findings[J]. Radiology, 2012, 263(3): 732-740. DOI: 10.1148/radiol.12111222.

[17] Wiggermann P, Grützmann R, Weissenb?ck A,et al. Apparent diffusion coefficient measurements of the pancreas, pancreas carcinoma, and mass-forming focal pancreatitis[J]. Acta Radiol, 2012, 53(2): 135-9. DOI: 10.1258/ar.2011.100252.

[18] Rosenkrantz AB, Matza BW, Sabach A, et al. Pancreatic cancer: lack of association between apparent diffusion coefficient values and adverse pathological features[J]. Clin Radiol, 2013, 68(4): e191-e197. DOI: 10.1016/j.crad.2012.11.006.

[19] Philipp W, Robert G, Angelika W, et al. Apparent diffusion coefficient measurements of the pancreas, pancreas carcinoma and mass-forming focal pancreatitis[J]. Acta Radiol, 2012: 1-5. DOI: 10.1258/ar.2011.100252. DOI: 10.1258/ar.2011.100252.

[20] Buetow PC, Parrino TV, Buck JL, et al. Islet cell tumors of the pancreas: pathologic-imaging correlation among size, necrosis and cysts, calcification, malignant behavior, and functional status[J]. Am J Roentgenol, 1995, 165(5): 1175-1179. DOI: 10.2214/ajr.165.5.7572498.

[21] Takeuchi M, Matsuzaki K, Kubo H, et al. High-b-value diffusion-weighted magnetic resonance imaging of pancreatic cancer and mass-forming chronic pancreatitis: preliminary results[J]. Acta Radiol, 2008, 49(4): 383-386. DOI: 10.1080/02841850801895381.

[22] Ichikawa T, Erturk SM, Motosugi U, et al. High-b value diffusion-weighted MRI for detecting pancreatic adenocarcinoma: preliminary results[J]. Am J Roentgenol, 2007, 188(2): 409-414. DOI: 10.2214/AJR.05.1918.

[23] Huang WC, Sheng J, Chen SY, et al. Differentiation between pancreatic carcinoma and mass-forming chronic pancreatitis: usefulness of high b value diffusion-weighted imaging[J]. J Dig Dis, 2011, 12(5): 401-408. DOI: 10.1111/j.1751-2980.2011.00517.x.

[24] Klauss M, Lemke A, Grünberg K, et al. Intravoxel incoherent motion MRI for the differentiation between mass forming chronic pancreatitis and pancreatic carcinoma[J]. Invest Radiol, 2011, 46(1): 57-63. DOI: 10.1097/RLI.0b013e3181fb3bf2.

[25] Lemke A, Laun FB, Klauss M, et al. Differentiation of pancreas carcinoma from healthy pancreatic tissue using multiple b-values: comparison of apparent diffusion coefficient and intravoxel incoherent motion derived parameters[J]. Invest Radiol, 2009, 44(12): 769-775. DOI:10.1097/RLI.0b013e3181b62271.

ValueofreducedfieldofviewDWIindifferentiatingsolidpancreaticfocallesions

LiJing,MaChao,BianYun,WangXinrui,ShiZhang,WangLi,ShaoChengwei,ChenShiyue,LuJianping.

DepartmentofRadiology,ChanghaiHospital,SecondMilitaryMedicalUniversity,Shanghai200433,China

LuJianping,Email:cjr.lujianping@vip.163.com

ObjectiveTo study the value of reduced field-of-view (rFOV DWI) in differentiating patients with solid pancreatic focal lesions.Methods139 patients with solid pancreatic mass were enrolled, including 105 patients with pancreatic ductal acinar carcinoma (PDAC), 16 patients with neuroendocrine neoplasms, 7 patients with mass forming chronic pancreatitis (MFCP) and 11 patients with solid papillary tumor (SPT). 38 healthy adult volunteers served as controls, and underwent single stimulated echo planar imaging (ss-EPI)DWI and rFOV DWI(b value=0 and 600 s/mm2)MRI examination. Quartation method was used to evaluate the image quality of ss-EPI)DWI and rFOV DWI in the three terms of the visibility of anatomical structure, contrast of pancreatic lesions, motion and the susceptibility artifacts during MRI. Work station self-carried software was used to measure the ADC value of the region of interest (ROI).The image quality and ADC values of different pancreatic diseases and normal pancreas were compared. ROC curve for ADC value was drawn to evaluate the difference among PDAC, other benign pancreatic masses and normal pancreas.ResultsAt b value of 0 and 600 s/mm2, rFOV DWI was superior to ss-EPI DWI in terms of showing pancreatic anatomic structure, the contrast of the lesion and the score evaluation for susceptibility artifacts(b=0 s/mm22.99±0.51vs2.79±0.64, 2.37±0.48vs1.81±0.63, 3.17±0.56vs2.91±0.60；b=600 s/mm23.63±0.50vs3.32±0.56, 3.45±0.50vs3.01±0.49, 3.74±0.44vs3.12±0.37), and the differences were statistically significant (P<0.001). ADC values of PDAC, NET, MFCP, SPT and normal pancreas were (1.38±0.17)×10-3,(1.22±0.35)×10-3,(1.29±0.13)×10-3,(1.04±0.38)×10-3and(1.86±0.15)×10-3mm2/s for rFOV DWI, and(1.73±0.24)×10-3,(1.63±0.39)×10-3,(1.58±0.19)×10-3,(1.25±0.26)×10-3and(2.04±0.20)×10-3mm2/s for ss-EPI DWI. The difference on ADC values among different groups and within one group were all statistically significant (P<0.001). There were no statistical significant differences on ADC values between MFCP and PDAC, between MFCP and SPT as well as on ss-EPI DWI ADC values between PDAC and NET, but statistical differences were found between other two groups (P<0.05). The area under the ROC curve of rFOV and ss-EPI DWI was 0.983 (95%CI0.944-0.998) and 0.889 (95%CI0.822-0.936), respectively, and the difference was statistically significant (P=0.0004), but rFOV DWI and ss-EPI DWI ADC values for PDAC and all benign solid diseases were 0.799 (95%CI0.719-0.864) and 0.755 (95%CI0.672-0.827), and the difference was not statistically significant.ConclusionsrFOV DWI could significantly enhance the quality of DWI images, and its diagnostic efficacy was much better than ss-EPI DWI.

Pancreatic neoplasms; Magnetic resonance imaging; Diagnosis, differential

FundprogramShanghai Natural Science Foundation(14ZR1408300)；Medical Guidance Project of Shanghai Science and Technology Commission(14411960100)

10.3760/cma.j.issn.1674-1935.2017.06.009

200433 上海，第二軍醫大學長海醫院放射科

陸建平，Email：cjr.lujianping@vip.163.com

上海市自然科學基金(14ZR1408300)；上海市科委醫學引導項目(14411960100)

2017-06-18)

呂芳萍)