胡婷丹，彭衛(wèi)軍，童 彤

復(fù)旦大學(xué)附屬腫瘤醫(yī)院放射診斷科，復(fù)旦大學(xué)上海醫(yī)學(xué)院腫瘤學(xué)系，上海 200032

·綜述·

紋理分析在結(jié)直腸癌診治中的應(yīng)用進(jìn)展

胡婷丹，彭衛(wèi)軍，童 彤

復(fù)旦大學(xué)附屬腫瘤醫(yī)院放射診斷科，復(fù)旦大學(xué)上海醫(yī)學(xué)院腫瘤學(xué)系，上海 200032

結(jié)直腸癌是常見惡性腫瘤，也是癌癥相關(guān)死亡的第二大原因，其早期診斷和腫瘤分期對(duì)治療方式的選擇有重要意義。紋理分析是一種新型影像分析技術(shù)，通過(guò)紋理分析對(duì)影像學(xué)圖像進(jìn)行處理，能預(yù)測(cè)腫瘤侵襲程度、微小轉(zhuǎn)移灶和治療反應(yīng)性等信息，有助于腫瘤的早期診斷、分期和預(yù)后評(píng)估。該文綜述了紋理分析在結(jié)直腸癌中的應(yīng)用進(jìn)展。

結(jié)直腸癌；紋理分析；治療反應(yīng)

結(jié)直腸癌(colorectal cancer，CRC)居常見惡性腫瘤第3位、腫瘤相關(guān)死亡第2位[1]。其確診依靠病理，影像學(xué)手段主要起提示診斷、進(jìn)行治療前評(píng)估及監(jiān)測(cè)預(yù)后的作用。傳統(tǒng)影像學(xué)檢查在反映腫瘤形態(tài)、大小、部位、侵襲程度、遠(yuǎn)處轉(zhuǎn)移等方面有一定價(jià)值，但對(duì)腫瘤的精確評(píng)估價(jià)值有限。近年來(lái)，越來(lái)越多的研究表明，紋理分析(texture analysis，TA)通過(guò)分析影像圖像中像素或體素灰度的分布和聯(lián)系，深度挖掘其細(xì)微結(jié)構(gòu)和變化規(guī)律，能更精確地評(píng)估腫瘤異質(zhì)性[2-4]、內(nèi)在侵襲性和治療抗性。因此，紋理分析有望成為更有效的腫瘤評(píng)估方式，指導(dǎo)臨床選擇合適治療方案，有助于謀求患者效益最大化。

1 紋理分析

紋理分析是指通過(guò)對(duì)傳統(tǒng)影像圖像進(jìn)行后處理，對(duì)圖像像素的強(qiáng)度和空間分布特點(diǎn)進(jìn)行數(shù)學(xué)分析與運(yùn)算，從而量化紋理參數(shù)的一種方法[5]。其圖像來(lái)源主要為CT、MRI和一些后處理圖像等：基于平掃CT的紋理分析主要反映腫瘤內(nèi)部結(jié)構(gòu)特點(diǎn)及壞死、出血、囊變等密度的改變；基于增強(qiáng)CT的紋理分析能進(jìn)一步反映造影劑在血管內(nèi)外分布的非均質(zhì)性；基于MRI圖像的紋理分析具有更高的對(duì)比度和分辨率，可減少圖像噪聲對(duì)紋理參數(shù)的影響[6]，從而更準(zhǔn)確地反映紋理特征。

1.1 紋理分析方法

常用的紋理分析方法主要分為四大類：統(tǒng)計(jì)分析法、結(jié)構(gòu)分析法、模型分析法和頻譜分析法[7]。統(tǒng)計(jì)分析法根據(jù)圖像像素的強(qiáng)度和分布規(guī)律來(lái)反映圖像紋理特征，常用的有灰度共生矩陣法和灰度-梯度共生矩陣法、長(zhǎng)游程法等；結(jié)構(gòu)分析法通過(guò)提取紋理內(nèi)部空間排列和位置關(guān)系來(lái)獲取結(jié)構(gòu)信息，可反映組織空間結(jié)構(gòu)及周期性規(guī)律；模型分析法將圖像中的每一個(gè)像素都看作與其相鄰像素存在某種函數(shù)關(guān)系，進(jìn)而從整體估算所有像素的空間位置關(guān)系；頻譜分析法常采用Laplacian of Gaussian過(guò)濾技術(shù)來(lái)處理圖像，將過(guò)濾條件調(diào)整至合適范圍，然后經(jīng)變換提取出紋理特征。

1.2 紋理參數(shù)的意義

紋理分析中，描述灰度變化規(guī)律的數(shù)字特征稱為圖像的紋理特征，用紋理參數(shù)來(lái)量化，常用的紋理參數(shù)主要有熵(entropy)、均勻性(uniformity)、對(duì)比度(contrast)、標(biāo)準(zhǔn)差(standard deviation，SD)、峰度(kurtosis)、偏度(skewness)等[8]。峰度和偏度是紋理的一階統(tǒng)計(jì)參數(shù)，主要描述灰度直方圖的分布特征；熵反映圖像紋理的復(fù)雜度和混亂度，若紋理灰度分布隨機(jī)則熵值較大；均勻性反映紋理的規(guī)則程度，紋理雜亂無(wú)章時(shí)均勻性較低；對(duì)比度反映圖像的清晰度和紋理溝紋的深淺程度，紋理溝紋深，對(duì)比度大。

2 紋理分析在結(jié)直腸癌分期中的應(yīng)用

2.1 評(píng)估腫瘤T分期

研究表明，與根治性手術(shù)相比，對(duì)局部晚期直腸癌(pT3-4期和(或)pN1-2期)患者行新輔助治療(neoadjuvant therapy，NAT)后手術(shù)可降低50%～61%的局部復(fù)發(fā)風(fēng)險(xiǎn)[9]。然而，NAT也有骨髓抑制、藥物性肝損傷等嚴(yán)重不良反應(yīng)[10-11]。因此，準(zhǔn)確的分期有助于患者選擇正確的治療方式，避免不必要的藥物毒性反應(yīng)。Liu等[10]研究表明，基于表觀擴(kuò)散系數(shù)(apparent diffusion coefficient，ADC)圖的紋理分析可評(píng)估腫瘤T分期。該研究發(fā)現(xiàn)，與pT3-4期患者相比，pT1-2期患者圖像的偏度、熵和對(duì)比度明顯偏低，其中偏度和熵是腫瘤外侵的獨(dú)立預(yù)測(cè)因子。理論上，偏度可反映感興趣區(qū)(region of interest，ROI)像素灰度強(qiáng)度分布的不對(duì)稱性，熵可反映像素的空間分布情況，較高的偏度和熵反映了ROI紋理的復(fù)雜性，提示病變區(qū)域異質(zhì)性增加。因此，與pT1-2期相比，pT3-4期腫瘤具有更復(fù)雜的紋理特征。

2.2 診斷淋巴結(jié)轉(zhuǎn)移

淋巴結(jié)轉(zhuǎn)移對(duì)腫瘤分期及治療方式的選擇非常重要，同時(shí)也是判斷預(yù)后的重要指標(biāo)。影像學(xué)檢查?；诹馨徒Y(jié)的形態(tài)學(xué)特征來(lái)鑒別其良惡性。但一項(xiàng)大樣本研究指出，轉(zhuǎn)移也可發(fā)生于正常大小形態(tài)的淋巴結(jié)中，常規(guī)測(cè)量短軸直徑不能作為淋巴結(jié)受累的可靠指標(biāo)[9]。Langman等[12]也認(rèn)為，影像學(xué)檢查對(duì)腫大的非轉(zhuǎn)移性淋巴結(jié)的鑒別和小淋巴結(jié)中微轉(zhuǎn)移灶的檢出有很大局限性。紋理分析為淋巴結(jié)轉(zhuǎn)移的確診提供了新的方法。Liu等[10]應(yīng)用結(jié)直腸癌原發(fā)灶的紋理特征來(lái)評(píng)估淋巴結(jié)狀態(tài)，發(fā)現(xiàn)熵在無(wú)淋巴結(jié)累及(pN0)與淋巴結(jié)受累(pN1-2)的腫瘤之間差異有統(tǒng)計(jì)學(xué)意義。熵作為淋巴結(jié)轉(zhuǎn)移的獨(dú)立預(yù)測(cè)因子，其區(qū)分pN0期與pN1-2期患者受試者工作特征(receiver operating characteristic，ROC)曲線的曲線下面積(area under curve，AUC)為0.751，靈敏度和特異度分別為96.2%、45.3%。另一項(xiàng)針對(duì)220例淋巴結(jié)轉(zhuǎn)移的結(jié)直腸癌患者的研究表明，良惡性淋巴結(jié)的CT紋理特征不同，惡性淋巴結(jié)的分?jǐn)?shù)維度較良性淋巴結(jié)高，據(jù)此判斷淋巴結(jié)轉(zhuǎn)移的準(zhǔn)確率達(dá)88%[13]。因此，對(duì)原發(fā)腫瘤及淋巴結(jié)紋理屬性的評(píng)估有助于預(yù)測(cè)淋巴結(jié)轉(zhuǎn)移情況。

2.3 檢測(cè)肝臟微小轉(zhuǎn)移灶

肝臟是結(jié)直腸癌患者最常發(fā)生轉(zhuǎn)移的器官。有研究表明，結(jié)直腸癌患者同時(shí)性肝轉(zhuǎn)移的發(fā)生率約20%[1]，隨訪1年后延遲性肝轉(zhuǎn)移發(fā)生率為4.3%，而隨訪5年后發(fā)生率為14.5%[14]。有文獻(xiàn)指出，58%～81%的結(jié)直腸癌肝轉(zhuǎn)移患者在肝臟手術(shù)切除標(biāo)本中可見微轉(zhuǎn)移灶[15]，因此有學(xué)者認(rèn)為延遲性肝轉(zhuǎn)移可能是由肝內(nèi)已存在的隱匿性轉(zhuǎn)移發(fā)展而來(lái)[16]。相關(guān)研究表明，紋理分析有望發(fā)現(xiàn)形態(tài)學(xué)尚不可見的微轉(zhuǎn)移灶，從而指導(dǎo)后續(xù)治療。Rao等[17]對(duì)29例結(jié)直腸癌患者肝臟CT圖像的非病灶區(qū)進(jìn)行紋理分析，發(fā)現(xiàn)無(wú)轉(zhuǎn)移灶、同時(shí)性肝轉(zhuǎn)移和延遲性肝轉(zhuǎn)移3組病例之間紋理參數(shù)存在差異，當(dāng)微轉(zhuǎn)移灶存在于無(wú)明顯轉(zhuǎn)移的肝組織中，會(huì)引起肝臟血流動(dòng)力學(xué)發(fā)生改變，導(dǎo)致更高的空間異質(zhì)性[18]，從而在紋理參數(shù)上與正常肝組織區(qū)分。但該結(jié)論尚需大樣本研究的支持。

3 紋理分析在結(jié)直腸癌臨床療效和評(píng)估生存中的應(yīng)用

3.1 預(yù)測(cè)原發(fā)腫瘤的治療反應(yīng)

NAT后行全直腸切除術(shù)是局部晚期直腸癌的首選治療方法[19]，可有效減少?gòu)?fù)發(fā)率[20]，但也會(huì)導(dǎo)致對(duì)治療部分反應(yīng)(partial response，PR)或無(wú)反應(yīng)(non-response，NR)患者的過(guò)度治療[23]，因此有必要對(duì)治療反應(yīng)進(jìn)行準(zhǔn)確和早期的評(píng)估。De等[22]研究表明，MRI圖像的紋理參數(shù)可評(píng)估NAT治療反應(yīng)：NAT治療前，病理完全反應(yīng)(pathological complete response，pCR)組患者峰度比PR+NR組高，NAT治療后pCR組峰度較PR+NR組低；pCR組患者NAT治療后峰度下降程度明顯較PR+NR組高。另一項(xiàng)研究[23]也發(fā)現(xiàn)，與其他參數(shù)相比，峰度是預(yù)測(cè)pCR的最有效參數(shù)，靈敏度和特異度分別為100%和67% (AUC=0.861，P=0.001)。這與理論結(jié)果一致，即具有較強(qiáng)治療抗性和高侵襲性的腫瘤具有較高的異質(zhì)性，因此紋理參數(shù)能有效量化腫瘤異質(zhì)性，有望成為預(yù)測(cè)NAT治療反應(yīng)的生物成像標(biāo)記。

3.2 預(yù)測(cè)肝轉(zhuǎn)移灶的治療反應(yīng)

對(duì)于結(jié)直腸癌肝轉(zhuǎn)移患者，手術(shù)切除是實(shí)現(xiàn)長(zhǎng)期生存的最有效方法[24]。對(duì)暫時(shí)無(wú)法手術(shù)的患者可先行系統(tǒng)化療，使轉(zhuǎn)移灶體積縮小，分期下降后再行根治性手術(shù)[25]。臨床研究表明，化療反應(yīng)良好者行根治性切除后可獲得長(zhǎng)期生存，但如果術(shù)前化療不敏感，肝轉(zhuǎn)移灶切除后患者預(yù)后也不理想[26]，因此有必要在術(shù)前正確預(yù)測(cè)治療反應(yīng)。以前對(duì)治療反應(yīng)的評(píng)估主要根據(jù)腫瘤退縮分級(jí)標(biāo)準(zhǔn)(tumor regression grading，TRG)，有研究表明紋理分析可提供相關(guān)信息。Rao等[27]研究發(fā)現(xiàn)，熵和均勻性是區(qū)分化療后良好反應(yīng)(TRG 1～2)與不良反應(yīng)(TRG 3～5)的重要預(yù)測(cè)因素；Ahn等[28]發(fā)現(xiàn)，較低的偏度和較小的標(biāo)準(zhǔn)差是治療反應(yīng)有效的獨(dú)立預(yù)測(cè)因子，在驗(yàn)證隊(duì)列中也顯示出良好的預(yù)測(cè)價(jià)值(AUC=0.797)。理論上，對(duì)治療反應(yīng)良好的轉(zhuǎn)移病灶常被壞死和纖維組織代替，具有更均質(zhì)的內(nèi)部結(jié)構(gòu)；而不良反應(yīng)病灶中仍有大量存活的腫瘤細(xì)胞，導(dǎo)致病灶仍有較大異質(zhì)性。因此，紋理分析有望評(píng)估轉(zhuǎn)移灶對(duì)化療的敏感性，從而指導(dǎo)治療方案的正確選擇。

3.3 評(píng)估放化療后總體生存

紋理分析還可被用于預(yù)測(cè)NAT治療后患者的總體生存(overall survival，OS)。Ng等[29]發(fā)現(xiàn)，熵、均勻性、峰度、偏度和標(biāo)準(zhǔn)偏差均可作為總體生存率的獨(dú)立預(yù)測(cè)因子。另一項(xiàng)基于MRI圖像的紋理分析也發(fā)現(xiàn)了類似結(jié)果，該研究發(fā)現(xiàn)平均陽(yáng)性像素值、平均強(qiáng)度均可預(yù)測(cè)總體生存和無(wú)病生存(disease free survival，DFS)，峰度可獨(dú)立預(yù)測(cè)無(wú)復(fù)發(fā)生存(recurrence free survival，RFS)[30]。Miles等[31]的研究也表明，均勻性是長(zhǎng)期生存的最佳預(yù)測(cè)參數(shù)。此外，Ganeshan等[32]研究發(fā)現(xiàn)，對(duì)結(jié)直腸癌肝轉(zhuǎn)移患者，紋理分析也顯示出預(yù)測(cè)生存的潛力，熵和均勻性與CRC肝轉(zhuǎn)移患者的總體生存有關(guān)，在熵＜2.0時(shí)準(zhǔn)確鑒別了4例于36個(gè)月內(nèi)死亡的患者，靈敏度和特異度分別為100%和65%。這些研究均表明，紋理參數(shù)有助于評(píng)估結(jié)直腸癌患者的預(yù)后和生存。

4 紋理分析的缺陷

目前，紋理分析處于發(fā)展階段，其應(yīng)用仍存在較多問(wèn)題：① 多種干擾因素對(duì)紋理參數(shù)的影響無(wú)法準(zhǔn)確評(píng)估，如由胰島素抵抗引起肝臟脂肪含量發(fā)生的微小變化[33]、放化療導(dǎo)致腫瘤組織的纖維化和壞死程度[20]、原發(fā)病灶存在與否對(duì)血流動(dòng)力學(xué)的改變等，因此需嚴(yán)格控制實(shí)驗(yàn)條件來(lái)獲得準(zhǔn)確的數(shù)據(jù)。② 選擇最大層面(2D)還是對(duì)腫瘤整體(3D)進(jìn)行紋理分析仍值得探討，有學(xué)者認(rèn)為3D比2D分析更具代表性[34]，也有研究表明2D與3D分析的紋理參數(shù)并無(wú)統(tǒng)計(jì)學(xué)差異[35]。這可能是由于選取的ROI和所研究的類型不同所致，也提示不同腫瘤之間存在生物學(xué)差異，還需對(duì)不同器官的不同腫瘤類型進(jìn)一步分析。③ 由于管電壓和管電流均會(huì)對(duì)CT圖像的采集產(chǎn)生不同程度的影響[39]，選取的過(guò)濾器不同也會(huì)使紋理參數(shù)的P值發(fā)生改變[22]，故有必要確定統(tǒng)一的采集參數(shù)和合適的過(guò)濾模式，從而使紋理分析更廣泛地應(yīng)用于臨床實(shí)踐。

5 展望

總之，越來(lái)越多的研究證據(jù)表明，紋理分析可用于判斷結(jié)直腸癌患者的預(yù)后、評(píng)估遠(yuǎn)處轉(zhuǎn)移、預(yù)測(cè)生存等，并有望作為治療反應(yīng)的新的生物成像標(biāo)記來(lái)指導(dǎo)臨床治療。隨著研究的進(jìn)一步深入，紋理分析將在結(jié)直腸癌的應(yīng)用中發(fā)揮更加積極的作用。

[1] VAN DER POOL A E M, DAMHUIS R A, IJZERMANS J N M, et al. Trends in incidence, treatment and survival of patients with stage IV colorectal cancer: apopulation-based series [J]. Colorec Dis, 2012, 14(1): 56-61.

[2] GANESHAN B, MILES K A. Quantifying tumour heterogeneity with CT [J]. Cancer Imaging, 2013, 13(1): 140-149.

[3] ROLLAND Y, BEZY-WENDLING J, GESTIN H, et al. Analysis of texture in medical imaging. Review of the literature [J]. Ann Radiol (Paris), 1995, 38(6): 315-347.

[4] GANESHAN B, GOH V, MANDEVILLE H C, et al. Non-small cell lung cancer: histopathologic correlates for texture parameters at CT [J]. Radiology, 2013, 266(1): 326-336.

[5] TOMITA F, TSUJI S. Statistical Texture Analysis [M]. Springer, 1990, 13-36.

[6] MILES K A, GANESHAN B, HAYBALL M P. CT texture analysis using the filtration-histogram method: what do the measurements mean? [J]. Cancer Imaging, 2013, 13(3): 400-406.

[7] VAN GIJN W, MARIJNEN C A, NAGTEGAAL I D, et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer: 12-year follow-up of the multicentre, randomised controlled TME trial [J]. Lancet Oncol, 2011, 12(6): 575-582.

[8] BIRGISSON H, PAHLMAN L, GUNNARSSON U, et al. Adverse effects of preoperative radiation therapy for rectal cancer: long-term follow-up of the Swedish Rectal Cancer Trial [J]. J Clin Oncol, 2005, 23(34): 8697-8705.

[9] FERNáNDEZ-MARTOS C, PERICAY C, APARICIO J, et al. Phase II, randomized study of concomitant chemoradiotherapy followed by surgery and adjuvant capecitabine plus oxaliplatin (CAPOX) compared with induction CAPOX followed by concomitant chemoradiotherapy and surgery in magnetic resonance imaging—defined, locally advanced rectal cancer: Grupo Cáncer de Recto 3 Study [J]. J Clin Oncol, 2010, 28(5): 859-865.

[10] LIU L, LIU Y, XU L, et al. Application of texture analysis based on apparent diffusion coefficient maps in discriminating different stages of rectal cancer [J]. J Magn Reson Imaging, 2017, 45(6): 1798-1808.

[11] DWORAK O. Morphology of lymph nodes in the resected rectum of patients with rectal carcinoma [J]. Pathol Res Pract, 1991, 187(8):1 020-1024.

[12] LANGMAN G, PATEL A, BOWLEY D M. Size and distribution of lymph nodes in rectal cancer resection specimens [J]. Dis Colon Rectum, 2015, 58(4): 406-414.

[13] CUI C, CAI H, LIU L, et al. Quantitative analysis and prediction of regional lymph node status in rectal cancer based on computed tomography imaging [J]. Eur Radiol, 2011, 21(11): 2318-2325.

[14] MANFREDI S, LEPAGE C M, HATEM C, et al. Epidemiology and management of liver metastases from colorectal cancer [J]. Ann Surg, 2006, 244(2): 254-259.

[15] WAKAI T, SHIRAI Y, SAKATA J, et al. Histologic evaluation of intrahepatic micrometastases in patients treated with or without neoadjuvant chemotherapy for colorectal carcinoma liver metastasis [J]. Int J Clin Exp Pathol, 2012, 5(4): 308-314.

[16] LEEN E. The detection of occult liver metastases of colorectal carcinoma [J]. J Hepato-Biliary-Pancreatic Surg, 1999, 6(1): 7-15.

[17] RAO S X, LAMBREGTS D M, SCHNERR R S, et al. Whole-liver CT texture analysis in colorectal cancer: Does the presence of liver metastases affect the texture of the remaining liver? [J]. United European Gastroenterol J, 2014, 2(6): 530-538.

[18] GANESHAN B, MILES K A, YOUNG R C, et al. In search of biologic correlates for liver texture on portalphase CT [J]. Acad Radiol, 2007, 14(9): 1058-1068.

[19] KAISER A M, KLARISTENFELD D, BEART R W. Preoperative versus postoperative radiotherapy for rectal cancer in a decision analysis and outcome prediction model [J]. Ann Surg Oncol, 2012, 19(13): 4150-4160.

[20] SAUER R, LIERSCH T, MERKEL S, et al. Preoperative versus postoperative chemoradiotherapy for locally advanced rectal cancer: results of the German CAO/ARO/AIO-94 randomized phase III trial after a median follow-up of 11 years [J]. J Clin Oncol, 2012, 30(16): 1926-1933.

[21] HARTLEY A, HO K F, MCCONKEY C, et al. Pathological complete response following pre-operative chemoradiotherapy in rectal cancer: analysis of phase II/III trials [J]. Br J Radiol, 2005, 78(934): 934-938.

[22] DE CECCO C N, GANESHAN B, CIOLINA M, et al. Texture analysis as imaging biomarker of tumoral response to neoadjuvant chemoradiotherapy in rectal cancer patients studied with 3-T magnetic resonance [J]. Invest Radiol, 2015, 50(4): 239-245.

[23] DE CECCO C N, CIOLINA M, CARUSO D, et al. Performance of diffusion-weighted imaging, perfusion imaging, and texture analysis in predicting tumoral response to neoadjuvant chemoradiotherapy in rectal cancer patients studied with 3T MR: initial experience [J]. Abdom Radiol (NY), 2016, 41(9): 1728-1735.

[24] CHOTI M A, SITZMANN J V, TIBURI M F, et al. Trends in long-term survival following liver resection forhepatic colorectal metastases [J]. Ann Surg, 2002, 235(6): 759-766.

[25] BELGAUMKAR A P, LOW N, RIGA A T, et al. Chemotherapy before liver resection of colorectal metastases: friend or foe? [J]. Ann Surg, 2015, 261(2): e36.

[26] BROUQUET A, BENOIST S, JULIE C, et al. Risk factors for chemotherapy-associated liver injuries: A multivariate analysis of a group of 146 patients with colorectal metastases [J]. Surgery, 2009, 145(4): 362-371.

[27] RAO S, LAMBREGTS D M, SCHNERR R S, et al. CT texture analysis in colorectal liver metastases: A better way than size and volume measurements to assess response to chemotherapy? [J]. United European Gastroenterol J, 2016, 4(2): 257-263.

[28] AHN S J, KIM J H, PARK S J, et al. Prediction of the therapeutic response after FOLFOX and FOLFIRI treatment for patients with liver metastasis from colorectal cancer using computerized CT texture analysis [J]. Eur J Radiol, 2016, 85(10): 1867-1874.

[29] NG F, KOZARSKI R, GANESHAN B, et al. Assessment of primary colorectal cancer heterogeneity by using whole-tumor texture analysis: contrast-enhanced CT texture as a biomarker of 5-year survival [J]. Radiology, 2013, 266(1): 177-184.

[30] JALIL O, AFAQ A, GANESHAN B, et al. Magnetic resonance-based texture parameters as potential imaging biomarkers for predicting long term survival in locally advanced rectal cancer treated by chemoradiotherapy [J]. Colorectal Dis, 2017, 19(4): 349-362.

[31] MILES K A, GANESHAN B, GRIFFITHS M R, et al. Colorectal cancer: texture analysis of portal phase hepatic CT images as a potential marker of survival [J]. Radiology, 2009, 250(2): 444.

[32] GANESHAN B, MILES K A, YOUNG R C D, et al. Hepatic enhancement in colorectal cancer [J]. Acad Radiol, 2007, 14(12): 1520-1530.

[33] GANESHAN B, MILES K A, YOUNG R C, et al. In search of biologic correlates for liver texture on portalphase CT [J]. Acad Radiol, 2007, 14(9): 1058-1068.

[34] NG F, KOZARSKI R, GANESHAN B, et al. Assessment of tumor heterogeneity by CT texture analysis: Can the largest cross-sectional area be used as an alternative to whole tumor analysis? [J]. Eur J Radiol, 2013, 82(2): 342-348.

[35] LUBNER M G, STABO N, LUBNER S J, et al. CT textural analysis of hepatic metastatic colorectal cancer: pre-treatment tumor heterogeneity correlates with pathology and clinical outcomes [J]. Abdom Imaging, 2015, 40(7): 2331-2337.

[36] MILES K A, YOUNG H, CHICA S L, et al. Quantitative contrast-enhanced computed tomography: is there a need for system calibration? [J]. Eur Radiol, 2007, 17(4): 919-926.

Progress on texture analysis in diagnosis and treatment of colorectal cancer

HU Tingdan, PENG Weijun, TONG Tong
(Department of Diagnostic Radiology, Fudan University Shanghai Cancer Center; Department of Oncology,Shanghai Medical College, Fudan University, Shanghai 200032, China)

TONG Tong E-mail: t983352@126.com

Colorectal cancer is a common malignant tumor, and it is also the second leading cause of cancer-related death.Early diagnosis and tumor staging are of great importance for the choice of treatment. Texture analysis, as a new type of image analysis technique, can be used to predict the degree of tumor invasion, micrometastasis and therapeutic response thus contributing to the early diagnosis, staging and assessment of prognosis. This article summarizes the application of texture analysis in colorectal cancer.

Colorectal cancer; Texture analysis; Therapeutic response

R445.2; R445.3

A

1008-617X(2017)04-0306-05

2017-06-17

2017-08-19)

國(guó)家自然科學(xué)青年基金(No：81501437)

童彤 E-mail：t983352@126.com