【摘要】：目的：研究在非增強(qiáng)掃描中寶石能譜CT減輕肝臟高密度植入物引起的射線束硬化偽影最佳單能量，通過對(duì)基物質(zhì)對(duì)圖的分析辨別肝內(nèi)植入物的物質(zhì)。 資料與方法：選取肝癌術(shù)后的62個(gè)病例，進(jìn)行能譜掃描（Gestone spectralimaging,GSI),其中獲得碘油硬化偽影圖像99層，125I粒子金屬偽影圖像19層，鈦止血夾金屬偽影28層，所選層面以lOkeV為間距，在40keV~140keV間進(jìn)行圖像重建，共獲得11組不同能量水平的單能量圖像，分別為單能量圖像mono圖像和mono+MARs(Metalartifacts reducing system)圖像，同時(shí)獲得平均能量圖（QC圖）并重建6種基物質(zhì)對(duì)圖像。量化分析：以每一層為一個(gè)單元，在高密度植入物區(qū)邊緣硬化偽影明顯處測(cè)量并計(jì)算獲得偽影指數(shù)（artifact index， AI）AI=。測(cè)量同層腹壁皮下脂肪作為層面的背景噪聲。主觀分析：對(duì)各種基物質(zhì)對(duì)圖像進(jìn)行比較分析，并對(duì)物質(zhì)密度及物質(zhì)所在片區(qū)邊緣密度這兩個(gè)重要特征進(jìn)行評(píng)分，以區(qū)分碘油、金屬粒子與鈦止血夾金屬三種物質(zhì)在不同基物質(zhì)對(duì)圖中的差異。采用SPSS18.0對(duì)所獲得的全部數(shù)據(jù)進(jìn)行分析處理，，所有數(shù)據(jù)均進(jìn)行方差齊性檢驗(yàn)和正態(tài)性檢驗(yàn)，以P0.05為差異有統(tǒng)計(jì)學(xué)意義。 結(jié)果分析：80keV是降低碘油和鈦夾的硬化偽影的最佳單能量，而粒子的最佳單能量出現(xiàn)在70keV。各組物質(zhì)mono+MARs圖像和mono圖像的圖像噪聲在70keV下能到達(dá)最低。mono+MARs圖像和mono圖對(duì)比，除對(duì)125I粒子組AI值在40~90keV時(shí)有統(tǒng)計(jì)學(xué)差異外，其他均沒表現(xiàn)出統(tǒng)計(jì)學(xué)差異。與QC圖的比較分析中，單能量圖像與QC圖AI比較有明顯的統(tǒng)計(jì)學(xué)差異（P0.05）；而SDn值與QC圖比較沒有有明顯的統(tǒng)計(jì)學(xué)差異。 結(jié)合mono+MARs70keV圖像，對(duì)六組基物質(zhì)圖的比較分析及其物質(zhì)密度和物質(zhì)片區(qū)邊緣密度這兩個(gè)重要特征進(jìn)行評(píng)分，根據(jù)碘油、125I粒子及鈦夾在基物質(zhì)圖像中的不同表現(xiàn)，可在CT圖像中對(duì)不同基物質(zhì)進(jìn)行區(qū)分，并可在混合物質(zhì)圖像中將碘油和鈦夾，碘油和125I粒子進(jìn)行有效區(qū)分。 結(jié)論：單能量圖像與QC圖像相比較，它能有效減輕高密度物質(zhì)的硬化偽影；根據(jù)基物質(zhì)對(duì)圖像對(duì)比可以初步判斷物質(zhì)的類型，可以通過分析不同基物質(zhì)對(duì)圖像的的密度和物質(zhì)片區(qū)邊緣密度有效區(qū)分三種植入物。
[Abstract]:Aim: to study the optimal single energy of gemstone energy spectrum (CT) to reduce the X-ray beam hardening artifact induced by liver high density implants in non-contrast scanning, and to identify the implantable substance in liver by analyzing the image of the base substance. Materials and methods: 62 cases of liver cancer after operation were selected for energy spectrum scanning (Gestone spectralimaging,GSI). 99 images of iodized oil hardening artifacts, 19 images of 125i seed metal artifacts and 28 layers of metal artifacts of titanium hemostatic clips were obtained. The selected layers are spaced by lOkeV and reconstructed between 40keV~140keV. Eleven groups of single energy images with different energy levels are obtained, which are single energy image mono image and mono MARs (Metalartifacts reducing system) image, respectively. At the same time, the average energy map (QC) was obtained and the images of 6 base matter pairs were reconstructed. Quantitative analysis: using each layer as a unit, the artifact index (artifact index, AI) AI=. was obtained by measuring and calculating the hardening artifacts at the edge of the high density implant region. The subcutaneous fat of the same abdominal wall was measured as the background noise on the plane. Subjective analysis: to compare and analyze the images of various base substances, and to grade the two important characteristics of the density of the material and the density of the edge of the area in which the substance is located, in order to distinguish the iodized oil. Differences between metal particles and titanium hemostatic clamp metal in different base material pairs. All the data were analyzed by SPSS18.0, and all the data were tested by variance homogeneity test and normality test. The difference was statistically significant (P0.05). The results showed that 80keV was the best single energy to reduce the hardening artifact of iodized oil and titanium clip, and the optimal single energy of particles appeared at 70 Kev. The image noise of mono MARs images and mono images was the lowest in 70keV. The comparison between mono MARs images and mono images showed that there was no statistical difference in the AI value of 125i seed group except for the 40~90keV of 125i seed group. Compared with QC, the AI of single-energy image was significantly different from that of QC (P0.05), but there was no significant difference between SDN and QC (P0.05). Combined with mono MARs70keV images, the comparative analysis of six groups of basic material maps and their two important characteristics, material density and edge density of material slice, were evaluated. According to the different manifestations of iodized oil, 125i seeds and titanium clamped in the base material image, Different base substances can be distinguished in CT images, and iodized oil and titanium clamps, iodized oil and 125i seeds can be effectively distinguished in mixed material images. Conclusion: compared with QC images, single energy images can effectively reduce the hardening artifacts of high density substances. According to the comparison of the base material to the image, the type of the substance can be preliminarily judged. By analyzing the density of the image and the density of the edge of the material region, the three planting inputs can be effectively distinguished by the analysis of the density of the base material to the image.
【學(xué)位授予單位】：華南理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：R816.5

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 董誠;馮衛(wèi)華;李曉莉;李曉飛;曹會(huì)志;徐文堅(jiān);;能譜成像技術(shù)在去除碘對(duì)比劑硬化偽影方面的作用[J];青島大學(xué)醫(yī)學(xué)院學(xué)報(bào);2012年02期

2 周澤俊;胡永勝;高斌;巢惠民;;多層螺旋CT圖像偽影的分析[J];中國CT和MRI雜志;2008年05期

3 韓冰;王飛;李志民;劉煒煒;;碘125放射性粒子植入治療口腔頜面部惡性腫瘤的療效觀察[J];吉林醫(yī)學(xué);2010年31期

4 劉嶸;王建華;周康榮;嚴(yán)福華;顏志平;;原發(fā)性肝癌碘油栓塞治療后磁共振成像(MRI)隨訪研究[J];中國臨床醫(yī)學(xué);2006年04期

5 孫博;董越;方春曉;王兵;鄭邵微;;能譜CT基物質(zhì)成像技術(shù)定量測(cè)量活體甲狀腺碘含量[J];中國介入影像與治療學(xué);2013年02期

6 曾昭沖;;肝細(xì)胞肝癌放射治療現(xiàn)狀與展望[J];中華腫瘤防治雜志;2009年13期

7 曾昭沖;;肝細(xì)胞癌伴門靜脈/下腔靜脈癌栓放療現(xiàn)狀與展望[J];實(shí)用腫瘤雜志;2009年05期

8 毛學(xué)群;陳寶定;薛玉;任重陽;;多次肝動(dòng)脈化療栓塞術(shù)聯(lián)合射頻消融在肝癌治療中的臨床應(yīng)用[J];實(shí)用醫(yī)學(xué)影像雜志;2012年05期

9 劉蒙;董衛(wèi)國;;原發(fā)性肝癌非手術(shù)治療的研究進(jìn)展[J];世界華人消化雜志;2009年12期

10 程路得;劉斌;吳興旺;汪潔;王樂;王萬勤;鮑麗君;劉咸英;張帥;;能譜CT單能成像對(duì)乏血供肝癌的診斷價(jià)值[J];安徽醫(yī)科大學(xué)學(xué)報(bào);2012年01期

本文編號(hào)：2448906