本文關(guān)鍵詞： 磁共振成像 肝細胞癌分化程度 IDEAL-IQ序列 R2*值 脂肪分數(shù)值　出處：《大連醫(yī)科大學》2017年碩士論文　論文類型：學位論文

【摘要】：目的探討磁共振IDEAL-IQ序列的R2*值及脂肪分數(shù)值評估原發(fā)性肝細胞癌分化程度的價值。方法回顧性分析2016年1月至2017年1月期間于我院行肝臟MRI檢查,并經(jīng)手術(shù)或穿刺病理證實為原發(fā)性HCC,且行MR檢查前未進行任何治療的患者,MR檢查需包含IDEAL—IQ序列。最終入組患者20例,共22個病灶,其中4個病灶經(jīng)穿刺活檢證實,18個病灶行手術(shù)切除。男16例,女4例,平均年齡58.7歲。IDEAL—IQ掃描參數(shù)為:翻轉(zhuǎn)角3°;回波時間分別為0.9、1.6、2.4、3.1、3.9、4.7、5.4、6.2、7.0、7.7、8.5 及 9.3ms,重復時間 6.9ms,帶寬 200kHz,視野 40×32cm,層厚6mm,層間距1mm,掃描層數(shù)24層。掃描結(jié)束后在操作平臺進行圖像重建,獲得T2*water圖、T2*fat圖、R2*弛豫率圖、脂肪分數(shù)圖四組圖像。將IDEAL-IQ重建圖像傳至AW4.5工作站,經(jīng)compare軟件進行測量。ROI放置于HCC病灶內(nèi)實質(zhì)成分處,另在距病灶3cm以上增強掃描無明顯占位肝實質(zhì)內(nèi)放置等大ROI,并避開肝內(nèi)大血管及膽管。實驗數(shù)據(jù)由兩名MRI診斷經(jīng)驗3年的影像醫(yī)生,采用雙盲法測量各ROI的R2*值及脂肪分數(shù)值。22個HCC灶的病理學分級按照根據(jù)Edmondson-Steiner分級法分為高、中、低分化組。根據(jù)病理學或影像學檢查確定肝背景是否有肝硬化,分為肝硬化組和非肝硬化組。采用SPSS 21.0統(tǒng)計分析軟件進行統(tǒng)計學分析。采用組內(nèi)相關(guān)系數(shù)檢驗兩位觀察者所測得各參數(shù)一致性。HCC與其肝背景各值比較,采用獨立樣本t檢驗,p0.05,認為差異具有統(tǒng)計學意義。HCC的R2*值及脂肪分數(shù)值與病理學分化程度相關(guān)性,采用Spearman相關(guān)分析。HCC不同病理級別組間的R2*值、脂肪分數(shù)值差異性采用ANOVA的LSD法檢驗。采用ROC曲線評價R2*值及脂肪分數(shù)值預估以肝背景為基準各病理學分化程度HCC的診斷效能。結(jié)果HCC各病理學分化組、不同肝背景組腫瘤瘤體及肝背景R2*值和脂肪分數(shù),兩名觀察者測量結(jié)果一致性良好。HCC與其肝背景R2*值差異均具有統(tǒng)計學意義,(p0.05),病灶內(nèi)R2*值均低于其肝背景。HCC病灶內(nèi)與肝背景脂肪分數(shù)無統(tǒng)計學意義。HCC的R2*值與其病理學分化程度間呈中度相關(guān)(p0.05,rs=0.453)。隨著HCC分化程度減低,其R2*值程增高趨勢。HCC高、中和低分化組三組間總體R2*值及脂肪分數(shù)值差異具有統(tǒng)計學意義。應用脂肪分數(shù)值預估高分化HCC曲線下面積(AUC)為0.856,以脂肪分數(shù)值=6.05%為界,敏感度為66.7%,特異度為100%。結(jié)論IDEAL-IQ序列獲得的脂肪分數(shù)值可作為判定高分化HCC的定量指標。IDEAL-IQ測得R2*值與HCC的病理學分化程度,呈中度相關(guān)。
[Abstract]:Objective to evaluate the value of R2 * value and fat fraction of IDEAL-IQ sequence in evaluating the differentiation of primary hepatocellular carcinoma (HCC). Methods MRI examination of liver was performed in our hospital from January 2016 to January 2017. IDEAL-IQ sequence should be included in Mr examination of the patients who were proved to be primary HCC by operation or biopsy and without any treatment before Mr examination. Finally, 20 patients (22 lesions) were included in the group. Among them, 4 lesions were confirmed by puncture biopsy, 18 lesions were surgically resected. The average age was 58.7 years old. IDEAL-IQ scan parameters were as follows: turnover angle 3 擄, echo time 0.91.6U 2.4C 3.1C 3.9U 4.75.4ng 7.78.5 and 9.3msrespectively, repetition time 6.9ms, bandwidth 200kHz, visual field 40 脳 32cm, layer thickness 6mm, interval between layers 1mm. Image reconstruction was carried out on the operating platform after scanning. Four sets of images were obtained from T2water and T2FAT-R2 * relaxation rate and fat fraction images. The reconstructed images were transferred to AW4.5 workstation, and measured by compare software. ROI was placed in the solid components of HCC lesions. In addition, the enhancement scan more than 3 cm away from the lesion had no significant space occupying in the hepatic parenchyma and avoided the large vessels and bile ducts in the liver. The experimental data were obtained from two imaging doctors with 3 years' experience in the diagnosis of MRI. The R2 * value and fat score of each ROI were measured by double blind method. The pathological grades of 22 HCC foci were divided into high, middle and low differentiation groups according to Edmondson-Steiner classification. It was divided into two groups: cirrhosis group and non-cirrhosis group. The statistical analysis was carried out with SPSS 21. 0 software. The consistency of parameters measured by two observers was tested by intragroup correlation coefficient, and the liver background values were compared with each other. The independent sample t test (p0.05) showed that the difference was statistically significant. The R2 * value and fat fraction of HCC were correlated with the degree of pathological differentiation. The R2 * value of different pathological grades of HCC was analyzed by Spearman correlation analysis. The difference of fat content was tested by LSD method of ANOVA. R2 * value and fat fraction value were used to evaluate the diagnostic efficacy of HCC based on liver background. Results all pathological differentiation groups of HCC were evaluated. R2 * value and fat fraction of tumor body and liver background in different liver background groups, The difference of R2 * between HCC and its liver background was statistically significant. The R2 * value in the lesion was lower than that in the liver background. There was no significant difference between the fat fraction in the HCC lesion and the liver background. The R2 * value of HCC and its disease was not statistically significant. The degree of differentiation was moderately correlated with the degree of differentiation of HCC. The R2 * value of HCC was higher than that of low differentiation group. The difference of total R2 * value and fat fraction between the three groups was statistically significant. The area under the well-differentiated HCC curve predicted by fat fraction value was 0.856, and the fat fraction value was 6.05%. The sensitivity is 66.7 and the specificity is 1000.Conclusion the fat fraction obtained by IDEAL-IQ sequence can be used as a quantitative index to judge highly differentiated HCC. The R2 * value measured by IDEAL-IQ is moderately correlated with the degree of pathological differentiation of HCC.
【學位授予單位】：大連醫(yī)科大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：R735.7;R445.2
，

本文編號：1547407