Study Of DWI Of Pancreas At 3-T And The Values Of IVIM DWI Erived Quantitative Parameters In Diagnosing Solid Pancreatic Focal Lesions

Chapter 1:Comparisons of DWI techniques and methods of ADC measurements of pancreas at 3-TPart 1:Comparisons of image quality and ADCs in breath-hold,respiratory-triggered and free-breathing DWI of pancreas at 3-TObjective:To compare image quality and apparent diffusion coefficients (ADC) of the normal pancreas parenchymas in breath-hold, respiratory-triggered and free-breathing diffusion weighted imaging (DWI) at 3.0-Tesla.Methods:DWI of the pancreas was performed at 3.0-Tesla in 21 healthy volunteers with breath-hold, respiratory-triggered and free-breathing using b-values of 0 and 500 s/mm2. For all three sequences, two readers assigned an image quality score to images at bo and b500, and two independent readers measured ADCs for the head, body and tail of pancreas. Image quality scores and ADCs of pancreas in the three DWIs were compared.Results:For bo, image quality scores was not significantly different among the three sequences (p= 0.103). For b500, image quality score was significantly lower in free-breathing DWI than breath-hold or respiratory-triggered DWI (P<0.001), and not significantly different between breath-hold and respiratory-triggered DWI (P= 0.212). Mean ADCs differed significantly among the anatomical regions with the lowest values measured in the pancreatic tail both at breath-hold and respiratory-triggered DWIs whereas no significant difference was found at free-breathing DWI.Conclusion:Breath-hold or respiratory-triggered technique provided DW images of pancreas with acceptable quality at 3.0-Tesla. Breath-hold is the preferred DWI technique for ADC measurements of pancreas.Part 2:DWI of pancreatic adenocarcinoma at 3-T:the effects of ROI methods on ADC valuesObjective:To investigate three region of interest (ROI) methods determined apparent diffusion coefficient (ADC) measurements in patients with pancreatic ductal adenocarcinoma (PDAC) and determine the optimum ADC measurement method for the differentiation of PDAC from normal pancreas.Methods:Twenty-one patients with surgical pathology-proven PDAC and eighteen healthy volunteers were included. Respiratory triggered single-shot echo-planar DWI (b-values= 0,600 s/mm2) was employed to calculate the ADC-maps across all participants. Two readers independently measured the ADCs according to three ROI methods: whole-volume, single-slice and small solid samples of tumor. Mean ADCs for the healthy pancreas were calculated using three measurements from pancreatic head to tail. The inter-observer variability for the three methods was compared using the interclass correlation coefficient (ICC). The mean ADCs among ROI methods were compared, and compared with normal pancreas as well. The diagnostic performances were calculated and compared by using the receiver operating characteristic curves (ROC).Results:ICCs were 0.928 for whole-volume ROIs,0.877 for single-slice ROIs,0.397 for small solid sample ROIs and 0.707 for normal pancreas. All of the ADCs calculated from the three ROI methods of PDAC were significantly lower than the normal pancreas. ADCs of solid tumor samples were significant lower than that calculated from whole-volume or single slice (both P< 0.001). Areas under the ROC curve for the identification of PDAC, based on small solid samples, single-slice and whole-volume ROIs, respectively, were 0.939,0.791 and 0.735.Conclusion:ADC based on the small solid samples of tumor provided the highest diagnostic performance in assessing PDAC and was more accurate than ADCs measured from single-slice or whole-volume ROI.Part 3:Effects of ROI sizes on pancreatic ADC values and the diagnostic accuracies of ADC values and signal intensity of DWI for PDACObjective:To investigate the effect of different ROI sizes on pancreatic ADC values, and compare the diagnostic performances of signal intensity on DWI and ADC for differentiating PDAC from normal pancreas.Methods:DWI datas of sisty-three patients with surgical pathology-proven PDAC and eighteen healthy volunteers were retrospective studied. SIs on DW images (both bo and b600) and ADCs of tumors and normal pancreas were measured by two observers independently using two different ROI size. The inter-observer variability for the SI andADC measurements was compared using the intraclass correlation coefficient (ICC). Mean SIs and ADCs between the two ROI size were compared. Using two independent sample t-test for the comparison of SI and ADC values between normal pancreas and PDAC groups. The diagnostic performances were calculated and compared by using the receiver operating characteristic (ROC) curves.Results:ICCs for all parameters were higher than 0.950. Paired sample t-test revealed Sis and ADCs differed significantly among the two groups of different ROI sizes, however, ICCs for all parameters between the two groups were higher than 0.980. Both SIb6oo and SIbo of PDAC calculated from the two different ROI sizes were significantly higher than the normal pancreas, while the mean ADCs were significantly lower than the normal pancreas. In addition, all the areas under the ROC curve for the identification of PDAC based on SIs were significantly higher than ADCs.Conclusion:ROI size had no obvious effect on the measured results of pancreatic DWI parameters. Both SIbo and SIb6oo provided higher diagnostic performance in differentiating PDAC from normal pancreas than ADC at 3-Tesla.Chapter 2:Multiple-b-value DWI of solid pancreatic focal lesions:the values of parameters based on monoexponential and biexponential modelPart 1:Multiple-b-value DWI of the healthy pancreas:the effects of age and gender on the DWI derived quantitative parametersObjective:To compare the multiple-b-value DWI derived quantitative parameters based on monoexponential and biexponential model in different anatomical regions of the healthy adult pancreas, and investigate the effect of gender and age on the diffusion parameters.Methods:Fifty-seven healthy volunteers (36 male,21 female, range from 21 to 68 years of age) were recruited. DWI of the pancreas was performed with 9 b-values (0,20,50,100, 200,400,600,800 and 1000 s/mm2, respectively). The ADC was calculated for all b-values using linear regression yielding ADCtotal-The ADCb, value of the monoexponential DWI, slow component of diffusion (ADCslow), incoherent microcirculation (ADCfast) and perfusion fraction (f) of the biexponential DWI were calculated for the pancreas head, body and tail. Dependency of the parameters on the anatomical regions was analyzed using Friedman test. Effects of gender on the parameters were analyzed using the independent sample t-test. Relationships of the four parameters on age were characterized using a Spearman rank-order correlation test.Results:All of the mean ADC400, ADC600, ADC800, ADC1000, ADCtotal and f values differed significantly among the anatomical regions with the lowest values were observed in the tail of pancreas (P< 0.05). The Friedman test results demonstrated a significant decline of the mean ADC values of the monoexponential DWI from b2o to biooo for the three anatomical regions respectively (P< 0.001). The mean global pancreatic ADCtotal, ADCfast, ADCslow and/values in female group (1.45 ± 0.15,13.17±4.91,0.88±0.18 X 10-3 mm2/s and 0.370 ± 0.063) were lower than that in male group (1.55 ± 0.17,15.77±7.83, 0.91±0.15×10-3 mm2/s and 0.372 ± 0.056), but the independent sample t-tests showed no significant differences Call of P> 0.05) between the two groups in all of the four parameters. Across the age spectrum, it was shown that there were significant correlation between the average ADCtotal, ADCfast and ADCslow values with age (r=-0.315, P= 0.017; r=-0.340, P= 0.010 and r=-0.300, P= 0.023, respectively); and no significant difference across the age in mean/values was observed (r= 0.048, P= 0.725).Conclusion:Multi-b-value DWI derived quantitative parameters including ADC400, ADC600, ADC800, ADC1000, ADCtotal and/differed significantly among the pancreatic head, body and tail, with the lowest values obtained in the tail. Multi-b-value DWI derived parameters including ADCtotal, ADCfast, ADCslow and/of normal adult pancreas showed no significant difference between male and female. Significant correlations between ADCtotal, ADCfast, ADCslow and age were observed, however, there was no significant difference across the age for mean f value.Part 2:Multiple-b-value DWI of pancreas:the values of IVIM DWI derived quantitative parameters and signal intensity in diagnosing solid pancreatic focal lesionsObjective:To evaluate the diagnostic potential of signal intensity (SI) and multiple-b-value DWI derived quantitative parameters for differentiation of solid pancreatic focal lesions.Methods:Methods:Eighty-five patients with surgical pathology-proven pancreatic focal lesions [66 pancreatic adenocarcinomas (PAC),13 neuroendocrine tumors (NET),6 chronic pancreatitis (CP)] and 23 healthy volunteers were included in this study. DWI of the pancreas was performed with 10 b-values (0,25,50,75,100,150,200,400,600 and 800 s/mm2, respectively). SIs on DW images and multiple-b-value DWI derived quantitative parameters of lesions and normal pancreases were measured by two observers independently using two different ROI size. The inter-observer variability for the SIs and diffusion parameters measurements was compared using the intraclass correlation coefficient (ICC). Mean SIs and diffusion parameters between the two ROI sizes were compared.Using Mann-Whitney U tests for the comparisons of SIs and diffusion parameters between normal pancreas and PAC groups. Kruskal-Wallis test was used for the comparison between normal pancreas and pancreatic lesions with different pathologic results.The diagnostic performances were calculated and compared by using the receiver operating characteristic (ROC) curves.Results:The inter-observer consistency for all SIs and diffusion parameters were excellent, and ICCs ranged from 0.847 to 0.991.Wilcoxon tests revealed SIs and ADCtotal differed significantly among the two groups of different ROI sizes, however, ICCs for all parameters between the two groups were higher than 0.90. Mann-Whitney U tests showed, SIs and parameters (except for ADCfast) between PAC and NP groups had statistically significant differences. Kruskal-Wallis test revealed SIs and ADCtotal differed significantly between normal pancreas and pancreatic lesions with different pathologic results.In addition, areas under the ROC curve for the identification of PAC from NP based on Sis were significantly higher than diffusion parameters.Conclusion:ROI size had no obvious effect on the pancreatic multiple-b-value DWI parameters. SIs for all b values provided higher diagnostic performance in differentiating common solid pancreatic focal lesions.