| Objective:1.Verify that the image quality based on IMR algorithm is better than the FBP and iDose~4algorithms under conventional dose.2.To explore whether the IMR technique at low dose can still achieve the diagnostic accuracy of craniocerebral low-contrast lesion and is superior to the FBP and iDose~4algorithms.Methods1.Case materialsPatients with suspected acute cerebral infarction,white matter demyelination and cerebral hemorrhage who underwent CT scan of the brain from June 2018 to January2019 were collected from the Second Affiliated Hospital of Dalian Medical University.According to the random principle,Divide them into groups A and B,low-dose(100KV)and conventional dose(120KV)were performed.There were 9 males and 11 females in group A,with an average age of(60.8±9.3)years and a body mass index(BMI)of(20.8±1.7).)kg/m~2.There were 7 males and 13 females in group B,with an average age of(61.1±8.6)years and a body mass index(BMI)of(21.6±1.5)kg/m~2.General clinical data such as age,gender,and BMI were not significantly different between the two groups and were comparable.2.Scanners and parametersUsing a Philips Brilliance 256-slice iCT scan,the patient was placed in a supine position with a baseline of auditory sacral lines,scanned from the posterior fossa to the cranial apex,and continuously scanned from the foot to the head.Scanning parameters:tube voltage group A(100KV),group B(120KV),according to the patient’s body mass index tube current using automatic milliampere adjustment,other scanning parameters are consistent,the collimator width is 128×0.625 mm,the spacing is 0.91 mm,The layer thickness is 3 mm,FOV 23 cm x 20 cm,and matrix 512 x 512.MRI is used as a reference,and patients with acute stroke must scan for DWI sequences.3.Radiation dose analysisThe CT dose length product(DLP)and the computed tomographic dose index volume(CTDIvol)generated by the two groups of CT examination equipment were recorded separately.Calculate the effective dose,ED(mSv).4.Image quality assessment(1)Objective evaluation of image qualityThe source data of the two groups A and B are respectively three-dimensional iterative model reconstruction(IMR),advanced hybrid iterative reconstruction(HIR),iDose~4 and filtered back-projection(FBP).The method reconstructs the image.The basal ganglia level was selected as the Region Of Interest(ROI),the lenticular nucleus was used as the gray matter ROI,the posterior limb of the internal capsule was used as the white matter ROI,and the cerebrospinal fluid ROI was selected as the anterior horn of the lateral ventricle to avoid the choroid plexus as much as possible.The average CT value and background noise value(SD)of gray matter,white matter and cerebrospinal fluid were measured,and the signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR)of the region of interest were calculated according to the formula.Treatment of low-contrast lesions:Select a large enough ROI,select the same size ROI on the healthy side of the lesion,measure the CT value,and find the difference between the two CT values.If there is lesion on the contralateral side,select the lesion adjacent to the normal brain tissue for measurement.The lesions with CT value difference<10HU were selected as low-contrast lesions.All of the above CT values and SD values were fixed by ROI and repeated 3times,and averaged.(2)Subjective assessment of image qualityThe reconstructed images were independently scored by two radiologists with extensive experience in the diagnosis of central nervous system diseases.The scores include:(1)Normal brain tissue evaluation grading:Mainly observe brain tissue anatomy,such as cerebral hemisphere ash,white matter boundary,sulci,cerebral ventricle,ventricle display,image graininess,image artifact degree.(2)Evaluation of low-contrast lesions:Mainly observe the contour display of the lesion and the confidence of diagnosis.A consistency test analysis was performed on the obtained results.When the results of the two physicians are inconsistent,a central nervous system physician with more than20 years of experience is required to determine the difference and finally reach a consensus score.5.Statistical method analysisStatistical analysis was performed using SPSS 22.0 statistical analysis software.Clinical data of groups A and B were measured by independent sample t-test and chi-square test(x~2).The CT volume dose index,dose length product,and effective radiation dose were compared using an independent sample t test.One-way analysis of variance was used to compare the objective evaluation indexes of IMR,iDose~4 and FBP reconstructed images in the two groups.The independent sample t test was used to compare the objective evaluation indexes of IMR reconstructed images between the two groups.The Friedman test was used to compare the reconstructed images of IMR,iDose~4and FBP in the two groups and the subjective evaluation indexes of craniocerebral low contrast lesion.The subjective evaluation index of IMR reconstruction between the two groups was performed by Mann-Whitney U test.The Kappa test line consistency test was used to analyze the subjective scores of the image quality of the two physicians.P<0.05was considered statistically significant.Results1.Comparison of the average radiation dose between the two groups:Compared with group B(120 KV),CTDIvol,DLP and ED in group A(100KV)are reduced by 47.91%,39.1%and 38.73%,respectively.2.Objective image quality analysis:Under the condition of 100KV scanning,the signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR)of IMR algorithm were higher than those of IDose~4 and FBP algorithms,and the differences were statistically significant(P<0.05).The background noise value(SD)of the IMR algorithm was lower than the iDose~4 and FBP algorithms at100KV scanning,and the difference was statistically significant(P<0.05).Since the signal-to-noise ratio and the contrast-to-noise ratio reflect the indicators of excellent image quality,the higher the value,the better the image quality,that is,the lower the background noise value of the target brain anatomy indicates the higher the quality of the brain tissue imaging.It can be seen that the image quality of the IMR algorithm is significantly better than the iDose~4 and FBP algorithms.Comparison of signal-to-noise ratio(SNR),contrast signal-to-noise ratio(CNR),CT value and background noise value(SD)of IMR algorithm between 120KV and 100KV groups:Gray and white matter CT values of 120KV group and 100KV group are statistically significant Difference,gray matter and white matter in the 100KV group CT value is slightly lower than the 120KV group,however,there is no significant difference in background noise,signal-to-noise ratio and contrast signal-to-noise ratio between the two groups,indicating that the low-dose scan and normal dose scan are in terms of imaging quality.There is no obvious disadvantage.3.Subjective image quality analysis:The results of the study showed that the image scores of the IMR algorithm in the100KV group and the 120KV group were higher than iDose~4 and FBP in the subjective scoring of the images,and there was no statistical difference in the scores of the IMR reconstruction algorithm between the 100KV and 120KV groups.The results showed that the IMR algorithm was superior to iDose~4 and FBP in the observation of low contrast lesions and brain tissue anatomy under the same dose scan.And IMR combined with low-dose scan showed no significant difference between the craniocerebral low contrast lesion and the conventional dose.Conclusions1.IMR technology has better image quality than traditional FBP and iDose~4 technology under conventional and a range of low dose scanning.2.Under low dose scanning,IMR has higher accuracy than FBP and iDose~4 for the diagnosis of craniocerebral low-contrast lesion. |
PDF Full Text Request