ABSTRACT
Objective To establish a neural network model based on enhanced CT for distinguishing ISUP grade of clear cell renal cell carcinoma (ccRCC). Methods We collected 131 cases of ccRCC, with 92 cases of low ISUP grade and 39 cases of high ISUP grade. Patients were divided into training set and validation set according to 5:5 stratified sampling. The enhanced CT images of each ccRCC patient were evaluated by the radiologist. Recursive feature elimination (RFE) was used to reduce the dimension of patients' general features and enhanced CT features, which was used for neural network modeling and validation. Results Patients' general features and enhanced CT features were verified by RFE method and then reduced to 14 features. The top 5 features were growth pattern, necrosis, enlargement of lymph nodes, tumor size and capsule. The AUC of the neural network model based on these 5 features in training set was 0.8844 (95%CI: 0.8062-0.9626), sensitivity was 89.47% and specificity was 82.61%; and those in validation set were 0.7924 (95%CI: 0.6567-0.9280), 75.00% and 86.96%, respectively. Conclusion The neural network model of ccRCC ISUP grade based on enhanced CT has relatively high diagnostic efficiency.
ABSTRACT
Objective@#To assess the relationship between body mass index (BMI) and mortality in adults of Shanxi, China.@*Methods@#Baseline data were from the '2002 China Nutrition and Health Survey’ in Shanxi province. All the death-related investigation and follow-up visits were carried out from December 2015 to March 2016. The follow-up program covered 5 360 people from all the 7 007 participants aged 18 years and over that having complete core information, with a rate as 76.5%. Participants of this study were divided into eight groups, according to the appearance of BMI. Taking the group with the lowest mortality density as the reference group, Cox regression model was used to estimate the hazard ratio (HR) and 95% confidence interval (CI) of deaths by the whole population, gender and age groups (≥60 years, <60 years). Results were then adjusted by age, gender, smoking, alcohol use and education level from the baseline survey. Sensitivity analysis was also conducted.@*Results@#Results from the study showed that among the total number of 67 129 person- years from the average period of 12.5 years, there were 615 deaths occurred, with the mortality density as 916 per 100 000 person-years. Taking the BMI range of 26.0-27.9 kg/m2 as the reference, the aHRs of death increased to 1.90 (95%CI: 1.26-2.86), 1.68 (95%CI: 1.15-2.45), 1.49 (95%CI: 1.08-2.06) and 1.72 (95%CI: 1.07-2.76) after the multivariate adjustment, in these four groups (BMI<18.5, 18.5-19.9, 22.0-23.9 and ≥30.0 kg/m2), respectively. Low body weight (BMI<18.5 kg/m2) was associated with higher risks of death in the elderly of ≥60 years, with the aHR of death as 1.94 (95%CI: 1.20-3.15).@*Conclusions@#When BMI appeared as ≤19.9 kg/m2, 22.0-23.9 kg/m2 and ≥30.0 kg/m2, the risks of death would increase. In addition to programs that focusing on obesity, special attention should be paid to the high risk of mortality which was caused by low-weight and malnutrition in the elderly.
ABSTRACT
Objective To compare the values of the lung nodule image quality by using two adaptive statistical iterative reconstruction (ASIR)in a chest phantom CT imaging study.Methods Non-enhanced CT of chest phantom with routine dose was obtained on both GE Discovery CT and Revolution CT.The scan parameters included tube voltage of 120 kVp,a noise index of 14 HU,slice thickness of 5 mm.All scans were reconstructed with FBP and different percentages of (20%－100% ASIR)when using Discovery CT,and different percentages of ASIR-V (20%－100% ASIR-V)when using Revolution CT.The CT attenuation values and SD (the SD represented image noise)of the subcutaneous uniform tissue were measured at the level of thoracic entry,trachea carina and hepatic portal.Based on the standard FBP(FBPs )algorithm,the image noise reduction rate of reconstructed images using different iterative percentage of standard algorithms ASIR and ASIR-V(ASIRs and ASIR-Vs )were compared.The paired t test was used to compare the noise of ASIRs and ASIR-Vs images under the same iterative percentage.The subj ective image quality was independently evaluated by two experienced radiologists,Kappa test was used to evaluate consistency,Wilcoxon test was used to compare subjective scores.Results For all ASIRs and ASIR-Vs images,the objective image noise of subcutaneous uniform tissue decreased along with increasing percentage.Comparison with the average noises of CT image reconstructed with the FBPs algorithm,the image noise reconstructed with ASIRs from 20% to 1 00% were reduced by 1 1.0 6%－48.9 7%,while the image noise reconstructed with ASIR-Vs from 20% to 100% were reduced by 17.06%－79.50%.The image noise of different percentage ASIR-Vs was significantly lower than that of same percentage ASIRs between 40% and 1 00% (P＜0.05 ).In terms of subj ective image evaluation,all ASIR1 and ASIR-V1 reconstructions had good diagnostic acceptability.Two readers scored good or moderate consistency in each iterative reconstruction of each lung nodule.Nodules of smaller size and less dense were more likely to have score differences.The 60%ASIR1 showed significantly superior visibility of lung nodules when compared to other percentages (P＜0.05).60% and 80% ASIR-V1 series showed significantly superior visibility of lung nodules when compared to the 60% ASIR1 and ASIR-V1 of other percentages (P＜0.05).Conclusion In routine dose non-enhanced chest CT, ASIR-V shows greater potential in reducing image noise and improves lung nodule image quality when compared to the same level ASIR algorithm. Combining both the obj ective and subj ective evaluation of images,reconstructed with 60% and 80% ASIR-V in the non-enhanced chest CT image is recommended for the observation and evaluation of pulmonary nodules.
ABSTRACT
Objective Toexploretheroleofmodel-basediterativereconstruction (MBIR)algorithminimprovingthequalityof thyroidCTimagesbyreducinghardeningartifactsattheentrancetothethorax.Methods ChestCTscansof20patientswiththyroidnoduleswere retrospectivelyreviewed.AlgorithmsofFBP,ASIR40,MBIRSTNDandMBIRNR40 wereusedtoreconstructatthe0.625mmslicethickness.Region ofinterestwasplacedonthecoronalimageswiththemostobvioushardeningartifacts.ThestandarddeviationsoftheCTvaluesof theleftandrightthyroidarteriesandthesurroundingnormalthyroidtissueweremeasured,furthermore,thevalueofthyroidartifact index(AI)andAI=sqrt(SDa2-SDb2)werecalculated.Tworadiologistsused4-pointmethodtoassessimageartifactssubjectivly.(1 point,severeartifacts,unclearthyroidrimsandnodules,undiagnosed;2points,moderateartifacts,poorlydisplayingthyroid margins andinternaldetails,affectivediagnosis;3points,mildartifacts,showingthyroid marginandinternaldetails,noaffectingthediagnosis;4 points ,no ribbon artifacts ,displaying thyroid edge and internal details very w ell ).ANOVA and paired t-test w ere used to co m pare the CTvaluesofnormalthyroidtissueofleftandrightlobeamongdifferentreconstructedimagesofthyroid.Subjectivescoredifferenceswere testedusingthe W ilcoxon symbolscale.Results ThevalueofCTreductionandAIatFBPandASIR40reconstructionimagesweresignificantly greaterthanthoseatMBIRSTNDandMBIRNR40reconstructionimages. Whiletherewerenodifferenceofleftandrightlobeofthyroid CTreductionand AIvalueatFBPand ASIR40reconstruction imagesnorthoseatMBIRSTNDandMBIRNR40reconstructionimages. SubjectiveevaluationofclavicleartifactswasincreasedatMBIRSTND and MBIRNR40image,andMBIRNR40imagehadoptimalsubjectiveevaluationresults(P<0.05).Conclusion MBIRcansignificantlyreducethe impactoftheclavicularharnesshardeningartifactsandnoiseonthethyroidanditsnodulesduringCTscans,especiallyoptimizingthe low-densitycontrastsettingofMBIRNR40.
ABSTRACT
Objective To compare the spatial resolution and density resolution balance algorithm(MBIRSTND)and spatial resolution preference algorithm (MBIRRP20)from new version of model-based iterative reconstruction(MBIRn),and adaptive statistical iterative reconstruction(ASIR) with lung kernel in routine dose about the performance of computer-aided detection (CAD)for quantitative analysis of airway.Methods 30 patients were involved who were scanned for pulmonary disease with spectrum CT.Data with a slice thinkness 0.625 mm were reconstructed with ASIR,MBIRSTNDand MBIRRP20.Airway dimensions from three reconstruction algorithm images were measured using an automated and quantitative software(Dexin-FACT)that was designed to segment and quantify the bronchial tree,and a skeletonization algorithm to extract the center-line of airway trees automatically.For each patient,reconstruction algorithm chose the right middle lobe bronchus,and the bronchial length of the matched airways was measured by this scheme.Two radiologists used a semiquantitative 5 scale (Score 0 stands for its image quality is similar to that with ASIR;Score±1 stand for a little better or a little worse;Score±2 stand for obviously better or obviously worse)to rate subjective image quality of airway trees about images reconstructed with MBIRSTNDand MBIRRP20.Paired t test and Wilcoxon signed-rank test were used.Results Algorithm impacts the measurement variability of bronchus length in chest CT.The bronchial length with MBIRRP20was longer than with MBIRSTND, while the length with ASIR were the shortest(P<0.05).In addition, the optimal reconstruction algorithm was found to affect the subjective noise,the continuity and completeness of bronchial wall,and the show of bronchial end.The subjective noise of MBIRSTNDwas better than that of MBIRRP20.The show of bronchial end of MBIRRP20was better than that of MBIRSTND(P<0.05).There was no significant difference in the continuity and completeness of bronchial wall compared with MBIRRP20and MBIRSTND(P>0.05),which was much better than with ASIR(P<0.05).Conclusion MBIRn can inmprove the analyzing ability of CAD airway.The MBIRSTNDcan significantly reduce the image noise,the MBIRRP20significantly improve the branching of the bronchial arteries,both of which can allow the desired airway quantification accuracy of CAD for chest CT of the bronchial wall.
ABSTRACT
Objective To demonstrate the feasibility of high-resolution computed tomography(HRCT)reconstructed with a model-based iterative reconstruction (MBIR)for evaluating early peripheral lung cancer (≤3 cm),by comparing image quality obtained from MBIR,filtered back proj ection reconstruction(FBP)and state of the art adaptive statistical iterative reconstruction(ASIR)algorithm respectively.Methods A total of 30 patients confirmed with lung cancer by postoperative pathology were enrolled in the study.A chest phantom was also used to evaluate image noise,spatial resolution and density resolution.Both patients and chest phantom were received HRCT,and the images were reconstructed using FBP,ASIR(40% ASIR and FBP mix)and MBIR.The objective CT value, standard deviation(SD)and signal noise ratio(SNR)were measured.Two radiologists used a semi-quantitative to rate subjective image quality of lung nodules.Results There was no significant difference in CT value between the three reconstruction algorithms (P>0.05).But significant improvements in objective image noise were observed in MBIR compared with FBP and ASIR (P<0.05):including the SD value in back muscle [(12.63±1.70)with MBIR vs (31.58±5.21)with FBP and (24.55±4.14)with ASIR],and in subcutaneous fat [(12.77±2.53)vs (24.39±5.08)and (19.20±4.11)].Subjective image noise of the three group were also significantly difference:FBP with lowest subjective noise score;and MBIR with highest subjective noise score.The sharpness of small vessels and bronchi and diagnostic acceptability with MBIR were significantly better than with FBP and ASIR (P< 0.05).Conclusion Lung HRCT reconstructed with MBIR provides diagnostically more acceptable images for the detailed analyses of peripheral lung cancer compared with FBP and ASIR.
ABSTRACT
Objective To explore the clinical value of adaptive statistical iterative reconstruction(ASIR)combined with automatic tube current modulation in low dose scan of chest CT.Methods 80 patients who underwent chest CT scan were randomly divided into four groups,with a noise index(NI)of 14,16,18,and 20,respectively.Automatic tube current modulation technique for chest CT scan was adopted for all patients.The standard dose group,with a NI of 14,was reconstructed with FBP.While the low-dose groups,with a NI of 16,18 and 20,were reconstructed with four ASIR levels(20% ASIR,40% ASIR,60% ASIR and 80% ASIR) in each group.The volume CT dose index(CTDIvol),dose length product(DLP)and effective dose(ED)were recorded of the four groups.The signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR)were calculated of the different reconstruction groups. Two radiologists with more than five years of work experience blindly scored the subjective image quality.Results Effective dose of the four groups with different NI were(3.29 ± 1.21)mSv,(3.10 ± 1.07)mSv,(2.20 ± 0.82)mSv and(1.97 ± 1.21)mSv,respectively. In all the reconstruction groups,when NI was constant,the greater the ASIR,the lower the SD.When ASIR percentage was constant, the SD was increased and the SNR was reduced along with NI rising up.When the parameters were set as NI 18 and ASIR 60%,the SD would be smaller and SNR would be higher than that of a standard dose group.In aspect of the scores from subjective image quality evaluation,there was no significant difference(P> 0.05).The average ED was decreased by about 33.2% when compared to the standard dose group.Conclusion Combining automatic tube current modulation with ASIR has a certain clinical practicality.When the NI is set at 18 and ASIR is 60%,the image quality can be optimal for not only satisfying the requirements of clinical diagnosis, but also reducing radiation dose in chest CT scanning.
ABSTRACT
Objective To evaluate the image quality improvement on abdominal CT imaging by using new model-based iterative reconstruction (MBIRn)in comparison with adaptive statistical iterative reconstruction (ASiR).Methods 40 patients who underwent upper abdominal three-phase contrast-enhanced scan were included.After scanning,all the scans obtained at 180 s later injection were reconstructed by three protocol,including ASiR (combined reconstruction of 40%FBP and 60%ASiR),the MBIRn of the noise reduction settings (MBIRNR40)and the spatinal resolution settings(MBIRSTND).The thickness of the slice was 0.625 mm.The values of CT and SD of the subcutaneous fat,left erector spinae,inferior vena cava and hepatic vein (left and right branches)were measured at the branch level of hepatic vein,and the contrast noise ratio (CNR)between inferior vena cava and hepatic vein were calculated.The subj ective image quality was evaluated by two radiologists according to the noise,smoothness and small branches of the inferior vena cava and hepatic vein using 5-scoring method.The quality images obtained from ASiR method were treated as reference standard.Results For MBIRSTNDand MBIRNR40images,the subjective noise decreased and image quality increased comparing with ASiR images.Among which the MBIRNR40images had the best image with vascular smoothness score and the lowest subjective noise.Conclusion Compared with ASiR,MBIRSTNDand MBIRNR40,especially MBIRNR40improves the quality of CT images of the inferior vena cava and its branches.
ABSTRACT
[Abstraet] Purpose To explore the application of noise reduction settings among modelbased iterative reconstruction (MBIRn) on decreasing radiation dose in CTA through experimental study.Materials and Methods Spectral CT was used to scan vessel model which could accommodate 8 tubes under static condition under 120 kVp fixed tube voltage at 10 mA,50 mA,150 mA and 600 mA.The tubes were filled with purified water,1,2,5,10,20 and 30 mgI/ml solutions and 30 mgCa/ml solution.Algorithms of FBP (standard),ASIR40 (combined by 40％ ASIR and FBP),MBIRc and MBIRn with optimized spatial resolution setting as MBIRRP20,MBIRstnd (standard setting),and MBIRNR40 (noise reduction setting) were adopted for original scanning data to reconstruct images of 0.625 mm slice thickness for contrastive analysis.Three fixed layers were chosen and regions of interest were placed on central tube and surrounding ester matrix,and CT value and standard variation (SD) were measured to represent noise.Noise reduction setting MBIRNR40 and solution noise of other reconstruction algorithm images were calculated and compared and average value of contrast noise ratio (CNR) was compared.Results Compared to FBP under different tube currents,ASIR40,MBIRc,MBIRRP20,MBIRstnd and MBIRNR40 reduce noise and increase CNR at different levels.Average noise of MBIRNR40 was the lowest (reduced by 78.33％) and CNR the highest (increased by 241.74％),making it superior to other reconstruction algorithm images (P＜0.05).Meanwhile,the lower the radiation dose was,the more obvious its advantage was.Reconstruction image noise of MBIRNR40 at 10 mA was close to that of FBP and ASIR40 at 600 mA.CNR was obvious greater than that of FBP and ASIR40 at other tube voltage.Conclusion Reconstruction algorithms of MBIRc,MBIRn and ASIR can help enhance CTA image quality and reduce radiation dose.Noise reduction setting MBIRNR40 from MBIRn has the lowest noise and greatest CNR.The lower the radiation dose is,the more obvious its advantage is.
ABSTRACT
Objective To compare the effect on image quality of low-dose upper abdominal CT reconstructed with the new version of model-based iterative reconstruction (MBIRn) focused on low-contrast resolution (MBIRNR40),conventional model-based iterative reconstruction (MBIRc),adaptive statistical iterative reconstruction (ASIR) and routine-dose CT reconstructed with filtered back projection (FBP).Methods Water plantom at rest was scanned with CT,and spatial resolution and density resolution were compared among FBP,ASIR,MBIRc,and MBIRNR40.Sixty patients with 2 times CT in the upper abdomen within a year were enrolled.The initial examination was acquired at a standard radiation dose (noise index [NI] of 10 HU) and reconstructed with the conventional FBP algorithm.The follow-up scan was acquired at a lowdose (NI=20 HU) and reconstructed with the standard ASIR,MBIRc and MBIRNR40.All images were obtained with 0.625 mm slice thickness.CT values and noise of fat,muscle as well as the liver and kidney parenchyma were measured and CNR of liver and kidney parenchyma using the fat SD as background image noise were calculated.Two radiologists independently graded images for noise,sharpness of details of structures and lesion.The quantitative image quality scores of different reconstructions were analyzed with one-way ANOVA using FBP reconstruction as reference of standard.The degree of interobserver consistency was evaluated using Kappa test.Results The phantom study revealed the highest spatial resolution with MBIRc and highest density resolution with MBIRNR40 among all reconstructions.The dose-length product and radiation dose for the first inspection was (93.18 ± 41.21) mGy · cm,(1.40 ± 0.62) mSv,respectively,and were (368.03 ± 146.25)mGy · cm,(5.52 ± 2.19)mSv for the second inspection,representing an approximate overall dose reduction of 74.68％ and 74.64％.The mean image noise of muscle and fat for MBIRNR40 was significantly lower than that of MBIR,ASIR and FBP(P＜0.05).The mean CNR values of liver and spleen for MBIRNR40 were significantly higher than that of ASIR,MBIRc and FBP (P＜0.05).Two radiologists had a good subjective score consistency.Low-dose MBIRNR40 subjective image noise was the lowest,showing the most detailed on the upper abdominal detail structure and lesion edge,better than MBIRc,MBIRc was superior to routine-dose FBP,low dose ASIR was worst,the difference was statistically significant (P＜0.05).Conclusion With 75％ dose reduction in upper abdominal CT,the MBIRNR40 can provide well objective and subjective image quality than MBIRc and ASIR40,and the routine-dose FBP.
ABSTRACT
Objective To assess image quality of adaptive statistical iterative reconstruction (ASIR),conventional modelbased iterative reconstruction (MBIRc) and a new lung-specific setting (MBIRRP20 and MBIRNR40) from the new version of model-based iterative reconstruction (MBIRn) in submillisievert chest CT comparing with ASIR in standard-dose.Methods Two chest CT examinations were performed with standard-dose and low-dose in 30 patients.Low-dose CT images were reconstructed with ASIR,MBIRc and MBIRn,while standard-dose CT images were reconstructed with ASIR only.Objective image noise and SNR were measured on the same part from the back muscle and subcutaneous fat which located at the level of thoracic entry,trachea carina and hepatic portal.Image quality of lung,mediastinum and upper abdomen structures were evaluated on a 5-point scale.The results were compared with one-way ANOVA and Wilcoxon signed-rank tests.Results The effective dose equivalent for standard-dose CT was (3.01 ± 1.89) mSv,compared with (0.88 ± 0.83) mSv for low dose CT,which decreased by 70.76％.The mean image noise for low-dose MBIRNR40 was significantly lower than that of conventional-dose ASIR,low-dose ASIR and MBIRc (P＜0.05).The mean SNR for low-dose MRIRNR40 was significant ly higher than that of conventional dose ASIR,low-dose ASIR and MBIRc (P＜0.05).The subjective image noise score was significantly lower than that of ASIR and MBIRc,and the score of sharpness of details of the structures score for low dose MBIRn was significantly better than that of the ASIR and MBIRc (P＜0.05).Conclusion MBIRNR40 can significantly reduce image noise and improve SNR compared to ASIR and MBIRc in low-dose,even better than ASIR in standard dose,which reduce radiation dose by about 70％.In low-dose,MBIRPP20 can well display lung structures,and MBIRNR40 can display mediastinal and the upper abdominal structures.
ABSTRACT
Objective To assess image quality of adaptive statistical iterative reconstruction (ASIR),conventional modelbased iterative reconstruction (MBIRc) and a new lung-specific setting (MBIRRP20 and MBIRNR40) from the new version of model-based iterative reconstruction (MBIRn) in submillisievert chest CT comparing with ASIR in standard-dose.Methods Two chest CT examinations were performed with standard-dose and low-dose in 30 patients.Low-dose CT images were reconstructed with ASIR,MBIRc and MBIRn,while standard-dose CT images were reconstructed with ASIR only.Objective image noise and SNR were measured on the same part from the back muscle and subcutaneous fat which located at the level of thoracic entry,trachea carina and hepatic portal.Image quality of lung,mediastinum and upper abdomen structures were evaluated on a 5-point scale.The results were compared with one-way ANOVA and Wilcoxon signed-rank tests.Results The effective dose equivalent for standard-dose CT was (3.01 ± 1.89) mSv,compared with (0.88 ± 0.83) mSv for low dose CT,which decreased by 70.76％.The mean image noise for low-dose MBIRNR40 was significantly lower than that of conventional-dose ASIR,low-dose ASIR and MBIRc (P＜0.05).The mean SNR for low-dose MRIRNR40 was significant ly higher than that of conventional dose ASIR,low-dose ASIR and MBIRc (P＜0.05).The subjective image noise score was significantly lower than that of ASIR and MBIRc,and the score of sharpness of details of the structures score for low dose MBIRn was significantly better than that of the ASIR and MBIRc (P＜0.05).Conclusion MBIRNR40 can significantly reduce image noise and improve SNR compared to ASIR and MBIRc in low-dose,even better than ASIR in standard dose,which reduce radiation dose by about 70％.In low-dose,MBIRPP20 can well display lung structures,and MBIRNR40 can display mediastinal and the upper abdominal structures.
ABSTRACT
Purpose To explore the correlation between normal pancreas and abdominal aorta in the peak enhancement (PE) and the shift time at the peak by applying the multislices spiral CT perfusion imaging.Materials and Methods Prospectively analyzed 62 patients who received enhancement CT examination for the superior or the middle abdomen,underwent optimum level CT perfusion imaging after plain scanning.These data were processed on a Vitreal 2.0 worker-station by using Toshiba body software package.The time-density curves (TDC) of the normal pancreas and the abdominal aorta were drawn,the PE and the shift time of PE were recorded and their correlation was analyzed.Results Compared with abdominal aorta,the mean value of PE of the normal pancreas was lower,and the difference was statistically significant [(111.94± 14.42)HU vs (351.83 ± 74.93)HU,P＜0.05],the mean difference was (246.10± 65.86)HU.Compared with abdominal aorta,the mean shift times of PE of the normal pancreas was latter,and the difference was statistically significant [(37.56±6.90) s vs (30.82±6.73) s,P＜0.05],the mean difference was (6.54±2.97)s.The PE and shift time of PE of the normal pancreas were positively and linearly correlated with that of abdominal aorta (r=0.438,r=0.379).Conclusion The PE of the normal pancreas is not synchronous with that of the abdominal aorta.The shift time of the former is usually 6～8 seconds slower than that of the latter.This provides a basis to find the PE of the normal pancreas in enhanced scan.
ABSTRACT
Objective To explore the value of the spectral CT iodine-based material decomposition technique in differential diagnosis of central pulmonary carcinoma from obstructive pneumonia and atelectasis.Methods 25 cases with central pulmonary carcinoma complicating with obstructive pneumonia and atelectasis underwent CT plain scan and spectral contrast scan including pulmonary-arterial-phase (PAP) and bronchial-arterial-phase(BAP).Conventional CT images and iodine base images were generated in each phase by GSI viewer.The 4 groups of images differences between the tumor and the obstructive pneumonia and atelectasis were analyzed.Results The margin of the tumor was ill-defined in plain scan.The cases that showed difference between the tumor and the obstructive pneumonia and atelectasis in 4 groups of images were respectively as follows,10 in conventional CT images of PAP,1 6 in conventional CT images of BAP,1 9 in iodine base images of PAP,and 23 in iodine base images of BAP.The difference of the 4 groups was significant (χ2 =16.54,P0.05),but were better than that of the conventional CT images of PAP(χ2 =6.65,P <0.05).Conclusion Spectral CT iodine-based material decomposition technique is helpful to improve subjective diagnosis of central pulmonary carcinoma complicating with obstructive pneumonia and atelectasis,especially in iodine base images of BAP,it can provide accurate information for clinical stage and treatment.
ABSTRACT
PurposeSupport vector machine (SVM) is a machine learning method based on statistical learning theory of Vapnik-Chervonenkis (VC) dimension structure and risk minimization theory. We analyzed the gem spectrum CT scan data of patients with thyroid nodules and established the SVM diagnostic model. The experimental targets were then reduced and the forecast analysis was carried out based on SVM model. The diagnostic model and experimental methods were proved to provide guidance for clinical diagnosis of thyroid nodules.