| Part one Application of multi parameter quantitative analysis of energy spectrum CT in differential diagnosis of lung cancer and chronic inflammatory disease in lung Objective To investigate the value of spectral CT parameters in differentiating lung cancers from pulmonary chronic inflammatory lesions in lung.Methods 113 cases who had lung nodule or mass underwent chest plain and enhanced CT scan with spectral imaging CT in February 2016 to July 2017,including 82 cases of lung cancers,31 cases of pulmonary chronic inflammatory lesions,all cases were confirmed by operation and pathology.The slope of 40-100 keV spectrum curve(?_H),effective atomic number(Effective-Z),calcium concentration and HAP concentration were measured and compared respectively in plain scan(PS),and?_H,normalized iodine concentration(NIC),water concentration were measured and compared respectively in arterial phase(AP)and in venous phase(VP).The independent sample t test was used to analyze the difference of the energy spectrum parameters,and the receiver operating curve was generated,and the sensitivity and specificity of the parameters were compared.Results(1)In PS:Effective-Z??_H?HAP concentration were 7.764±0.222?0.396±0.381?10.185±9.147mg/cm~3 and 7.872±0.194?0.594±0.346?14.581±8.561 mg/cm~3 of lung cancers and pulmonary chronic inflammatory lesions,respectively.The parameters of pulmonary chronic inflammatory lesions were all higher than those of lung cancer,and the difference was statistically significant.(t=-2.176?P=0.035,t=-2.257?P=0.030,respectively).There was no statistically significant difference of calcium concentration(P>0.05).(2)In AP:NIC and?_H were 0.114±0.045?1.464±0.621 and 0.179±0.084?2.302±1.224 of lung cancers and pulmonary chronic inflammatory lesions,respectively.These parameters of pulmonary chronic inflammatory lesions were all higher than those of lung cancers(t=-5.253?P=0.000,t=-4.797?P=0.000,respectively).There was no statistically significant difference in Water concentration(P>0.05).(3)In VP:NIC and?_H were 0.339±0.105?2.102±0.739 and 0.417±0.144?2.545±0.897 of lung cancers and pulmonary chronic inflammatory lesions,respectively.These parameters of pulmonary chronic inflammatory lesions were all higher than those of lung cancers(t=-3.158?P=0.002,t=-2.450?P=0.018,respectively).There was no statistically significant difference in Water concentration(P>0.05).(4)NIC<0.18068 as a diagnostic threshold in AP.The sensitivity,specificity,positive predictive value,negative predictive value and accuracy of distinguishing lung cancers from pulmonary chronic inflammatory lesions were 54.8%,93.9%,77.3%,84.6%and 83.2%respectively..Conclusions multi-parameter quantitative analysis with spectral CT imaging is helpful in the identification of lung cancer and pulmonary chronic inflammatory lesions.Part two The value of spectral CT multi parameter quantitative analysis in differential diagnosis of different pathological types of lung cancerObjective To explore the value of multi-parameter quantitative analysis with spectral CT imaging in differentiating lung cancer with different pathological types.Methods 72 cases who had lung nodule or mass underwent chest plain and enhanced CT scan with spectral imaging CT in February 2016 to March 2017,including 44 cases of adenocarcinomas(ADC),23 cases of squamous cell carcinomas(SQCC)and 5 cases of small cell lung cancers(SCLC),all cases were confirmed by operation and pathology.The slope of 40-100 ke V spectrum curve(?H),effective atomic number(Effective-Z),calcium concentration and HAP concentration were measured and compared respectively in plain scan(PS),and ?H,normalized iodine concentration(NIC)?water concentration were measured and compared respectively in arterial phase(AP)and in venous phase(VP).These quantitative parameters that satisfy normal distribution and homogeneity of variance were compared by One-Way ANOVA,parameters with statistically significant differences were used to draw the receiver operating characteristic curve and calculate the area under the curve to evaluate the differential diagnostic efficiency of different parameters.Results(1)In PS: Effective-Z and ?H were 7.79±0.26,7.63±0.12,7.69±0.05 and 0.45±0.45,0.17±0.20,0.27±0.10 in ADC,SQCC and SCLC,respectively.These parameters of ADC were all higher than SCLC and SQCC,the difference between groups was statistically significant(F=3.423?P=0.04,F=3.476?P=0.038,respectively).There was no statistically significant difference in HAP concentration and calcium concentration between the groups(P>0.05).(2)In AP: water concentration and ?H were 1019.79±10.33mg/cm3,1027.97±7.42 mg/cm3,1027.22±8.52 mg/cm3 and 1.58±0.66,1.11±0.40,1.08±0.27 in ADC,SQCC and SCLC,respectively.The water concentration in SQCC were significantly higher than SCLC and ADC,The ?H in ADC were significantly higher than SCLC and SQCC,the difference between groups was statistically significant(F=6.303?P=0.003,F=5.833?P=0.005,respectively).There was no statistically significant difference in NIC between the groups(P>0.05).(3)In VP: NIC and ?H were 0.36±0.10,0.29±0.11,0.31±0.11 and 2.29±0.76,1.66±0.58,1.71±0.48 in ADC,SQCC and SCLC,respectively.The NIC and ?H in ADC were significantly higher than SCLC and SQCC,the difference between groups was statistically significant(F=3.974?P=0.023,F=6.766?P=0.002,respectively).There was no statistically significant difference in water concentration between the groups(P>0.05).(4)There was statistically significant difference in all quantitative parameters of spectral CT between ADC and SQCC(P<0.05).?H > 1.77697 as a diagnostic threshold in VP.The sensitivity,specificity,positive predictive value,negative predictive value and accuracy of distinguishing lung adenocarcinoma from squamous cell carcinoma were 79.5%,69.6%,83.3%,64% and 76.1%,respectively,Conclusions multi-parameter quantitative analysis with spectral CT imaging in the identification of lung cancer with different pathological types has a certain value.Part there Application of multi parameter quantitative CT analysis in evaluation of lymph node metastasis in lung cancerObjective Investigate the evaluation of multi parameter quantitative CT analysis in mediastinal lymph node metastasis of lung cancer.Methods 23 cases of non small cell lung cancer confirmed by pathology from February 2016 to July 2017 underwent lung cancer resection and mediastinal lymph node dissection.The CT chest enhanced examination was performed,and 92 cases of mediastinal lymph nodes were compared with the pathological results,including 41 cases of metastatic lymph nodes and 51 cases of non metastatic lymph nodes.The slope of 40-100 ke V spectrum curve,normalized iodine concentration(NIC)and water concentration in metastatic lymph nodes,non metastatic lymph nodes and primary lesions in AP and VP were measured respectively.Independent sample t test was used to compare the differences between metastatic lymph nodes and non metastatic lymph nodes,,and P < 0.05 indicated that the difference was statistically significant.Parameters with statistically significant differences were used to draw the receiver operating characteristic curve and calculate the area under the curve to evaluate the differential diagnostic efficiency of different parameters.Results 1?Comparison between metastatic lymph nodes and non lymph nodes:(1)The NIC??H and water concentration in metastatic lymph nodes and non metastatic lymph nodes in arterial phase were 0.188±0.073 ?2.542±0.881?1021.310±16.732 mg/cm3 and 0.232±0.072 ?2.942±0.811?1013.731±17.820 mg/cm3 respectively.(2)The NIC??H in non metastatic lymph nodes were significantly higher than metastatic lymph nodes,the water concentration in metastatic lymph nodes were significantly higher than non metastatic lymph nodes,and the difference was statistically significant(t= 2.920?P=0.004,t=2.241?P=0.028).The NIC in metastatic lymph nodes and non metastatic lymph nodes in venous phase were 0.438±0.128 and 0.541±0.129 respectively,the number of non metastatic lymph nodes was higher than metastatic lymph nodes,and the difference was statistically significant(t= 3.845,P=0.000).The water concentration and ?H difference was not obvious(P>0.05).2?NIC < 0.42135 as the diagnostic threshold,the diagnostic sensitivity,specificity,positive predictive value,negative predictive value and accuracy of the metastatic lymph nodes and non metastatic lymph nodes were 86.3%,43.9%,72%,65.7% and 67.4%,respectively.Conclusions multi parameter quantitative analysis of energy spectrum CT has certain clinical value in differentiating metastatic lymph nodes from non metastatic lymph nodes of non-small cell lung cancer. |
PDF Full Text Request