Threshold Segmentation Of Pulmonary Subsolid Nodules On CT Images:Detection And Quantification Of The Solid Component

Background and ObjectiveOn computed tomography(CT)images,the term pulmonary subsolid nodule(SSN)refers to both pure groundglass nodules and part-solid nodules.The detection and quantifcation of solid components in pulmonary subsolid nodules(SSN)are of vital importance on differential diagnosis,pathological speculation and prognosis prediction.However,no objective and wide-accepted criterion has been built up to now.The purpose of this study is to explore the optimal threshold that can be used for the detection and quantifcation of solid components in SSNs by using threshold segmentation method on CT images.MethodsCT images of 102 subsolid nodules were retrospectively analysed.To establish reference standards,observers made judgments on whether the solid component existed in every SSN and did measurements of the volume of solid component with the help of sofware.Segmentations of every nodule were then performed,the total volume of everyone was recorded,then using different threshold setings,ranging from-500 HU to-130 HU,to quantify segmented nodules.All of the measured volumes were assumed to be solid volumes,then solid-to-total volume ratios were calculated.The results were compared with the reference standards using the receiver operating characteristic curve(ROC)and paired Wilcoxon test to find out the optimal threshold and the cutoff value of volume ratio was determined throught the largest Youden index.ResultsThresholds of-250 HU,-300 HU resulted in good accuracy of subsolid nodules classification with AUC values 0.982,0.977 and volume ratio cutoff values 1.10%,6.14% respectively;median volumes of solid components 202.7 mm~3(598.2 mm~3),247.1 mm~3(696.0 mm~3)were not significantly different from the reference standard 199.5 mm~3(743.1 mm~3)(P=0.1251,0.0613).ConclusionThe threshold segmentation on chest CT images is valuable to detect and quantify the solid component of SSNs;thresholds as-250 HU and-300 HU are recommended.