Evaluation with low-dose dual-phase helical computed tomography of patients with thyroid lesions

Background The incidence of thyroid cancer has been increasing.Our aim was to evaluate the efficacy of low-dose dualphase helical computed tomography （CT） in the characterization of thyroid lesions,and to discuss the relationship between image characteristics and their pathology.Methods One hundred and six patients with thyroid lesions underwent low-dose dual-phase helical CT after the injection of contrast material.CT scans were obtained at arterial and venous phase with delays of 25 and 65 seconds,and tube current of 60 and 120 mA,respectively.The attenuation change in the lesion between the arterial and venous phase was analyzed and categorized as ＂increased,＂ ＂decreased,＂ ＂mixed＂ or ＂no change.＂ Results Histopathologic diagnosis was obtained by surgery in 106 patients （115 lesions）.Of the 106 patients,45 had nodular goiter,5 thyroid adenoma,6 thyroiditis,and 50 papillary thyroid carcinoma （PTC） （59 lesions）.The attenuation value showed a significant difference （P ＜0.05） between the arterial and venous phase for the high attenuation area.There was statistical significant difference in terms of attenuation value in high attenuation areas at both phases and in low attenuation areas on arterial phase between nodular goiter and PTC （P ＜0.05）.However,there was no significant difference in attenuation value between adenoma and PTC.Twenty-nine cases （76.3％） of goiter manifested mixed type,3 cases （3/5） of adenoma showed decreased type,6 cases （6/6） of thyroiditis showed increased type,and 55 cases （93.2％） of PTC showed decreased type attenuation.The sensitivity,specificity for thyroid carcinoma by dual-phase CT were 94.9％ and 80.4％ respectively.The overall diagnostic accuracy for thyroid lesions by dual-phase CT was 87.8％.Conclusions The performance of dual-phase helical CT is related to the pathological structure of the lesions.The analysis of enhancement patterns by using dual-phase helical CT will be helpful in the differential diagnosis of thyroid lesions.

3D Gray Level Co-Occurrence Matrix Based Classification of Favor Benign and Borderline Types in Follicular Neoplasm Images

Since the efficiency of treatment of thyroid disorder depends on the risk of malignancy, indeterminate follicular neoplasm (FN) images should be classified. The diagnosis process has been done by visual interpretation of experienced pathologists. However, it is difficult to separate the favor benign from borderline types. Thus, this paper presents a classification approach based on 3D nuclei model to classify favor benign and borderline types of follicular thyroid adenoma (FTA) in cytological specimens. The proposed method utilized 3D gray level co-occurrence matrix (GLCM) and random forest classifier. It was applied to 22 data sets of FN images. Furthermore, the use of 3D GLCM was compared with 2D GLCM to evaluate the classification results. From experimental results, the proposed system achieved 95.45% of the classification. The use of 3D GLCM was better than 2D GLCM according to the accuracy of classification. Consequently, the proposed method probably helps a pathologist as a prescreening tool.