Computerized Quantification of Ultrasound Heterogeneity in Thyroid Nodules

Abstract

Brief Description of the Purpose of the StudyIn order to test whether the computerized quantification of ultrasonic heterogeneity can aid the diagnosis of thyroid malignancy, we evaluate ultrasound heterogeneity by an objective and quantitative computerized method in a prospective setting.MethodsA total of 347 participants with 400 nodules were evaluated. There were 271 benign thyroid nodules and 129 malignant thyroid nodules. Patient clinical data were collected, and the grading of heterogeneity on conventional gray scale ultrasound images was retrospectively reviewed by a thyroid specialist. Quantification of ultrasound heterogeneity was performed by a proprietary program.Main ResultsThe ultrasound heterogeneity index (HI) between benign and malignant nodules showed a significant difference (p < 0.001, AUC=0.714). To determine the relationship of the HI with conventional method, we compared HI to ultrasound heterogeneity evaluated by human investigators. In this study, the marked heterogeneity of ultrasound heterogeneity(US-H) showed higher HI than homogeneous US-H did. The US-H evaluated by human investigators did not show significant difference between benign and malignant thyroid nodules.Importance of the ConclusionsThis new computer-aided diagnosis method to evaluate the sonographic heterogeneity of thyroid nodules is an objective and quantitative method. This new computerized heterogeneity index can aid in the diagnosis of thyroid malignancy better than traditional US-H evaluated by human investigator. Brief Description of the Purpose of the StudyIn order to test whether the computerized quantification of ultrasonic heterogeneity can aid the diagnosis of thyroid malignancy, we evaluate ultrasound heterogeneity by an objective and quantitative computerized method in a prospective setting. In order to test whether the computerized quantification of ultrasonic heterogeneity can aid the diagnosis of thyroid malignancy, we evaluate ultrasound heterogeneity by an objective and quantitative computerized method in a prospective setting. MethodsA total of 347 participants with 400 nodules were evaluated. There were 271 benign thyroid nodules and 129 malignant thyroid nodules. Patient clinical data were collected, and the grading of heterogeneity on conventional gray scale ultrasound images was retrospectively reviewed by a thyroid specialist. Quantification of ultrasound heterogeneity was performed by a proprietary program. A total of 347 participants with 400 nodules were evaluated. There were 271 benign thyroid nodules and 129 malignant thyroid nodules. Patient clinical data were collected, and the grading of heterogeneity on conventional gray scale ultrasound images was retrospectively reviewed by a thyroid specialist. Quantification of ultrasound heterogeneity was performed by a proprietary program. Main ResultsThe ultrasound heterogeneity index (HI) between benign and malignant nodules showed a significant difference (p < 0.001, AUC=0.714). To determine the relationship of the HI with conventional method, we compared HI to ultrasound heterogeneity evaluated by human investigators. In this study, the marked heterogeneity of ultrasound heterogeneity(US-H) showed higher HI than homogeneous US-H did. The US-H evaluated by human investigators did not show significant difference between benign and malignant thyroid nodules. The ultrasound heterogeneity index (HI) between benign and malignant nodules showed a significant difference (p < 0.001, AUC=0.714). To determine the relationship of the HI with conventional method, we compared HI to ultrasound heterogeneity evaluated by human investigators. In this study, the marked heterogeneity of ultrasound heterogeneity(US-H) showed higher HI than homogeneous US-H did. The US-H evaluated by human investigators did not show significant difference between benign and malignant thyroid nodules. Importance of the ConclusionsThis new computer-aided diagnosis method to evaluate the sonographic heterogeneity of thyroid nodules is an objective and quantitative method. This new computerized heterogeneity index can aid in the diagnosis of thyroid malignancy better than traditional US-H evaluated by human investigator. This new computer-aided diagnosis method to evaluate the sonographic heterogeneity of thyroid nodules is an objective and quantitative method. This new computerized heterogeneity index can aid in the diagnosis of thyroid malignancy better than traditional US-H evaluated by human investigator.