Radiomics Study of Thyroid Ultrasound for Predicting BRAF Mutation in Papillary Thyroid Carcinoma: Preliminary Results

Abstract

BACKGROUND AND PURPOSE: It is not known how radiomics using ultrasound images contribute to the detection of BRAF mutation. This study aimed to evaluate whether a radiomics study of gray-scale ultrasound can predict the presence or absence of B-Raf proto-oncogene, serine/threonine kinase (BRAF) mutation in papillary thyroid cancer. MATERIALS AND METHODS: The study retrospectively included 96 thyroid nodules that were surgically confirmed papillary thyroid cancers between January 2012 and June 2013. BRAF mutation was positive in 48 nodules and negative in 48 nodules. For analysis, ROIs from the nodules were demarcated manually on both longitudinal and transverse sonographic images. We extracted a total of 86 radiomics features derived from histogram parameters, gray-level co-occurrence matrix, intensity size zone matrix, and shape features. These features were used to build 3 different classifier models, including logistic regression, support vector machine, and random forest using 5-fold cross-validation. The performance including accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating characteristic curve, of the different models was evaluated. RESULTS: The incidence of high-suspicion nodules diagnosed on ultrasound was higher in the BRAF mutation-positive group than in the mutation-negative group (P ¼ .004). The radiomics approach demonstrated that all classification models showed moderate performance for predicting the presence of BRAF mutation in papillary thyroid cancers with an area under the curve value of 0.651, accuracy of 64.3%, sensitivity of 66.8%, and specificity of 61.8%, on average, for the 3 models. CONCLUSIONS: Radiomics study using thyroid sonography is limited in predicting the BRAF mutation status of papillary thyroid carcinoma. Further studies will be needed to validate our results using various diagnostic methods. ABBREVIATIONS: AUC ¼ area under the receiver operating characteristic curve; GLCM ¼ gray-level co-occurrence matrix; ISZM ¼ intensity size zone matrix; K-TIRADS ¼ Korean Thyroid Imaging Reporting and Data System; mRMR ¼ minimum redundancy maximum relevance; PTC ¼ papillary thyroid carcinoma; US ¼ ultrasound

Radiomics Based on Thyroid Ultrasound Can Predict Distant Metastasis of Follicular Thyroid Carcinoma

“…Since the introduction of radiomics, many studies concluded that it has potential power for analyzing thyroid cancers, including the differentiation between malignant and benign nodules or the prediction of prognosis or gene mutations in papillary thyroid carcinoma [ 12 , 20 , 21 , 22 , 23 , 24 , 25 ]. However, little is known about the association of radiomics and FTC.…”

“…Contrary to tomographic images, US has several limitations in quantitative analysis including uncommon 3D data available, lack of reproducibility, lack of representative features due to limited amount of image data, operator dependency, and dependency on US machines. However, US is the most widely used standard imaging tool for evaluating thyroid pathology and many recent studies concluded that quantitative features extracted from thyroid US images have favorable results [ 12 , 20 , 21 , 22 , 23 , 24 , 25 ].…”

Radiomics in Differentiated Thyroid Cancer and Nodules: Explorations, Application, and Limitations

“…Thus, if we can predict genes mutated in thyroid cancer through convenient and feasible approaches, this information would contribute to improving tumor diagnosis, judging the prognosis, and personalizing the treatment. To date, two studies have applied radiomics to estimate BRAF mutations in PTC patients [ 73 , 81 ]. These two studies offered a consistent outcome that ultrasound radiomics has a limited value in predicting BRAF nutation.…”

Artificial Intelligence in Thyroid Field—A Comprehensive Review

“…In fact, to this date, several studies [ 7 , 20 , 28 , 43 ] recommend that the lesion evaluation should be achieved from a combination between the clinician evaluation and ML or DL outcome. Moreover, it is worth noticing that most AI-based studies focused on thyroid pathologies are performed using retrospectively collected data [ 9 , 11 , 33 , 40 , 42 , 51 , 55 , 60 , 61 , 62 , 63 , 65 , 66 , 67 ]. Conversely, studies that prospectively evaluate AI predictive models concerning thyroid disease diagnosis are limited in the literature [ 22 , 41 ].…”

“…In fact, [ 42 , 60 , 61 ] implemented radiomics models that analyze the risk stratification and predict the aggressiveness of the thyroid carcinoma with high values of accuracy, i.e., roughly 85 percent. Radiomics analysis has the potential to determine tumor phenotypes or the presence of gene mutations [ 62 , 63 ]. Furthermore, several studies have investigated by means of radiomic features the occurrence of metastases [ 64 ] or disease-free survival [ 65 ].…”