Xavier M Keutgen1, Hui Li2, Kelvin Memeh1, Julian Conn Busch1, Jelani Williams1, Li Lan2, David Sarne3, Brendan Finnerty4, Peter Angelos1, Thomas J Fahey4, Maryellen L Giger2. 1. The University of Chicago Medicine, Endocrine Surgery Research Program, Division of General Surgery and Surgical Oncology, Department of Surgery, Chicago, Illinois, United States. 2. The University of Chicago, Department of Radiology, Chicago, Illinois, United States. 3. The University of Chicago Medicine, Division of Endocrinology, Department of Medicine, Chicago, Illinois, United States. 4. New York Presbyterian Hospital-Weill Cornell Medicine, Endocrine Oncology Research Program, Division of Endocrine Surgery, Department of Surgery, New York, United States.
Abstract
Background: Ultrasound (US)-guided fine needle aspiration (FNA) cytology is the gold standard for the evaluation of thyroid nodules. However, up to 30% of FNA results are indeterminate, requiring further testing. In this study, we present a machine-learning analysis of indeterminate thyroid nodules on ultrasound with the aim to improve cancer diagnosis. Methods: Ultrasound images were collected from two institutions and labeled according to their FNA (F) and surgical pathology (S) diagnoses [malignant (M), benign (B), and indeterminate (I)]. Subgroup breakdown (FS) included: 90 BB, 83 IB, 70 MM, and 59 IM thyroid nodules. Margins of thyroid nodules were manually annotated, and computerized radiomic texture analysis was conducted within tumor contours. Initial investigation was conducted using five-fold cross-validation paradigm with a two-class Bayesian artificial neural networks classifier, including stepwise feature selection. Testing was conducted on an independent set and compared with a commercial molecular testing platform. Performance was evaluated using receiver operating characteristic analysis in the task of distinguishing between malignant and benign nodules. Results: About 1052 ultrasound images from 302 thyroid nodules were used for radiomic feature extraction and analysis. On the training/validation set comprising 263 nodules, five-fold cross-validation yielded area under curves (AUCs) of 0.75 [Standard Error (SE) = 0.04; P < 0.001 ] and 0.67 (SE = 0.05; P = 0.0012 ) for the classification tasks of MM versus BB, and IM versus IB, respectively. On an independent test set of 19 IM/IB cases, the algorithm for distinguishing indeterminate nodules yielded an AUC value of 0.88 (SE = 0.09; P < 0.001 ), which was higher than the AUC of a commercially available molecular testing platform (AUC = 0.81, SE = 0.11; P < 0.005 ). Conclusion: Machine learning of computer-extracted texture features on gray-scale ultrasound images showed promising results classifying indeterminate thyroid nodules according to their surgical pathology.
Background: Ultrasound (US)-guided fine needle aspiration (FNA) cytology is the gold standard for the evaluation of thyroid nodules. However, up to 30% of FNA results are indeterminate, requiring further testing. In this study, we present a machine-learning analysis of indeterminate thyroid nodules on ultrasound with the aim to improve cancer diagnosis. Methods: Ultrasound images were collected from two institutions and labeled according to their FNA (F) and surgical pathology (S) diagnoses [malignant (M), benign (B), and indeterminate (I)]. Subgroup breakdown (FS) included: 90 BB, 83 IB, 70 MM, and 59 IM thyroid nodules. Margins of thyroid nodules were manually annotated, and computerized radiomic texture analysis was conducted within tumor contours. Initial investigation was conducted using five-fold cross-validation paradigm with a two-class Bayesian artificial neural networks classifier, including stepwise feature selection. Testing was conducted on an independent set and compared with a commercial molecular testing platform. Performance was evaluated using receiver operating characteristic analysis in the task of distinguishing between malignant and benign nodules. Results: About 1052 ultrasound images from 302 thyroid nodules were used for radiomic feature extraction and analysis. On the training/validation set comprising 263 nodules, five-fold cross-validation yielded area under curves (AUCs) of 0.75 [Standard Error (SE) = 0.04; P < 0.001 ] and 0.67 (SE = 0.05; P = 0.0012 ) for the classification tasks of MM versus BB, and IM versus IB, respectively. On an independent test set of 19 IM/IB cases, the algorithm for distinguishing indeterminate nodules yielded an AUC value of 0.88 (SE = 0.09; P < 0.001 ), which was higher than the AUC of a commercially available molecular testing platform (AUC = 0.81, SE = 0.11; P < 0.005 ). Conclusion: Machine learning of computer-extracted texture features on gray-scale ultrasound images showed promising results classifying indeterminate thyroid nodules according to their surgical pathology.
Authors: Brooke Nickel; Tessa Tan; Erin Cvejic; Peter Baade; Donald S A McLeod; Nirmala Pandeya; Philippa Youl; Kirsten McCaffery; Susan Jordan Journal: JAMA Otolaryngol Head Neck Surg Date: 2019-03-01 Impact factor: 6.223
Authors: Nicole A Cipriani; Alexander T Pearson; James M Dolezal; Anna Trzcinska; Chih-Yi Liao; Sara Kochanny; Elizabeth Blair; Nishant Agrawal; Xavier M Keutgen; Peter Angelos Journal: Mod Pathol Date: 2020-12-10 Impact factor: 7.842
Authors: Mark A Lupo; Ann E Walts; J Woody Sistrunk; Thomas J Giordano; Peter M Sadow; Nicole Massoll; Ryan Campbell; Sara A Jackson; Nicole Toney; Christina M Narick; Gyanendra Kumar; Alidad Mireskandari; Sydney D Finkelstein; Shikha Bose Journal: Diagn Cytopathol Date: 2020-08-07 Impact factor: 1.582