Jialin Zhu1, Sheng Zhang1, Ruiguo Yu2, Zhiqiang Liu2, Hongyan Gao3, Bing Yue1, Xun Liu4, Xiangqian Zheng5, Ming Gao5, Xi Wei1. 1. Department of Diagnostic and Therapeutic Ultrasonography, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center of Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin's Clinical Research Center for Cancer, Tianjin, China. 2. College of Intelligence and Computing, Tianjin University, Tianjin Key Laboratory of Cognitive Computing and Application, Tianjin Key Laboratory of Advanced Networking, Tianjin, China. 3. Tianjin Xiqing District Women and Children's Health and Family Planning Service Center, Tianjin, China. 4. Department of Ultrasonography, the Fifth Central Hospital of Tianjin, Tianjin, China. 5. Department of Thyroid and Neck Tumor, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center of Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin's Clinical Research Center for Cancer, Tianjin, China.
Abstract
BACKGROUND: The aim of this study was to construct a deep convolutional neural network (CNN) model for localization and diagnosis of thyroid nodules on ultrasound and evaluate its diagnostic performance. METHODS: We developed and trained a deep CNN model called the Brief Efficient Thyroid Network (BETNET) using 16,401 ultrasound images. According to the parameters of the model, we developed a computer-aided diagnosis (CAD) system to localize and differentiate thyroid nodules. The validation dataset (1,000 images) was used to compare the diagnostic performance of the model using three state-of-the-art algorithms. We used an internal test set (300 images) to evaluate the BETNET model by comparing it with diagnoses from five radiologists with varying degrees of experience in thyroid nodule diagnosis. Lastly, we demonstrated the general applicability of our artificial intelligence (AI) system for diagnosing thyroid cancer in an external test set (1,032 images). RESULTS: The BETNET model accurately detected thyroid nodules in visualization experiments. The model demonstrated higher values for area under the receiver operating characteristic (AUC-ROC) curve [0.983, 95% confidence interval (CI): 0.973-0.990], sensitivity (99.19%), accuracy (98.30%), and Youden index (0.9663) than the three state-of-the-art algorithms (P<0.05). In the internal test dataset, the diagnostic accuracy of the BETNET model was 91.33%, which was markedly higher than the accuracy of one experienced (85.67%) and two less experienced radiologists (77.67% and 69.33%). The area under the ROC curve of the BETNET model (0.951) was similar to that of the two highly skilled radiologists (0.940 and 0.953) and significantly higher than that of one experienced and two less experienced radiologists (P<0.01). The kappa coefficient of the BETNET model and the pathology results showed good agreement (0.769). In addition, the BETNET model achieved an excellent diagnostic performance (AUC =0.970, 95% CI: 0.958-0.980) when applied to ultrasound images from another independent hospital. CONCLUSIONS: We developed a deep learning model which could accurately locate and automatically diagnose thyroid nodules on ultrasound images. The BETNET model exhibited better diagnostic performance than three state-of-the-art algorithms, which in turn performed similarly in diagnosis as the experienced radiologists. The BETNET model has the potential to be applied to ultrasound images from other hospitals. 2021 Quantitative Imaging in Medicine and Surgery. All rights reserved.
BACKGROUND: The aim of this study was to construct a deep convolutional neural network (CNN) model for localization and diagnosis of thyroid nodules on ultrasound and evaluate its diagnostic performance. METHODS: We developed and trained a deep CNN model called the Brief Efficient Thyroid Network (BETNET) using 16,401 ultrasound images. According to the parameters of the model, we developed a computer-aided diagnosis (CAD) system to localize and differentiate thyroid nodules. The validation dataset (1,000 images) was used to compare the diagnostic performance of the model using three state-of-the-art algorithms. We used an internal test set (300 images) to evaluate the BETNET model by comparing it with diagnoses from five radiologists with varying degrees of experience in thyroid nodule diagnosis. Lastly, we demonstrated the general applicability of our artificial intelligence (AI) system for diagnosing thyroid cancer in an external test set (1,032 images). RESULTS: The BETNET model accurately detected thyroid nodules in visualization experiments. The model demonstrated higher values for area under the receiver operating characteristic (AUC-ROC) curve [0.983, 95% confidence interval (CI): 0.973-0.990], sensitivity (99.19%), accuracy (98.30%), and Youden index (0.9663) than the three state-of-the-art algorithms (P<0.05). In the internal test dataset, the diagnostic accuracy of the BETNET model was 91.33%, which was markedly higher than the accuracy of one experienced (85.67%) and two less experienced radiologists (77.67% and 69.33%). The area under the ROC curve of the BETNET model (0.951) was similar to that of the two highly skilled radiologists (0.940 and 0.953) and significantly higher than that of one experienced and two less experienced radiologists (P<0.01). The kappa coefficient of the BETNET model and the pathology results showed good agreement (0.769). In addition, the BETNET model achieved an excellent diagnostic performance (AUC =0.970, 95% CI: 0.958-0.980) when applied to ultrasound images from another independent hospital. CONCLUSIONS: We developed a deep learning model which could accurately locate and automatically diagnose thyroid nodules on ultrasound images. The BETNET model exhibited better diagnostic performance than three state-of-the-art algorithms, which in turn performed similarly in diagnosis as the experienced radiologists. The BETNET model has the potential to be applied to ultrasound images from other hospitals. 2021 Quantitative Imaging in Medicine and Surgery. All rights reserved.
Authors: Hyung Jin Lee; Dae Young Yoon; Young Lan Seo; Jin Ho Kim; Sora Baek; Kyoung Ja Lim; Young Kwon Cho; Eun Joo Yun Journal: J Ultrasound Med Date: 2017-07-24 Impact factor: 2.153
Authors: Jenny K Hoang; William D Middleton; Alfredo E Farjat; Jill E Langer; Carl C Reading; Sharlene A Teefey; Nicole Abinanti; Fernando J Boschini; Abraham J Bronner; Nirvikar Dahiya; Barbara S Hertzberg; Justin R Newman; Daniel Scanga; Robert C Vogler; Franklin N Tessler Journal: Radiology Date: 2018-03-02 Impact factor: 11.105
Authors: Eoin F Cleere; Matthew G Davey; Shane O'Neill; Mel Corbett; John P O'Donnell; Sean Hacking; Ivan J Keogh; Aoife J Lowery; Michael J Kerin Journal: Diagnostics (Basel) Date: 2022-03-24