Evaluation of a deep learning‐based computer‐aided diagnosis system for distinguishing benign from malignant thyroid nodules in ultrasound images

Abstract

Purpose: Computer-aided diagnosis (CAD) systems assist in solving subjective diagnosis problems that typically rely on personal experience. A CAD system has been developed to differentiate malignant thyroid nodules from benign thyroid nodules in ultrasound images based on deep learning methods. The diagnostic performance was compared between the CAD system and the experienced attending radiologists. Methods: The ultrasound image dataset for training the CAD system included 651 malignant nodules and 386 benign nodules while the database for testing included 422 malignant nodules and 128 benign nodules. All the nodules were confirmed by pathology results. In the proposed CAD system, a support vector machine (SVM) is used for classification and fused features which combined the deep features extracted by a convolutional neural network (CNN) with the hand-crafted features such as the histogram of oriented gradient (HOG), local binary patterns (LBP), and scale invariant feature transform (SIFT) were obtained. The optimal feature subset was formed by selecting these fused features based on the maximum class separation distance and used as the training sample for the SVM. Results: The accuracy, sensitivity, and specificity of the CAD system were 92.5%, 96.4%, and 83.1%, respectively, which were higher than those of the experienced attending radiologists. The areas under the ROC curves of the CAD system and the attending radiologists were 0.881 and 0.819, respectively. Conclusions: The CAD system for thyroid nodules exhibited a better diagnostic performance than experienced attending radiologists. The CAD system could be a reliable supplementary tool to diagnose thyroid nodules using ultrasonography. Macroscopic features in ultrasound images, such as the margins and shape of thyroid nodules, could influence the diagnostic efficiency of the CAD system.

A Computer-Aided Diagnosis System and Thyroid Imaging Reporting and Data System for Dual Validation of Ultrasound-Guided Fine-Needle Aspiration of Indeterminate Thyroid Nodules

“…Over the past decade, artificial intelligence (AI)-aided US techniques, which integrate US and computer science, have become increasingly employed for the detection and diagnosis of thyroid diseases (10,11). From traditional machine learning to deep learning methods, many algorithms, such as the Support Vector Machine (12), GoogleNet (13), and a convolutional neural network (14), have been shown to be effective for USbased diagnosis of thyroid nodules. These advanced methods have become increasingly used in recent years due to advances in commercial software applications, such as AmCad-UT (AmCad BioMed Corporation, Taipei City, Taiwan) (15), S-Detect (Samsung Medison Co., Ltd., Seoul, Korea) (3), and AI-SONIC (Demetics Medical Technology, Zhejiang, China).…”

Application of Artificial Intelligence in Medicine: An Overview

“…For example, Nguyen has proposed a new ultrasonographic image analysis method based on AI that successfully enhanced the consequence of thyroid nodule classification [38] . Other scholars have also confirmed that the use of AI can promote the traditional ultrasonographic detection of tumors in the thyroid, breast, bronchia, puborectalis muscle, and urogenital hiatus as well as other obstetric and gynecological lesions, with a high efficiency and accuracy [39][40][41][42][43][44] .…”

AF-SENet: Classification of Cancer in Cervical Tissue Pathological Images Based on Fusing Deep Convolution Features

“…The boundary point i hood of the new central pixel, and so on. HOG [32,33] features have a strong image structure and contour de ities as well as a strong recognition effect on the description of local are are also suitable for describing texture features. Texture features have and macro regularity.…”

“…If cell unit is too large, the local feature description is incomplete, which is generalizing the macro features. The cell unit size used in this paper wa HOG [32,33] features have a strong image structure and contour description capabilities as well as a strong recognition effect on the description of local areas. HOG features are also suitable for describing texture features.…”