Automatic thyroid nodule recognition and diagnosis in ultrasound imaging with the YOLOv2 neural network

Abstract

BackgroundIn this study, images of 2450 benign thyroid nodules and 2557 malignant thyroid nodules were collected and labeled, and an automatic image recognition and diagnosis system was established by deep learning using the YOLOv2 neural network. The performance of the system in the diagnosis of thyroid nodules was evaluated, and the application value of artificial intelligence in clinical practice was investigated.MethodsThe ultrasound images of 276 patients were retrospectively selected. The diagnoses of the radiologists were determined according to the Thyroid Imaging Reporting and Data System; the images were automatically recognized and diagnosed by the established artificial intelligence system. Pathological diagnosis was the gold standard for the final diagnosis. The performances of the established system and the radiologists in diagnosing the benign and malignant thyroid nodules were compared.ResultsThe artificial intelligence diagnosis system correctly identified the lesion area, with an area under the receiver operating characteristic (ROC) curve of 0.902, which is higher than that of the radiologists (0.859). This finding indicates a higher diagnostic accuracy (p = 0.0434). The sensitivity, positive predictive value, negative predictive value, and accuracy of the artificial intelligence diagnosis system for the diagnosis of malignant thyroid nodules were 90.5%, 95.22%, 80.99%, and 90.31%, respectively, and the performance did not significantly differ from that of the radiologists (p > 0.05). The artificial intelligence diagnosis system had a higher specificity (89.91% vs 77.98%, p = 0.026).ConclusionsCompared with the performance of experienced radiologists, the artificial intelligence system has comparable sensitivity and accuracy for the diagnosis of malignant thyroid nodules and better diagnostic ability for benign thyroid nodules. As an auxiliary tool, this artificial intelligence diagnosis system can provide radiologists with sufficient assistance in the diagnosis of benign and malignant thyroid nodules.

Computer-Aided Diagnosis Systems in Diagnosing Malignant Thyroid Nodules on Ultrasonography: A Systematic Review and Meta-Analysis

“…Second, the rigid cutoff levels that were adopted to determine the diagnostic conclusion of the radiologists may also have influenced the per-formance of the radiologists. For instance, points 4a, 4b, and 5 of the TI-RADS criteria were adopted by researchers to determine the conclusion of radiologists during the diagnostic process [7,21,24]. It is probable that a different conclusion would have been drawn if the cutoff level had been adjusted.…”

“…The pooled sensitivity, specificity, AUC, and DOR are demonstrated in Figure 2b. Eleven of the 13 studies compared the diagnostic performances of CAD systems and radiologists [7,20,21,[23][24][25][26][27][28]. The pooled sensitivity, specificity, AUC, and DOR were comparable between the CAD systems and the radiologists (Fig.…”

“…The flow diagram of the literature search is shown in Figure 1. Nineteen studies with 4,781 nodules used in external validation sets were included in the study, including 6 studies on classic machine learning-based CAD sys-DOI: 10.1159/000504390 tems [14][15][16][17][18][19] and 13 studies on deep learning-based CAD systems [7,[20][21][22][23][24][25][26][27][28][29][30][31]. The general characteristics of the included studies are shown in Table 1, and the detailed characteristics are demonstrated in online supplementary Table 1 (see www.karger.com/doi/10.1159/000504390 for all online suppl.…”

“…Thirteen studies with 1,667 malignant and 1,415 benign nodules were included in the analysis [7,[20][21][22][23][24][25][26][27][28][29][30][31]. The pooled sensitivity, specificity, AUC, and DOR are demonstrated in Figure 2b.…”

Commentary on the study of Li J et al: The KWAK TI‐RADS and 2015 ATA guidelines for medullary thyroid carcinoma, combined with cell block‐assisted ultrasound‐guided thyroid fine‐needle aspiration

“…Taking into consideration the most recent trend in utilization of artificial intelligence based on deep neural networks for building computer‐aided diagnosis systems for ultrasound assessment of thyroid lesions, and quite promising initial reports in this field for the diagnosis of PTC, it would be interesting to see in near‐future outcomes of research focused on MTC diagnosis based on artificial intelligence image analysis . Computer‐aided diagnosis systems may be considered an additional research tool available nowadays for further validation of sonographic patterns of suspected thyroid lesions in search for phenotype differences and signatures of different histological types of thyroid cancers.…”