Ultrasound is the best way to diagnose thyroid nodules. To discriminate benign and malignant nodules, calcification is an important characteristic. However, calcification in ultrasonic images cannot be extracted accurately because of capsule wall and other internal tissue. In this paper, deep learning was first proposed to extract calcification, and two improved methods were proposed on the basis of Alexnet convolutional neural network. First, adding the corresponding anti-pooling (unpooling) and deconvolution layers (deconv2D) made the network to be trained for the required features and finally extract the calcification feature. Second, modifying the number of convolution templates and full connection layer nodes made feature extraction more refined. The final network was the combination of two improved methods above. To verify the method presented in this article, we got 8 416 images with calcification, and 10 844 without calcification. The result showed that the accuracy of the calcification extraction was 86% by using the improved Alexnet convolutional neural network. Compared with traditional methods, it has been improved greatly, which provides effective means for the identification of benign and malignant thyroid nodules.
Objective To evaluate the diagnostic performance of thyroid imaging reporting and data system (TI-RADS)classification and elastography in differentiating benign and malignant thyroid nodules according to size. Methods A total of 222 thyroid nodules (209 cases) with solid or predominantly solid internal contentdiagnosed at pathological findings were enrolled in this study, all the 209 cases underwent surgery in our hospital from Jan.2014 to Jun.2014. The diagnostic performance of TI-RADS classification and elastography in differentiating benign and malignant thyroid nodules according to size nodules were evaluated (for≤1 cm and > 1 cm respectively). Results By using TI-RADS category and elastography, 178 thyroid nodules were diagnosed as malignant nodules, and 44 thyroid nodules were diagnosed as benign nodules. The high predictive factors for malignant thyroid nodules of > 1 cm were irregular shape(OR=6.376), microcalcification(OR=21.525), and capsule invasion(OR=3.852), P < 0.05. The factors for thyroid nodules of≤1 cm were anteroposterior to transverse diameter ratio≥1(OR=3.406), capsule invasion (OR=3.922), and high elastography score(OR=1.606), that suggested the possibility of malignant (P < 0.05). For nodules of > 1 cm, the sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and Youden index of TI-RADS combining with elastography were 98.3%(59/60), 68.6%(24/35), 87.4%(83/95), 84.3%(59/70), 96.0% (24/25), and 66.9% respectively; for nodules of≤1 cm, the sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and Youden index were 98.5%(67/68), 30.5%(18/59), 66.9%(85/127), 62.0%(67/108), 94.7%(18/19), and 29.0% respectively. Conclusion TI-RADS category combined with elastography for diagnosis of thyroid nodules in different size has just a bit differently diagnostic indicators, but that play a high diagnostic performance on the thyroid nodules with maximum diameter > 1 cm.
ObjectiveTo investigate the prevalence of thyroid nodules and its influencing factors among medical staff of a hospital in Hengshui. MethodsA total of 2 190 medical workers in a hospital in Hengshui who underwent physical examinations in the Medical Examination Center of Harrison International Peace Hospital from January 2012 to July 2013 were included as the study subjects. Through physical examination, thyroid ultrasonography and questionnaire, we collected data and investigated the prevalence of thyroid nodules and its influencing factors. ResultsThe prevalence rate of thyroid nodules was 39.45% (864/2 190) in these medical staff. The prevalence rates of thyroid nodules in males and females were 32.34% (238/736) and 42.92% (624/1 454), respectively, with statistically significant difference between them (χ2=22.913, P<0.001). With increasing age, the prevalence of thyroid nodules increased gradually, and the prevalence of the same sex increased gradually too (P<0.001). The prevalence rate of women was significantly higher than that of men for each age group (P<0.05). Among the 864 subjects, there were 550 (63.66%) with multiple nodules and 314 (36.34%) with solitary nodules; there were 454 (52.55%) with nodules in both lobes of thyroid, 238 (27.54%) with nodules in the right lobe of thyroid, and 172 (19.91%) with nodules in the left lobe of thyroid; there were 504 (58.33%) with solid modules, 172 (19.91%) with cystic nodules, 114 (13.19%) with solid and cystic nodules, and 74 (8.56%) with calcification. Logistic regression analysis showed that gender, age, educational degree, drinking history, taste habit, stress, agitation were correlated with the development of thyroid nodules. ConclusionThe prevalence of thyroid nodules is high among medical staff of this hospital. Thyroid nodules can be prevented through healthy lifestyle, limiting alcohol intake, emotional control, and relieving stress. Early diagnosis and early treatment should be done to prevent the development and spread of thyroid nodules.
ObjectiveTo explore the diagnostic value of ultrasound elastography (USE) combined with long non-coding RNA actin filament associated protein 1 anti-sense RNA 1 (AFAP1-AS1) mRNA in thyroid fine-needle aspiration (FNA) wash-out fluid for distinguishing benign from malignant thyroid nodules. MethodsThe patients with thyroid nodules who were treated in the Shenzhen Futian District Second People’s Hospital from January 2020 to June 2022 were collected. Before operation, the patients’ thyroid nodules were evaluated by the USE score and the AFAP1-AS1 mRNA in the thyroid FNA wash-out fluid was detected. The pathological result of the thyroid nodule after operation was as a gold standard for diagnosis of malignant thyroid nodules. The clinical diagnostic value of USE score combined with AFAP1-AS1 mRNA in the FNA wash-out fluid of the benign and malignant thyroid nodules were analyzed. ResultsA total of 174 thyroid nodules (124 patients) were detected in this study, of which 62 (45 patients) were histologically diagnosed as malignant. There was a statistical difference in the comparison of the composition ratio of USE score grading between the benign and malignant thyroid nodules (Z=8.82, P<0.001). The point of USE of the benign thyroid nodules was statistically lower than that of the malignant thyroid nodules [2.28±1.16 vs. 4.26±1.01, mean difference (MD) and 95% confidence interval (95%CI)=2.98 (2.76, 3.20), t=30.85, P<0.001]. The AFAP1-AS1 mRNA in the FNA wash-out fluid of the malignant thyroid nodules was statistically higher than that of the benign thyroid nodules [1.45±0.27 vs. 1.13±0.16, MD (95%CI)=1.45(1.39, 1.50), t=10.69, P<0.001]. Pearson correlation analysis showed that there was a positive correlation between the USE score of thyroid nodules and the expression of AFAP1-AS1 mRNA in the FNA wash-out fluid (r=0.58, P<0.001). The sensitivity and specificity of USE score in combination with expression of AFAP1-AS1 mRNA in the FNA wash-out fluid for diagnosing the malignant thyroid nodules by receiver operating characteristic (ROC) curve was 93.5% and 88.4% respectively. The area under the ROC curve (95%CI) was 0.91 (0.86, 0.96). Conclusion According to preliminary results of this study, USE score combined with AFAP1-AS1 mRNA in the thyroid FNA wash-out fluid is more sensitive and shows a potential diagnostic performance than USE score or AFAP1-AS1 mRNA detection alone for distinguishing benign from malignant thyroid nodules.
Objective To discuss the nursing measures for thyroid nodule patients who undergo core-needle biopsy (CNB) guided by ultrasound. Methods We retrospectively analyzed the experiences and main points of nursing for 1 900 thyroid nodule patients who underwent CNB guided by ultrasound between June 2010 and May 2014. Results All the 1 900 patients underwent CNB successfully. The nursing time was between 5 and 15 minutes, averaging (8.0±3.7) minutes. Complications included hematoma in 25 patients (1.3%) and needle syncope reaction in 30 patients (1.6%), which were cured through symptomatic treatment. No complications such as nerve injury, anesthesia accident or death occurred. No medical disputes happened due to specimen errors or loss. The success rate of specimen collection was 98.4% (1 870/1 900), and the diagnostic accuracy was 95.3% (1 812/1 900). Conclusions Ultrasonography-guided CNB is a safe and reliable operation with a high success rate, high diagnosis accuracy and few complications. Being familiar with the process of nursing cooperation and correct disposal and transfer of biopsy specimens are crucial for successful CNB in patients with thyroid nodules.
Objective To summary the change of surgical approaches, indications, contraindications, and complications of endoscopic thyroidectomy. Methods Endoscopic thyroidectomy related literatures at home and abroad were collected to summary the change of surgical approaches, indications, contraindications, and complications of it. Results The approaches of endoscopic thyroidectomy had their own advantages, wherein breast and complete areola approaches were mainstream surgical procedures; the indications were closely correlated with surgeons and devices, and there were no uniform indications yet. With the advance in endoscopic techniques, the contraindications would be gradually resolved, and the assistance of new devices and the accumulation of experience reduced the incidence of surgical complications, but there were still probability of occurrence of such complications. Conclusion Endoscopic thyroidectomy is the main means of treating thyroid nodules, which will be more widely used with the advance in techniques, the innovation of devices, the expansion of indications as well as the overcoming of contraindications.
Objective To report our experience in using The Bethesda System for Reporting Thyroid Cytopathology(TBSRTC), and to investigate the diagnostic value of the system based on the cytologic-histologic result. Methods Pathological data of 2 257 thyroid nodules classified by TBSRTC which were obtained from the Department of Pathology of West China Hospital between Jan.2010 to Dec.2012 were collected and analyzed, to investigate the diagnostic evaluation indicators, such as the sensitivity, specificity, and diagnostic accuracy of the system based on the cytologic-histologic result. Results Of the 2 257 thyroid nodules, 442 (19.6%) were diagnosed as categoryⅠ, 1 184(52.4%) were diagnosed as categoryⅡ, 216(9.6%) were diagnosed as categoryⅢ, 38(1.7%) were diagnosed as categoryⅣ, 172(7.6%) were diagnosed as categoryⅤ, 205(9.1%) were diagnosed as categoryⅥ. Using TBSRTC categoryⅡas the boundary point of diagnosing benign and malignant diseases, the sensitivity, specificity, and diagnostic accuracy were 93.7%(236/252), 86.6%(323/373), and 89.4%(559/625)respectively, while categoryⅢwas excluded from analysis. When including categoryⅢinto analysis, the sensitivity, specificity, and diagnostic accuracy were 94.3%(267/283), 74.9%(323/431), and 82.6%(590/714)respectively. Conclusion The validity of TBSRTC was high at our institution. TBSRTC has proven to be an accurate and reliable approach for the diagnosis of thyroid nodules.