Hybrid deep learning assisted multi classification: Grading of malignant thyroid nodules.

Thyroid nodules are commonly diagnosed with ultrasonography, which includes internal characteristics, varying looks, and hazy boundaries, making it challenging for a clinician to differentiate between malignant and benign forms based only on visual identification. The advancement of AI, particularly DL, provides significant breakthroughs in the domain of medical image identification. Yet, there are certain obstacles to achieving accuracy as well as efficacy in thyroid nodule detection. The thyroid nodules in this study are detected and classified using an inventive hybrid deep learning-assisted multi-classification method. The median blur method is applied in this work to eliminate the salt and pepper noise from the image. Then MPIU-Net-based segmentation is utilized to segment the image. The LGBPNP-based features are retrieved from the segmented image to obtain a single histogram sequence of the LGBP pattern in addition to other features like extraction of multi-texton and LTP-based features. After the feature extraction, the data augmentation process is applied and then the features are fed to the hybrid classification-based nodule classification model that comprises Deep Maxout and CNN, this hybrid classification trains the features and predicts the thyroid nodule. Additionally, the TIRADS score classification is used for the projected malignant thyroid nodule coupled with statistical features collected from the segmented. The DBNAAF with transfer learning model is employed to classify the grading of malignant thyroid nodules, where the weights of the model are learned with transfer learning. The MCC of the Hybrid Model is 0.9445, whereas the DCNN is 0.6858, YOLOV3-DMRF is 0.7229, CNN is 0.7780, DBN is 0.7601, Bi-GRU is 0.7038, Deep Maxout is 0.7528, and RNN is 0.8522, respectively.

Blood leukocytes as a non-invasive diagnostic tool for thyroid nodules: a prospective cohort study.

BACKGROUND: Thyroid nodule (TN) patients in China are subject to overdiagnosis and overtreatment. The implementation of existing technologies such as thyroid ultrasonography has indeed contributed to the improved diagnostic accuracy of TNs. However, a significant issue persists, where many patients undergo unnecessary biopsies, and patients with malignant thyroid nodules (MTNs) are advised to undergo surgery therapy. METHODS: This study included a total of 293 patients diagnosed with TNs. Differential methylation haplotype blocks (MHBs) in blood leukocytes between MTNs and benign thyroid nodules (BTNs) were detected using reduced representation bisulfite sequencing (RRBS). Subsequently, an artificial intelligence blood leukocyte DNA methylation (BLDM) model was designed to optimize the management and treatment of patients with TNs for more effective outcomes. RESULTS: The DNA methylation profiles of peripheral blood leukocytes exhibited distinctions between MTNs and BTNs. The BLDM model we developed for diagnosing TNs achieved an area under the curve (AUC) of 0.858 in the validation cohort and 0.863 in the independent test cohort. Its specificity reached 90.91% and 88.68% in the validation and independent test cohorts, respectively, outperforming the specificity of ultrasonography (43.64% in the validation cohort and 47.17% in the independent test cohort), albeit with a slightly lower sensitivity (83.33% in the validation cohort and 82.86% in the independent test cohort) compared to ultrasonography (97.62% in the validation cohort and 100.00% in the independent test cohort). The BLDM model could correctly identify 89.83% patients whose nodules were suspected malignant by ultrasonography but finally histological benign. In micronodules, the model displayed higher specificity (93.33% in the validation cohort and 92.00% in the independent test cohort) and accuracy (88.24% in the validation cohort and 87.50% in the independent test cohort) for diagnosing TNs. This performance surpassed the specificity and accuracy observed with ultrasonography. A TN diagnostic and treatment framework that prioritizes patients is provided, with fine-needle aspiration (FNA) biopsy performed only on patients with indications of MTNs in both BLDM and ultrasonography results, thus avoiding unnecessary biopsies. CONCLUSIONS: This is the first study to demonstrate the potential of non-invasive blood leukocytes in diagnosing TNs, thereby making TN diagnosis and treatment more efficient in China.

Thyroid nodule malignancy is associated with increased non-invasive hepatic fibrosis scores in metabolic subjects.

Introduction: Thyroid cancer incidence is increasing, and adiposity-related conditions are gaining space in its pathogenesis. In this study, we aimed to detect any anthropometric, biohumoral, and clinical features that might be associated with thyroid nodule malignancy, potentially representing novel non-invasive markers of thyroid cancer. Materials and methods: The study was conducted in a group of 142 consecutive outpatients (47 men and 95 women) who underwent fine-needle aspiration biopsy/cytology (FNAB/C) due to suspicion of malignancy from January 2018 to September 2022. We compared lipid and glycemic blood profiles as well as non-invasive liver fibrosis indexes such as aspartate aminotransferase (AST) to alanine aminotransferase (ALT) ratio (AAR), AST to platelet ratio index (APRI), and fibrosis index based on four factors (FIB-4) between patients with benign and malignant newly diagnosed nodules. Then, we performed receiver operating characteristic (ROC) analysis to assess their best cutoff values for discrimination of malignant nodules and chi-squared test to evaluate the association of specific dysmetabolic conditions with malignancy. To understand whether and to what degree dysmetabolic conditions increased the risk of thyroid nodule malignancy, we also calculated the odds ratio (OR) of the main biomarkers. Results: After FNAB/C, 121 (85%) patients were diagnosed with benign thyroid nodules, while 21 (15%) individuals were diagnosed with thyroid cancer. Comparing patients with benign and malignant nodules, we found that individuals with thyroid cancer exhibited increased body mass index (BMI) (p = 0.048) and fasting plasma glucose (p = 0.046). Intriguingly, considering non-invasive scores for liver fibrosis, subjects with thyroid cancer presented increased AAR (p < 0.001) and APRI (p = 0.007), and these scores were associated with malignancy (p < 0.005) with OR = 7.1 and OR = 5, respectively. Moreover, we showed that only in the cancer group, low levels of vitamin D correlated with stigmata of impaired metabolism. Discussion: In our study, AAR and APRI scores were associated with thyroid nodule malignancy and could be used to predict it and to speed up the diagnostic process. From a pathogenic point of view, we speculated that metabolic-associated fatty liver disease (MAFLD) along with hyperglycemia and vitamin D deficiency may represent putative drivers of thyroid carcinogenesis.

Recurrent Metastatic Colorectal Adenocarcinoma to the Thyroid Gland Presenting With Vocal Cord Paralysis and Inspiratory Stridor.

The most common sites for metastases of colorectal cancer include the liver, lungs, brain, and regional lymph nodes. However, a limited number of reported cases describe colon cancer metastasis to the thyroid gland. Metastatic colorectal adenocarcinoma to the thyroid gland is rare. The majority of these cases with colon cancer metastases to the thyroid gland are diagnosed years after initial treatment of colon cancer. The discovery is usually made after routine surveillance imaging, and often patients have minimal or absent symptoms. We report a case of a recurrence of metastatic colorectal adenocarcinoma to the thyroid gland presenting with vocal cord paralysis and inspiratory stridor.

Accuracy of Ultrasound Scans as Compared to Fine Needle Aspiration Cytology in the Diagnosis of Thyroid Nodules.

INTRODUCTION: Thyroid nodules (TNs) are among the more common findings on physical examinations. Due to the fear of the TN harboring malignancy and with the increasing incidence of thyroid cancer, ultrasound (US) scanning is used as an important diagnostic tool in the assessment of a TN. The American College of Radiology's Thyroid Imaging Reporting and Data System (TI-RADS) was established based on specific patterns composed of two or more features. According to the TI-RADS guidelines, a suspicious nodule by US findings should undergo fine-needle aspiration cytology (FNAC), in which results would guide further management. OBJECTIVE: This study was carried out to assess the accuracy of US as compared to FNAC in the diagnosis of a thyroid nodule. METHODOLOGY: This retrospective study involved 213 cases that were sent for FNAC after having done a US scan of the thyroid. Data was gathered from all patient files that were referred for FNAC thyroid between 01/02/2018 and 30/06/2021 in Al-Ahli Hospital in the state of Qatar. The US scans were interpreted and reported according to the TI-RADS criteria. The FNAC samples were interpreted and reported according to the Bethesda System for Reporting Thyroid Cytopathology. Data were tabulated and analyzed with Excel (Microsoft, Redmond, WA, USA) and SPSS version 25 (IBM Corp., Armonk, NY, USA). RESULTS: The study showed that US had a sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 73.9%, 72.6%, 24.6% and 95.8%, respectively, with a significant association between the results of US and the results of FNAC (X2 (1, n = 213) = 20.295, p < .001) and a significant positive correlation (phi coefficient = .309, p < .001). In addition, the data showed that the odds for having a positive FNAC were 7.519 (95% CI: 2.811, 20.112) times greater for cases with positive US compared with cases with negative US. The relative risk of having a positive FNAC when the US was positive was 5.913 (95% CI: 2.440, 14.332) times greater compared to when the US was negative. CONCLUSION: While our results showed that US cannot be solely relied on in diagnosing TNs, they did show that US can reliably rule out a malignancy in TNs. Recent studies have been showing increasing accuracy of US in diagnosing TNs and more studies are needed to explore this topic.

Could the level of nitrite/nitrate contribute to malignant thyroid nodule diagnostics?

Thyroid nodules are among highly prevalent thyroid diseases. To make a distinction between benign and malignant thyroid nodules are of cumbersome significance for each endocrinologist. There is no unique and completely accurate diagnostic test, method, or even biomarker that points to a malignant thyroid nodule. Many studies in modern thyroidology are conducted to determine the usefulness of individual biomarkers, which could help clinicians detect thyroid nodules' potential malignant nature. One interesting biomarker with a promising diagnostic potential for the thyroid gland pathological conditions is nitric oxide (NO). Inducible nitric oxide synthase expression is increased in thyroiditis cases and even more in thyroid carcinoma cases, directly connected with increased NO levels in both pathological conditions. We hypothesize that the basal levels of nitrite/nitrate in serum and biopsy washout could indicate nodules' malignant nature.

Ultrasound Image-Based Diagnosis of Malignant Thyroid Nodule Using Artificial Intelligence.

Computer-aided diagnosis systems have been developed to assist doctors in diagnosing thyroid nodules to reduce errors made by traditional diagnosis methods, which are mainly based on the experiences of doctors. Therefore, the performance of such systems plays an important role in enhancing the quality of a diagnosing task. Although there have been the state-of-the art studies regarding this problem, which are based on handcrafted features, deep features, or the combination of the two, their performances are still limited. To overcome these problems, we propose an ultrasound image-based diagnosis of the malignant thyroid nodule method using artificial intelligence based on the analysis in both spatial and frequency domains. Additionally, we propose the use of weighted binary cross-entropy loss function for the training of deep convolutional neural networks to reduce the effects of unbalanced training samples of the target classes in the training data. Through our experiments with a popular open dataset, namely the thyroid digital image database (TDID), we confirm the superiority of our method compared to the state-of-the-art methods.

Diagnostic performance of combination of ultrasound elastography and BRAF gene detection in malignant thyroid nodule: a retrospective study.

OBJECTIVES: The aims of our study were to explore the preoperative diagnostic value of ultrasound elastography combined with BRAF gene detection in malignant thyroid nodule, and find whether shear wave elastography (SWE) combined with BRAF gene detection can improve the diagnostic sensitivity and specificity. METHODS: From 1480 patients with thyroid nodule examined between January 2015 and December 2017, a retrospective analysis was performed on 161 patients who underwent thyroidectomy. Diagnosis was confirmed by postoperative pathology, including 139 malignant thyroid nodules and 22 benign thyroid nodules. All the patients underwent SWE, BRAF gene detection, and the combination for their preoperative evaluation. The sensitivities, specificities, and accuracies of SWE, BRAF gene detection, and the combination for detection of malignant thyroid nodules were calculated and then compared using Fisher's exact probability test, based on the original preoperative reports and postoperative pathology. A receiver-operating characteristic curve analysis was performed to compare the diagnostic performance of SWE, BRAF gene detection, and combination for detecting malignant thyroid nodules. RESULTS: Based on the original preoperative reports and postoperative pathology, SWE, BRAF gene detection, and the combination showed sensitivities of 88.67%, 78.41%, 92.8%, and specificities of 72.77%, 77.27%, 95.45%. A correct diagnosis was obtained in 85.82%, 78.26%, 93.16% and missed diagnosis rates were 12.23%, 21.58%, and 7.19%. The sensitivities, specificities, and correct diagnosis rate in the combination group were significantly higher than any single detection method (P<0.05). The missed diagnosis rate in the combination group was significantly lower than any single detection method (P<0.05). The receptor operating characteristics curve analysis showed a significantly higher diagnostic performance for the combination than for SWE and BRAF gene detection (P<0.05). The interobserver agreement for detecting malignant thyroid nodule was better for the combination than for SWE or BRAF gene detection alone. CONCLUSION: For the detection of a malignant thyroid nodule, SWE combined with BRAF gene detection was more sensitive and showed a higher diagnostic performance than SWE or BRAF gene detection alone.

Color Doppler ultrasonography diagnostic value in detection of malignant nodules in cysts with pathologically proven thyroid malignancy: a systematic review and meta-analysis.

AIMS: Today, the color Doppler ultrasonography is used to further evaluate suspected malignant tumors. This study investigates the malignant thyroid nodules using color Doppler. METHODS: After extracting true positive, false positive, false negative, and true negative among included studies, a quality was evaluated by the Quality Assessment of Diagnostic Accuracy Studies-2 tool. Sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio (with 95% confidence interval) were found using a random effect model. Summary receiver operating characteristic curves (SROC) were used to assess relationship between sensitivity and specificity. The area under the curve of the SROC was calculated to estimate the performance of color Doppler ultrasound to distinguish malignant thyroid nodules. Our registration code in PROSPERO is CRD42018111198. RESULTS: Of 1125 articles, 288 articles were selected for the further investigation. After excluding irrelevant and poor articles, 20 studies were included for the meta-analysis. According to a random effect model, the pooled sensitivity and specificity of color Doppler ultrasound to distinguish malignant thyroid nodules were estimated as 0.74 (95% CI 0.62-0.83; [Formula: see text]) and 0.70 (95% CI 0.56-0.81; [Formula: see text]), respectively. The SROC curve consists of representing the paired results for sensitivity and specificity. According to SROC, AUC = 0.78 (95% CI 0.74-0.81) is between 0.75 and 0.92, so that color Doppler ultrasound has a good accuracy. CONCLUSION: Color Doppler is a valuable non-invasive method for evaluating thyroid nodules, and it is a high-sensitivity diagnostic tool for assessing thyroid nodules. Resistive index > 0.75 and a pattern III or more in color Doppler predicts malignant with the confidence. Due to its precision, cost-efficiency, easy access, and non-invasive nature, color Doppler should be included in the standard clinical protocol for the decision-making period and the treatment evaluation.

Bayesian analysis of high-resolution ultrasonography and guided fine needle aspiration cytology in diagnosis of palpable thyroid nodules / Análise bayesiana de ultrassonografia de alta resolução e citologia de aspiração poragulha fina guiada no diagnóstico de nódulos palpáveis da tireoide

Abstract Introduction To evaluate diagnostic accuracy of high-resolution ultrasonography in differentiation of benign and malignant thyroid nodules in comparison to results of guided fine needle aspiration cytology based on the Bayes rule. Objective To assess the validity of ultrasonography results of thyroid nodules in comparison to guided fine needle aspiration cytology findings. Methods This study was done on randomly chosen 80 patients presented with palpable thyroid nodules, undergone real-time sonographic evaluation of thyroid nodules to characterize features, internal consistency, margins, echotexture, calcification, peripheral lucent halo and vascularity. Ultrasonography guided fine needle aspiration cytology studies of thyroid nodules were done. Results Palpable thyroid nodules were highly prevalent in fourth and fifth decades of life with female-male ratio, 4:1. Solid internal consistency was demonstrated by 75% malignant nodules. Hypoechogenicity and intra-nodular micro-calcifications were observed in 92% malignant nodules; 83% malignant nodules had intra-nodular vascularity and absence of peripheral halo. The pre-test prevalence of malignant nodules in the targeted population was 17.5%. As type I error, 2.5% false-positive cases and as type II error, 5.0% false-negative cases were detected. Values of sensitivity and specificity of the ultrasonography test were 71.43 and 96.97%, respectively. Conclusion Malignant thyroid nodules demonstrated ultrasonography characteristics of hypoechoic texture, intra-nodular micro-calcifications, solid consistency, internal vascularity and absence of peripheral halo. The ultrasonography test has 92.5% diagnostic accuracy to differentiate malignant from benign lesions in comparison to the gold standard fine needle aspiration cytology test.

Resumo Introdução Avaliar a precisão diagnóstica da ultrassonografia de alta resolução na diferenciação de nódulos tireoidianos benignos e malignos em comparação com os resultados da citologia de aspiração por agulha fina baseada na regra de Bayes. Objetivo Avaliar a validade dos resultados da USG de nódulos da tireoide em comparação com os resultados obtidos por citologia de aspiração por agulha fina. Método Este estudo foi feito em 80 pacientes selecionados aleatoriamente, que apresentavam nódulos palpáveis da tireoide, submetidos à avaliação ultrassonográfica em tempo real de nódulos da tireoide para estabelecer características, consistência interna, margens, ecotextura, calcificação, halo lucente periférico e vascularização. Foram feitos estudos por citologia de aspiração por agulha fina guiados pela USG dos nódulos de tireoide. Resultados Nódulos palpáveis da tireoide foram altamente prevalentes na quarta e quinta décadas de vida com uma razão sexo feminino-masculino de 4:1. A consistência interna sólida foi demonstrada em 75% de nódulos malignos. Hipoecogenicidade e microcalcificações intranodulares foram observadas em 92% de nódulos malignos; 83% dos nódulos malignos apresentaram vascularidade intranodular e ausência de halo periférico. A prevalência pré-teste de nódulos malignos na população alvo foi de 17,5%. Como erro tipo I, houve 2,5% de casos falso-positivos e como erro tipo II, foram detectados 5,0% de casos falso-negativos. Os valores de sensibilidade e especificidade do exame por USG foram de 71,43 e 96,97%, respectivamente. Conclusão Nódulos malignos da tireoide mostraram características de textura hipoecoica, microcalcificações intranodulares, consistência sólida, vascularidade interna e ausência de halo periférico na USG. O exame por USG tem 92,5% de precisão diagnóstica para diferenciar lesões malignas de benignas em comparação com o padrão ouro da citologia de aspiração por agulha fina.

Bayesian analysis of high-resolution ultrasonography and guided fine needle aspiration cytology in diagnosis of palpable thyroid nodules.

INTRODUCTION: To evaluate diagnostic accuracy of high-resolution ultrasonography in differentiation of benign and malignant thyroid nodules in comparison to results of guided fine needle aspiration cytology based on the Bayes rule. OBJECTIVE: To assess the validity of ultrasonography results of thyroid nodules in comparison to guided fine needle aspiration cytology findings. METHODS: This study was done on randomly chosen 80 patients presented with palpable thyroid nodules, undergone real-time sonographic evaluation of thyroid nodules to characterize features, internal consistency, margins, echotexture, calcification, peripheral lucent halo and vascularity. Ultrasonography guided fine needle aspiration cytology studies of thyroid nodules were done. RESULTS: Palpable thyroid nodules were highly prevalent in fourth and fifth decades of life with female-male ratio, 4:1. Solid internal consistency was demonstrated by 75% malignant nodules. Hypoechogenicity and intra-nodular micro-calcifications were observed in 92% malignant nodules; 83% malignant nodules had intra-nodular vascularity and absence of peripheral halo. The pre-test prevalence of malignant nodules in the targeted population was 17.5%. As type I error, 2.5% false-positive cases and as type II error, 5.0% false-negative cases were detected. Values of sensitivity and specificity of the ultrasonography test were 71.43 and 96.97%, respectively. CONCLUSION: Malignant thyroid nodules demonstrated ultrasonography characteristics of hypoechoic texture, intra-nodular micro-calcifications, solid consistency, internal vascularity and absence of peripheral halo. The ultrasonography test has 92.5% diagnostic accuracy to differentiate malignant from benign lesions in comparison to the gold standard fine needle aspiration cytology test.

Malignancy in solitary thyroid nodule: A clinicoradiopathological evaluation.

BACKGROUND: Thyroid nodules are common. They can be either benign or malignant. Solitary thyroid nodules (STN) have a high likelihood of being malignant. They should be characterized properly for optimum management. MATERIALS AND METHODS: In this study, we have analyzed our departmental data over a period of 5 years. All the patients who presented to the outpatient department with a clinically detected STN were included in the study group. Our approach was individualized. Preoperative ultrasonography (USG) and fine-needle aspiration cytology were planned in all these patients. Hemi thyroidectomy and total thyroidectomy with and without neck dissection were performed wherever appropriate. RESULTS: There were 162 cases of clinically detected STN. USG findings were available in 146 cases. Postoperative histopathology was reported as malignant in 58 cases. Malignant STN was more likely in males. Ultrasonographically detected solid STN were more prone for malignancy as compared to multinodular goiter (P = 0.000) Presence of micro calcification and cervical lymphadenopathy were more commonly noted in malignant thyroid swellings. CONCLUSION: Solitary thyroid nodules do have a high likelihood of harboring a malignancy. Solid echogenicity, micro calcification and cervical lymphadenopathy on USG were seen more frequently in malignant nodules.

Sonographic Analysis of Malignant Thyroid Nodules by Surgeon / 대한내분비외과학회지

PURPOSE: Ultrasound is most effective study for evaluating thyroid nodules. In this review, we discuss that sonographic findings to differentiate benign from malignant nodules and suggest recommendations for indications of fine needle aspiration biopsy and thyroid nodule management. METHODS: Sonographic scans of 206 thyroid nodules in 164 patients were candidated for this study. We evaluated sonographic findings by shape, calcification, margin, and echogenicity, retrospectively. Sonographic findings that suggested malignancy included microcalcifications, a speculated margin, marked hypoechogenicity and a shape that was taller than wide. The final diagnosis of lesion as benign (n=180) or malignant (n=26) was confirmed by fine needle aspiration biopsy and follow-up (>6 months). We demonstrated the difference of the sensitivity, specificity, positive predictive value, negative predictive value and accuracy. RESULTS: Of 206 thyroid nodules, 26 were malignant. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy based on our sonographic classification method were 84.6%, 73.9%, 31.9%, 97.0% and 75.2%. CONCLUSION: Sonography can be helpful for making the differentiation between and malignant nodules. So, when well trained surgeon find thyroid nodules on sonography, we can make correct diagnosis of malignant nodules.

Differentiation of Benign and Malignant Thyroid Nodules: Sonographic Analysis and a New Scoring System

PURPOSE: To analyze the synthetically created sonographic features of thyroid nodules, we present here a new scoring system for the sonographic features that are suggestive of malignant thyroid nodules. We also evaluated the accuracy and clinical significance of this system. MATERIALS AND METHODS: In this study, we included 725 thyroid nodules of 405 patients that were pathologically proven by USG-guided percutaneous fine-needle aspiration biopsy (FNAB) or surgery. Two radiologists analyzed the sonographic features according to the internal content, margin, echogenecity, shape and calcification. We scored from 0 point to 2 point for each feature, and then we calculated the total scores and classified them as three groups according to the total score such as low risk (0-3), intermediate risk (4-6) or high risk (7-10). We demonstrated the difference of the frequency and the positive predictive value among the three groups by using the Chi-square test (p<0.005). RESULTS: For 725 nodules, 654 (90.2%) were benign and 71 (9.8%) were malignant. For 589 nodules classified as low risk, 10 (1.7%) were malignant. For 102 nodules classified as intermediate risk, 32 (31.4%) were malignant. For 34 nodules classified as high risk, 29 (85.8%) were malignant. There was a statistically significant difference in the frequency and positive predictive value of malignancy among the three groups (p<0.001). CONCLUSION: There was a statistically significant difference in the frequency and positive predictive value of malignancy among the three groups for the new scoring system presented in this study to analyze the synthetically sonographic features of thyroid nodules. So, we think that sonography can be helpful for making the differentiation between benign and malignant nodules. When we find thyroid nodules on sonography, we can reduce the unnecessary FNAB and we can diagnose malignant nodules at an earlier stage.