Diagnosis Of Thyroid Nodules Research Articles

Stratified application of gene expression in diagnosis of thyroid nodules

Objective: To investigate the value of BRAF V600E and multigene detection and stratified application for the diagnosis of thyroid nodules. Methods: A total of 1 117 patients with thyroid nodules resection at Nanjing Gulou Hospital from December 2020 to July 2022 were enrolled in the study. Fine needle aspiration (FNA) and core biopsy samplings were performed for cytopathologic examination and genetic testings; the findings were combined with BSRTC classification. The diagnostic performance of BRAF V600E and multigene detection were compared. Results: Among the 1, 117 patients who underwent thyroid nodules resection, 285 were male and 832 were female, with a median age of 46 years (range: 24-76 years). Postoperative histopathologic examination confirmed 1 040 cases of thyroid cancer and 77 cases of benign nodules. The sensitivity (87.0% vs. 80.8%, P<0.01) and diagnostic accuracy (87.9% vs. 82.1%, P<0.01) of multigene detection were significantly higher than those of BRAF V600E detection. The result of multigene detection showed that BRAF V600E mutation was the most common finding, followed by CCDC6-RET (E1-E12) fusion, ETV6-NTRK3 fusion, and KRAS mutation. Multigene detection had a higher sensitivity (81.9% vs. 72.8%, P<0.01) and lower cancer risk in wild-type (47.6% vs. 57.7%, P=0.069) than BRAF V600E detection in BSRTCⅠ-Ⅴ lesions. Compared with BRAF V600E detection, multigene had no significant difference of sensitivity in BSRTC Ⅰ lesions, but significantly higher sensitivity (86.3% vs 74.0%, P<0.01) in BSRTC Ⅲ lesions. Conclusions: Genetic detection can be used as an effective tool for the diagnosis of thyroid nodules. A stratified application of molecular markers in the diagnosis of thyroid nodules is proposed. Combined with FNA, single gene or multigene detection both can effectively assist in the diagnosis of thyroid nodules. Moreover, multigene detection is superior to single gene detection. For BSRTC Ⅲ lesion with wild-type BRAF, multigene detection can be considered with a repeated FNA.

Accuracy of ACR-TIRADS in Assessment and Diagnosis of Thyroid Nodules in Patients Underwent Thyroid Surgery in Taksin Hospital

Accuracy of ACR-TIRADS in Assessment and Diagnosis of Thyroid Nodules in Patients Underwent Thyroid Surgery in Taksin Hospital

A Cytohistologic Correlation Study of Thyroid Lesions: Evaluation of Diagnostic Accuracy and Pitfalls of Fine Needle Aspiration Cytology.

Background Only about 5% of palpable thyroid nodules are malignant; the rest are entirely benign. In order to reduce the number of unnecessary treatments and properly identify situations that need surgical intervention, it is essential to distinguish between benign and malignant lesions prior to surgery. There exists a "grey zone" in thyroid cytology characterized by a significant decrease in diagnostic accuracy, making it difficult to precisely classify the lesion and leading to discrepancies. Aims and objectives The study aims to accomplish the following objectives: (1) assess the prevalence of thyroid lesions according to age and sex; (2) evaluate the accuracy of the fine needle aspiration cytology (FNAC) in diagnosing thyroid conditions; and (3) investigate the causes of cytohistological discordance within the context of this study. Materials and methods In our five-year study of thyroid lesions, 125 cases were studied for cytohistological correlation. Discrepant and likened FNAs were classified according to the diagnostic findings. A review of the cytological smears and histological sections was conducted. Results The cytological diagnoses were correlated with histopathology in 109 cases (90.83%). A total of 11 cases (09.16%) were discrepant. There were no false positives (FPs). The causes of false negative (FN) diagnoses in our study can be attributed to geographic misses and failure to recognize dual pathologies. Conclusion FNA is a very precise and time-saving technique for the diagnosis and subsequent management of palpable thyroid nodules. Patients having thyroidectomies have a much higher malignant yield, and the frequency of procedures performed on the thyroid is decreased. When FNA interpretation based on strict specimen sufficiency standards is considered along with clinical and imaging findings, the occurrence of FN and FP diagnoses is expected to decrease.

Collaborative Enhancement of Consistency and Accuracy in US Diagnosis of Thyroid Nodules Using Large Language Models.

Background Large language models (LLMs) hold substantial promise for medical imaging interpretation. However, there is a lack of studies on their feasibility in handling reasoning questions associated with medical diagnosis. Purpose To investigate the viability of leveraging three publicly available LLMs to enhance consistency and diagnostic accuracy in medical imaging based on standardized reporting, with pathology as the reference standard. Materials and Methods US images of thyroid nodules with pathologic results were retrospectively collected from a tertiary referral hospital between July 2022 and December 2022 and used to evaluate malignancy diagnoses generated by three LLMs-OpenAI's ChatGPT 3.5, ChatGPT 4.0, and Google's Bard. Inter- and intra-LLM agreement of diagnosis were evaluated. Then, diagnostic performance, including accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC), was evaluated and compared for the LLMs and three interactive approaches: human reader combined with LLMs, image-to-text model combined with LLMs, and an end-to-end convolutional neural network model. Results A total of 1161 US images of thyroid nodules (498 benign, 663 malignant) from 725 patients (mean age, 42.2 years ± 14.1 [SD]; 516 women) were evaluated. ChatGPT 4.0 and Bard displayed substantial to almost perfect intra-LLM agreement (κ range, 0.65-0.86 [95% CI: 0.64, 0.86]), while ChatGPT 3.5 showed fair to substantial agreement (κ range, 0.36-0.68 [95% CI: 0.36, 0.68]). ChatGPT 4.0 had an accuracy of 78%-86% (95% CI: 76%, 88%) and sensitivity of 86%-95% (95% CI: 83%, 96%), compared with 74%-86% (95% CI: 71%, 88%) and 74%-91% (95% CI: 71%, 93%), respectively, for Bard. Moreover, with ChatGPT 4.0, the image-to-text-LLM strategy exhibited an AUC (0.83 [95% CI: 0.80, 0.85]) and accuracy (84% [95% CI: 82%, 86%]) comparable to those of the human-LLM interaction strategy with two senior readers and one junior reader and exceeding those of the human-LLM interaction strategy with one junior reader. Conclusion LLMs, particularly integrated with image-to-text approaches, show potential in enhancing diagnostic medical imaging. ChatGPT 4.0 was optimal for consistency and diagnostic accuracy when compared with Bard and ChatGPT 3.5. © RSNA, 2024 Supplemental material is available for this article.

Prediction model based on MRI morphological features for distinguishing benign and malignant thyroid nodules

BackgroundThe low specificity of Thyroid Imaging Reporting and Data System (TI-RADS) for preoperative benign-malignant diagnosis leads to a large number of unnecessary biopsies. This study developed and validated a predictive model based on MRI morphological features to improve the specificity.MethodsA retrospective analysis was conducted on 825 thyroid nodules pathologically confirmed postoperatively. Univariate and multivariate logistic regression were used to obtain β coefficients, construct predictive models and nomogram incorporating MRI morphological features in the training cohort, and validated in the validation cohort. The discrimination, calibration, and decision curve analysis of the nomogram were performed. The diagnosis efficacy, area under the curve (AUC) and net reclassification index (NRI) were calculated and compared with TI-RADS.Results572 thyroid nodules were included (training cohort: n = 397, validation cohort: n = 175). Age, low signal intensity on T2WI, restricted diffusion, reversed halo sign in delay phase, cystic degeneration and wash-out pattern were independent predictors of malignancy. The nomogram demonstrated good discrimination and calibration both in the training cohort (AUC = 0.972) and the validation cohort (AUC = 0.968). The accuracy, sensitivity, specificity, PPV, NPV and AUC of MRI-based prediction were 94.4%, 96.0%, 93.4%, 89.9%, 96.5% and 0.947, respectively. The MRI-based prediction model exhibited enhanced accuracy (NRI>0) in comparison to TI-RADSs.ConclusionsThe prediction model for diagnosis of benign and malignant thyroid nodules demonstrated a more notable diagnostic efficacy than TI-RADS. Compared with the TI-RADSs, predictive model had better specificity along with a high sensitivity and can reduce overdiagnosis and unnecessary biopsies.

Integrated use of contrast-enhanced and grey-scale ultrasound in assigning American College of Radiology Thyroid Imaging-Reporting and Data System scores for characterisation of thyroid nodules: A prospective observational study.

The advent and increased use of high-resolution ultrasonography has resulted in improved detection of thyroid nodules. Even with the use of various Thyroid Imaging-Reporting and Data System, accurate imaging diagnosis of malignant thyroid nodules has been suboptimal, which necessitated use of newer modalities like contrast-enhanced ultrasonography alone and in combination for this purpose. Although the combined use of various Thyroid Imaging-Reporting and Data System and contrast-enhanced ultrasonography has turned out to be accurate in many studies, the ideal way to integrate contrast-enhanced ultrasonography into the Thyroid Imaging-Reporting and Data System algorithm is under-investigated. To estimate and compare the diagnostic accuracy of American College of Radiology Thyroid Imaging-Reporting and Data System and contrast-enhanced ultrasonography in differentiating benign and malignant nodules alone and in combination. To estimate the diagnostic accuracy of contrast-enhanced ultrasonography in re-categorisation of Thyroid Imaging-Reporting and Data System 3 and Thyroid Imaging-Reporting and Data System 4 thyroid nodules. This was a prospective cohort study performed in a tertiary care university-based hospital for 3 years. Adult patients with clinical or previous sonographic diagnosis of thyroid nodules were selected. Each of the nodules were assessed using ultrasonography and categorised using American College of Radiology Thyroid Imaging-Reporting and Data System criteria. The lesion was then assessed for contrast-enhanced ultrasonography features. The final diagnosis of the nodules was made using fine needle aspiration cytology. The diagnostic accuracy in diagnosis of malignant thyroid nodules for each of the American College of Radiology Thyroid Imaging-Reporting and Data System and contrast-enhanced ultrasonography alone and in combination was assessed. The diagnostic accuracy of contrast-enhanced ultrasonography in diagnosis of malignant thyroid nodules categorised as Thyroid Imaging-Reporting and Data System 3 and Thyroid Imaging-Reporting and Data System 4 was also assessed. American College of Radiology Thyroid Imaging-Reporting and Data System had a sensitivity, specificity, negative predictive value, positive predictive value and diagnostic accuracy of 86.6%, 54.5%, 17.4%, 97.3% and 57.7%, respectively, in diagnosis of malignant thyroid nodules. Contrast-enhanced ultrasonography had a sensitivity, specificity, negative predictive value, positive predictive value and diagnostic accuracy of 86.6%, 95.4%, 67.9%, 98.4% and 94.4%, respectively, in diagnosis of malignant thyroid nodules. Contrast-enhanced ultrasonography had sensitivity, specificity, negative predictive value, positive predictive value and diagnostic accuracy of 93.3%, 100.0%, 100.0%, 99.2% and 99.3%, respectively, in re-categorisation of Thyroid Imaging-Reporting and Data System 3 and Thyroid Imaging-Reporting and Data System 4 nodules. Contrast-enhanced ultrasonography can play a key role in diagnosis of malignant thyroid nodules which are categorised as indeterminate on grey-scale ultrasound.

Thyroid nodule diagnosis: A comparative analysis of ultrasonography and fine-needle aspiration cytology

Objectives: This paper aimed to investigate the association between Ultrasonography (USG) results and fine-needle aspiration cytology (FNAC) in the diagnosis of thyroid nodules (TNs) in private and public hospitals in Yemen. In addition, it aimed to determine the sensitivity, specificity of USG and FNAC in diagnosing thyroid nodules (TNs). Materials and Methods: The study utilized a retrospective comparative study design, analyzing medical records of patients diagnosed with TNs between June 01, 2022 and May 01, 2023, in four private and public hospitals in Yemen. USG and FNAC reports, along with demographic data were collected. The author categorized the USG findings based on the American Thyroid Association guidelines, while FNAC results were classified according to the Bethesda reporting system for thyroid cytopathology. Results: The study included 94 diverse participants with TNs. The average age was 39.96 years, with a majority of female participants (85.1%). The distribution of nodules revealed varying consistencies, sizes, and shapes. The association between USG and FNAC diagnoses showed a significant relationship (χ2 = 34.914, P &lt; 0.001), with a high proportion of benign cases in both USG and FNAC results. Validity analysis demonstrated a sensitivity of 88.68% and a specificity of 70.73% for USG in diagnosing thyroid carcinoma. The positive predictive value was 79.66%, and the negative predictive value was 82.86%. The overall accuracy of USG compared to FNAC was 80.85%. Conclusion: This study shows USG’s potential as a diagnostic tool for TNs in Yemen. Insights improve understanding of TN diagnosis in Yemen’s healthcare settings.

Clinicopathological Profile of Patient with Solitary Thyroid Nodule Presented to OPD at a Tertiary Care Hospital Pakistan

Objective: To determine the clinicopathological profile of patient with solitary thyroid nodule presented to opd at a tertiary care hospital Pakistan Study Design: Cross sectional study Settings: Department of Diabetes and endocrinology, Hayatabad Medical Complex, Peshawar Materials &amp; Methods: Patients with confirmed diagnosis of a solitary thyroid nodule through imaging or clinical examination were included. Patients with a history of thyroidectomy or thyroid cancer, those with multiple thyroid nodules were excluded. Results: The mean age of the patients in the study was 48.7 years (SD = 2.11). In terms of gender, 42.4% were female, while 57.6% were male. Swelling with dysphagia was reported by 25.6% of patients, swelling with pain by 20.8%, and swelling with fever by 1.6%. The distribution of thyroid nodule location revealed that 50.4% were in the left lobe, 48% in the right lobe, and a small percentage (1.6%) were bilateral. The ultrasonography findings of the patients revealed diverse characteristics of thyroid nodules. The most common were hypoechoic nodules, present in 38.4% of cases. Isoechoic nodules were observed in 26.4%, while mixed echogenicity was noted in 17.6% of patients. A smaller proportion, 9.6%, exhibited thyroid nodules with cystic components. A significant proportion, 57.6%, yielded benign results, while 22.4% were categorized as suspicious. Practical Implication: Understanding the clinicopathological profile of patients with solitary thyroid nodules presenting at a tertiary care hospital in Pakistan can aid in refining diagnostic and treatment strategies, ultimately improving patient outcomes and reducing unnecessary interventions. Conclusion: In conclusion, our study on the clinicopathological profile of patients with solitary thyroid nodules at a tertiary care hospital reveals a diverse presentation of clinical and ultrasonographic features. Keywords: Solitary thyroid nodule, Clinicopathological profile, Ultrasonography, Fine-Needle Aspiration, Thyroid Disorders

Comparative Cyto-Histological Genetic Profile in a Series of Differentiated Thyroid Carcinomas

Introduction: Molecular tests can contribute to improve the preoperative diagnosis of thyroid nodules. Tests available are expensive and not adapted to different populations. Aim: This study aimed to compare the cyto-histological genetic profile and to evaluate the reliability of molecular tests using ultrasound-guided fine needle aspiration cytology (US-FNAC) in accurately diagnosing differentiated thyroid carcinomas (DTCs) and predicting biologic behavior of papillary thyroid carcinomas (PTCs). Materials and Methods: The series included 259 patients with paired cyto-histological samples totaling 518 samples. The genetic alterations were analyzed via PCR/Sanger sequencing. The association with clinicopathologic features was evaluated in PTCs. Results/Discussion: From the 259 patients included, histologies were 50 (19.3%) benign controls and 209 (80.7%) DTC cases, from which 182 were PTCs; cytologies were 5.8% non-diagnostic, 18.2% benign, 39% indeterminate, and 37.1% malignant. In histology, indeterminate nodules (n = 101) were 22.8% benign and 77.2% malignant. Mutation frequencies in cytology and histology specimens were, respectively, TERTp: 3.7% vs. 7.9%; BRAF: 19.5% vs. 25.1%; and RAS: 11% vs. 17.5%. The overall cyto-histological agreement of the genetic mutations was 94.9%, with Cohen’s k = 0.67, and in indeterminate nodules agreement was 95.7%, k = 0.64. The identified mutations exhibited a discriminative ability in diagnosing DTC with a specificity of 100% for TERTp and BRAF, and of 94% for RAS, albeit with low sensitivity. TERTp and BRAF mutations were associated with aggressive clinicopathological features and tumor progression in PTCs (p &lt; 0.001). The obtained good cyto-histological agreement suggests that molecular analysis via US-FNAC may anticipate the genetic profile and the behavior of thyroid tumors, confirming malignancy and contributing to referring patients to surgery.

Clinical value of multi-gene testing in distinguishing benign and malignant thyroid nodules.

The newly released 2022 WHO Classification of Neuroendocrine Neoplasms (version 5) and a recent update on thyroid tumor classifications have emphasized genetic testing to an unprecedented level. Fine needle aspiration (FNA) has been widely applied for the preoperative diagnosis of thyroid nodules. However, it is limited mainly to testing for a single gene-BRAFV600E, whereas multi-gene testing data are scarce, especially in the Asian population. This study aimed to explore the clinical value of multi-gene testing in the differential diagnosis of benign and malignant thyroid nodules based on the 2023 Bethesda System for Reporting Thyroid Cytopathology (BSRTC). A total of 615 thyroid nodules underwent ultrasound-guided fine-needle aspiration cytology (FNAC) were collected from Sir Run Run Shaw Hospital, Zhejiang University School of Medicine. The next-generation sequencing platform was applied for multi-gene testing. A panel of well-recognized commonly mutated genes in thyroid cancer were analyzed, including BRAFV600E, KRAS, NRAS, HRAS, TERT, TP53, PAX8/PPARG, CCDC6/ RET and NCOA4/ RET. Gene mutations were identified in 324 nodules (52.7%), with BRAFV600E being the most prevalent driver gene alteration observed in this cohort (233/324; 79.1%), followed by RAS (77/324, 23.8%). The overall malignancy rate of gene mutations was 89.7% in our cohort, of which the lymph node metastasis rate was 45.3%. The combination of multi-gene testing and cytology resulted in 89.3% sensitivity, 95.2% specificity, 98.9% positive predictive value, 64.5% negative predictive value and 90.3% accuracy, which were significantly higher than those from mere cytology (sensitivity 68.6%, specificity 87.5%, positive predictive value 95.9%, negative predictive value 39.8%, accuracy 72.2%). Multi-gene testing could substantially enhance the detection rate of malignant thyroid nodules and protect patients with benign nodules from unnecessary surgeries. Multi-gene testing provides a valuable reference for individualized preoperative decision-making, which may serve as a crucial method for postoperative treatment and prognosis assessment.

TIRADS and BETHESDA correlation to histopathological results of thyroid nodules: a case series at Dr. Sardjito General Hospital, Yogyakarta

Early diagnosis of thyroid nodules is necessary to identify malignancy before surgery. Fine needle aspiration (FNA) and ultrasonography (USG) are supporting examinations for diagnosing thyroid tumors. There have been several previous studies related to the sensitivity of each investigation compared to postoperative biopsy findings. Both have different diagnostic values, and it is hoped that if these two diagnostic methods are applied before surgery, they will have a higher sensitivity value in determining the type of thyroid nodule compared to a single examination. A single-center retrospective case series study. Socio-demographic and clinical data were obtained from hospital records. This study included six thyroid nodule cases, consisting of three (50%) women and three (50%) men. Patients had the appearance and clinical symptoms of swelling in the neck (100%), of which 50% had symptoms of difficulty swallowing and 16% had symptoms of shortness of breath. Three (50%) patients underwent isthmolobectomy, and 3 (50%) patients underwent total thyroidectomy. Three (50%) patients suspected of approaching malignancy based on TIRADS and BETHESDA underwent total thyroidectomy with postoperative histopathological results of non-malignancy. Three (50%) patients with suspected benign nodules based on TIRADS and BETHESDA underwent isthmolobectomy, with postoperative histopathological results indicating non-malignancy. The concordance of the results of the ultrasound examination (TIRADS) and histopathology of thyroid nodules before surgery (BETHESDA) compared with the results of the postoperative histopathological examination was 50%.

Radiomics in the Diagnosis of Thyroid Nodules

The thyroid nodules (TNs) are widespread throughout the world: according to the pathological studies, they can be found in 50–60% of adults. Currently, ultrasound, computed tomography, magnetic resonance imaging and radionuclide diagnostics, such as positron emission tomography with computed tomography, are usually used to diagnose TNs in clinic. These techniques are mainly used to diagnose the nodile benignity and malignancy, the degree of invasion into adjacent tissues and metastases to lymph nodes. Thanks to the development of artificial intelligence, machine learning and the improvement of medical imaging equipment, radiomics has become a popular area of research in recent years. It allowes to obtain various quantitative characteristics from medical images, highlighting invisible features and significantly expanding the possibilities of identifying and predicting. Radiomics has a high potential in detecting and predicting TNs. We present the information on the development and workflow of radiomics. The article summarizes the application of various imaging techniques to identify benign and malignant TNs, determine invasiveness and metastases to lymph nodes, as well as some new advances in the field of molecular level and deep learning. The disadvantages of radiomics method are also given as well as prospects for its further development.

Unusual Intra-Thyroid Migration of Ingested Fish Bone: A Case Report and Literature Review.

Trans-esophageal migration of foreign bodies is a rare but serious complication of vulnerable foreign bodies. This clinical case is about a 37-year-old female patient, with a history of foreign body ingestion eight (08) months ago, received for anterior cervical swelling evolving since about seven (07) months ago. The diagnosis of thyroid nodule was evoked on clinical examination and explored by neck ultrasound which confirmed the thyroid nodule and highlighted the presence of an intra-thyroid fish bone. The extraction was carried out by cervicotomy under general anesthesia after fine needle markup and the evolution under medical treatment was favorable after 03 weeks. Extra-digestive migrations of ingested foreign bodies constitute often a source of diagnostic wandering and therapeutic difficulties due to their polymorphous clinical manifestations. This was the case in our context where it was confused with an authentic thyroid pathology. Its take of care often imposed neck surgery which could be avoided by earlier endoscopy. Vulnerable foreign bodies of the esophagus must imperatively be extracted for avoiding trans-esophageal migrating.

Diagnostic implication of thyroid spherules for cytological diagnosis of thyroid nodules.

Thyroid fine-needle aspiration (FNA) is a principal diagnostic procedure for thyroid nodules. A specific cytomorphological structure, known as the thyroid spherule, is often seen in FNA specimens. The clinical significance of these spherules in terms of diagnosis and prevalence remains largely unexplored. We performed a retrospective study on 310 thyroid FNA specimens and categorized them according to the Bethesda System for Reporting Thyroid Cytopathology. The presence, size and number of thyroid spherules in each specimen were examined and these data were subsequently correlated with the clinicopathological features. Thyroid spherules were almost exclusively detected in benign cases, comprising 7.6% of all benign diagnoses. The average diameter of spherules in benign cases was 84.9 μm. Benign cases and cases with atypia of undetermined significance cases primarily exhibited low cellularity, while follicular neoplasms and malignant cases typically showed moderate to high cellularity. In the subgroup of FNA cases with moderate to high cellularity, spherules were identified in 12 (20%) of 59 benign FNA cases. Within this group, the sensitivity and specificity of thyroid spherules for detecting benign FNA cases were 20% and 100%, respectively. Our results suggest that the presence of thyroid spherules in FNA specimens can serve as a highly specific marker for benign thyroid conditions. The prevalence of spherule detection is strongly influenced by the cellularity. In cases with moderate to high cellularity, the identification of spherules can assist the cytopathologists in diagnosing thyroid FNA cases as benign.

Diagnostic performance of ACR-TIRADS combined with superb microvascular imaging for differential diagnosis of mummified thyroid nodules and papillary thyroid carcinomas.

The aim was to investigate the ability of superb microvascular imaging (SMI) to improve the differential diagnosis of mummified thyroid nodules (MTNs) and papillary thyroid carcinomas (PTCs) using the 2017 American College of Radiology Thyroid Imaging Reporting and Data System (ACR-TIRADS). We enrolled 110 cases of MTNs and 110 cases of PTCs confirmed by fine needle aspiration (FNA) or surgery. Conventional ultrasound (US) and the quantity of microvessels detected by SMI were analyzed for all nodules. Thyroid nodules were initially categorized by ACR-TIRADS based on US imaging features and then reclassified based on ACR-TIRADS combined with SMI blood-flow grade (SMI-TIRADS). We compared the diagnostic performances of ACR-TIRADS and SMI-TIRADS by receiver operating characteristic curve, sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV). US-detected margin, shape, and echogenic foci differed between MTNs and PTCs (P < 0.05). The SMI blood-flow grade was significantly greater in PTCs compared with MTNs (Χ2 = 158.78, P < 0.05). There was no significant difference in ACR-TIRADS indicators between MTNs and PTCs (Χ2 = 1.585, P = 0.453); however, reclassification by SMI-TIRADS showed significant differences between the groups (Χ2 = 129.521, P < 0.001). The area under the curve was significantly lower for ACR-TIRADS compared with SMI-TIRADS (0.517 vs 0.887, P < 0.05). SMI-TIRADS had significantly higher diagnostic value for distinguishing MTNs and PTCs than ACR-TIRADS (sensitivity: 91.82% vs 74.55%, P < 0.05; specificity: 84.55% vs 21.82%, P < 0.05; accuracy: 88.18% vs 48.18%, P < 0.05; PPV: 85.59% vs 48.81%, P < 0.05; and NPV: 91.18% vs 46.15%, P < 0.05). The detection of microvascular flow and large vessels in thyroid nodules by SMI resulted in high diagnostic specificity and sensitivity. ACR-TIRADS combined with SMI could effectively distinguish between MTNs and PTCs, to avoid unnecessary FNA or surgical excision.

Precision medicine in endocrinology exemplified by medullary thyroid cancer

Medullary thyroid cancer (MTC) is a prime example for precision medicine in endocrinology and underlines the immediate benefits of basic, translational and healthcare research for patients with a rare disease in clinical . A mutation in the rearranged during transfection (RET) proto-oncogene that codes for a transmembrane receptor protein tyrosine kinase, leads to constitutive activation of the kinase, which is the decisive pathomechanism for the disease. The MTC occurs in a sporadic (somatic RET mutation) or hereditary form (RET germline mutation, multiple endocrine neoplasia types 2 and 3). For germline mutation carriers the timing of preventive thyroidectomy depends on the RET genotype. For advanced metastasized RET-mutant MTC, selective RET kinase inhibitors are available, which are currently considered to be game changers in the treatment. Based on the specific tumor marker calcitonin, MTC can be identified at an early stage during the differential diagnosis of thyroid nodules. The preoperative calcitonin level even enables statements on the degree of dissemination of the disease and on the probability of a cure through surgery. A new development is the consideration of desmoplasia as a histopathological biomarker for the metastatic potential of a MTC, which could possibly modify the operative approach as well as the future MTC nomenclature. Furthermore, the postoperative calcitonin level and the calcitonin doubling time are highly valid prognostic markers for tumor burden and biological aggressiveness of MTC and therefore decisive for patient follow-up. Biochemical, molecular and histological markers enable a risk-adapted surgical treatment and together with new targeted systemic treatments have contributed to a paradigm shift in the diagnostics, prognosis and treatment of MTC in recent years. Endocrine precision medicine for MTC therefore enabled a change from the previous purely symptom-oriented to a modern preventive and individualized treatment.

A multitask approach for automated detection and segmentation of thyroid nodules in ultrasound images

An increase in the incidence and diagnosis of thyroid nodules and thyroid cancer underscores the need for a better approach to nodule detection and risk stratification in ultrasound (US) images that can reduce healthcare costs, patient discomfort, and unnecessary invasive procedures. However, variability in ultrasound technique and interpretation makes the diagnostic process partially subjective. Therefore, an automated approach that detects and segments nodules could improve performance on downstream tasks, such as risk stratification. Ultrasound studies were acquired from 280 patients at UCLA Health, totaling 9888 images, and annotated by collaborating radiologists. Current deep learning architectures for segmentation are typically semi-automated because they are evaluated solely on images known to have nodules and do not assess ability to identify suspicious images. However, the proposed multitask approach both detects suspicious images and segments potential nodules; this allows for a clinically translatable model that aptly parallels the workflow for thyroid nodule assessment. The multitask approach is centered on an anomaly detection (AD) module that can be integrated with any UNet architecture variant to improve image-level nodule detection. Of the evaluated multitask models, a UNet with a ImageNet pretrained encoder and AD achieved the highest F1 score of 0.839 and image-wide Dice similarity coefficient of 0.808 on the hold-out test set. Furthermore, models were evaluated on two external validations datasets to demonstrate generalizability and robustness to data variability. Ultimately, the proposed architecture is an automated multitask method that expands on previous methods by successfully both detecting and segmenting nodules in ultrasound.

SRT: Swin-residual transformer for benign and malignant nodules classification in thyroid ultrasound images

With the advancement of deep learning technology, computer-aided diagnosis (CAD) is playing an increasing role in the field of medical diagnosis. In particular, the emergence of Transformer-based models has led to a wider application of computer vision technology in the field of medical image processing. In the diagnosis of thyroid diseases, the diagnosis of benign and malignant thyroid nodules based on the TI-RADS classification is greatly influenced by the subjective judgment of ultrasonographers, and at the same time, it also brings an extremely heavy workload to ultrasonographers. To address this, we propose Swin-Residual Transformer (SRT) in this paper, which incorporates residual blocks and triplet loss into Swin Transformer (SwinT). It improves the sensitivity to global and localized features of thyroid nodules and better distinguishes small feature differences. In our exploratory experiments, SRT model achieves an accuracy of 0.8832 with an AUC of 0.8660, outperforming state-of-the-art convolutional neural network (CNN) and Transformer models. Also, ablation experiments have demonstrated the improved performance in the thyroid nodule classification task after introducing residual blocks and triple loss. These results validate the potential of the proposed SRT model to improve the diagnosis of thyroid nodules' ultrasound images. It also provides a feasible guarantee to avoid excessive puncture sampling of thyroid nodules in future clinical diagnosis.

A study of machine learning models for rapid intraoperative diagnosis of thyroid nodules for clinical practice in China.

In China, rapid intraoperative diagnosis of frozen sections of thyroid nodules is used to guide surgery. However, the lack of subspecialty pathologists and delayed diagnoses are challenges in clinical treatment. This study aimed to develop novel diagnostic approaches to increase diagnostic effectiveness. Artificial intelligence and machine learning techniques were used to automatically diagnose histopathological slides. AI-based models were trained with annotations and selected as efficientnetV2-b0 from multi-set experiments. On 191 test slides, the proposed method predicted benign and malignant categories with a sensitivity of 72.65%, specificity of 100.0%, and AUC of 86.32%. For the subtype diagnosis, the best AUC was 99.46% for medullary thyroid cancer with an average of 237.6 s per slide. Within our testing dataset, the proposed method accurately diagnosed the thyroid nodules during surgery.

Shape-margin knowledge augmented network for thyroid nodule segmentation and diagnosis

Background and ObjectiveThyroid nodule segmentation is a crucial step in the diagnostic procedure of physicians and computer-aided diagnosis systems. However, prevailing studies often treat segmentation and diagnosis as independent tasks, overlooking the intrinsic relationship between these processes. The sequencial steps of these independent tasks in computer-aided diagnosis systems may lead to the accumulation of errors. Therefore, it is worth combining them as a whole by exploring the relationship between thyroid nodule segmentation and diagnosis. According to the diagnostic procedure of thyroid imaging reporting and data system (TI-RADS), the assessment of shape and margin characteristics is the prerequisite for radiologists to discriminate benign and malignant thyroid nodules. Inspired by TI-RADS, this study aims to integrate these tasks into a cohesive process, leveraging the insights from TI-RADS, thereby enhancing the accuracy and interpretability of thyroid nodule analysis. MethodsSpecifically, this paper proposes a shape-margin knowledge augmented network (SkaNet) for simultaneous thyroid nodule segmentation and diagnosis. Due to the visual feature similarities between segmentation and diagnosis, SkaNet shares visual features in the feature extraction stage and then utilizes a dual-branch architecture to perform thyroid nodule segmentation and diagnosis tasks respectively. In the shared feature extraction, the combination of convolutional feature maps and self-attention maps allows to exploitation of both local information and global patterns in thyroid nodule images. To enhance effective discriminative features, an exponential mixture module is introduced, combining convolutional feature maps and self-attention maps through exponential weighting. Then, SkaNet is jointly optimized by a knowledge augmented multi-task loss function with a constraint penalty term. The constraint penalty term embeds shape and margin characteristics through numerical computations, establishing a vital relationship between thyroid nodule diagnosis results and segmentation masks. ResultsWe evaluate the proposed approach on a public thyroid ultrasound dataset (DDTI) and a locally collected thyroid ultrasound dataset. The experimental results reveal the value of our contributions and demonstrate that our approach can yield significant improvements compared with state-of-the-art counterparts. ConclusionsSkaNet highlights the potential of combining thyroid nodule segmentation and diagnosis with knowledge augmented learning into a unified framework, which captures the key shape and margin characteristics for discriminating benign and malignant thyroid nodules. Our findings suggest promising insights for advancing computer-aided diagnosis joint with segmentation.