A 33-year-old man was admitted with a 4-day history of neck swelling and intermittent convulsions for 3 days. Physical examination revealed a soft, tender nodule measuring~4 cm×3 cm in the right lobe of the thyroid gland. Chvostek's and Trousseau's signs were positive. Laboratory tests revealed hypocalcemia, normal serum phosphorus and magnesium levels, elevated parathyroid hormone (PTH) levels, and reduced vitamin D levels. Bone turnover markers indicated active bone formation, and bone density testing revealed severe osteoporosis. Neck ultrasound and CT suggested a thyroid-origin mass; however, [99]Tc[m]-MIBI parathyroid scintigraphy was negative. A review of the previous thyroid ultrasound of the patient revealed a hypoechoic nodule on the dorsal side of the right lobe suspected to represent a parathyroid adenoma, along with elevated blood calcium levels. Combined with the significantly elevated PTH levels detected in the aspirate from the neck mass, these findings led to a final diagnosis of primary hyperparathyroidism, cystic degeneration and hemorrhage of a parathyroid adenoma, hungry bone syndrome, secondary osteoporosis, and vitamin D deficiency. After correcting the hypocalcemia, the patient underwent endoscopic resection of the parathyroid lesion. Postoperatively, serum PTH levels decreased significantly, blood calcium levels normalized, and no further convulsions occurred.

Cutaneous metastases of pulmonary origin are rare but clinically significant, typically indicating advanced disease with a poor prognosis. They may occasionally reveal an occult lung cancer, as illustrated by the case reported here. We report the case of a 78-year-old man, a former heavy smoker, who presented with a rapidly growing, ulcerated, exophytic nodule on the anterior chest wall. Skin biopsy revealed a poorly differentiated adenocarcinoma, with negative thyroid transcription factor-1 (TTF-1) and Napsin A staining, raising diagnostic difficulty regarding the primary origin. Chest CT subsequently identified a 4.1 cm tumor of the right lower lobe, and bronchoscopic biopsy confirmed a bronchial adenocarcinoma. Staging revealed disseminated disease (cT2b cN1 cM1c, Stage IVB) with cerebral, adrenal, and bone metastases. A multidisciplinary tumor board recommended whole-brain radiotherapy (30 Gy in 10 fractions) combined with palliative chemotherapy (cisplatin-gemcitabine), with stable disease observed at six months of follow-up. This case underlines the limitations of TTF-1 and Napsin A negativity in excluding a pulmonary origin, particularly in poorly differentiated tumors, and highlights the value of an integrated clinical, radiological, morphological, and immunohistochemical approach. Management of extensively disseminated disease remains essentially palliative; surgery has no impact on survival and is reserved for palliative or diagnostic indications.

ObjectivesThe American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) provides a standardized framework for thyroid nodule risk stratification based on predefined ultrasound features. This study aimed to evaluate the accuracy and robustness of ChatGPT-5.6 Sol in extracting and scoring ACR TI-RADS features and classifying thyroid nodules from synthetic free-text ultrasound reports.MethodsA primary dataset of 100 synthetic thyroid ultrasound reports was constructed, with each report describing 1 nodule and the 5 ACR TI-RADS feature categories. ChatGPT-5.6 Sol was evaluated using a structured prompting strategy for feature identification, feature scoring, total score calculation, and final Thyroid Imaging Reporting and Data System classification. A less-directive prompt was also tested for total score calculation and final classification. Robustness was further assessed using a 30-case synthetic challenge set balanced across TR1 to TR5. Outputs were compared with reference annotations based on the 2017 ACR TI-RADS criteria. Accuracy was reported with Wilson 95% confidence intervals, and final-category agreement was assessed using Cohen's kappa.ResultsIn the primary dataset, accuracy for both feature identification and scoring was 100.0% for each ACR TI-RADS feature category (100/100; 95% confidence interval: 96.3%-100.0%). Total score calculation and final classification were also correct in all cases (100/100; 95% confidence interval: 96.3%-100.0%), with a Cohen's κ of 1.000. In the challenge set, feature identification, feature scoring, total score calculation, and final classification were correct in all 30 cases (30/30; 95% confidence interval: 88.6%-100.0%), with a Cohen's κ of 1.000. The less-directive prompt produced identical total score and classification accuracy in both datasets, with no discordant cases.ConclusionsChatGPT-5.6 Sol demonstrated high accuracy and consistent ACR TI-RADS-based classification across both synthetic datasets. Large language models may have potential for standardized Thyroid Imaging Reporting and Data System interpretation of free-text ultrasound reports; however, further validation in larger and more diverse settings is needed to establish generalizability.

Thyroid nodules are a prevalent condition in clinical practice, typically identified via ultrasound imaging with visual assessment by physicians. However, the ambiguous features of malignant nodules, their proximity to surrounding tissues, image noise, and unclear boundaries can lead to variability in diagnostic outcomes among different practitioners. In this study, we initially employed nine established models to segment ultrasound images from two publicly available datasets, TN3K and DDTI. Evaluation metrics including IoU, F1-score, accuracy (Acc), specificity (Spec), and precision (Pre) were utilized to assess the segmentation performance of these models. The experimental results indicate that, given the current data scale, the overall segmentation performance of each model is relatively low. Analysis suggests this is primarily due to the limited number of samples in the medical imaging dataset. During training, the model struggles to fully learn the discriminative features of the lesion area. Particularly during the encoder's down-sampling stage, the feature expression ability is constrained, affecting the model's ability to accurately recognize lesion locations. To address this, we enhanced the classic model by incorporating the Residual Network ResNet34, constructing a new thyroid nodule segmentation model. The improved model demonstrated enhanced segmentation performance, with most IoU values on the TN3K dataset exceeding 0.8. However, models trained on the DDTI dataset rarely achieved high performance. Further analysis attributes this performance disparity to differences in the number of training samples. To further enhance model performance, we initialized the improved model with pre-trained ResNet34 weights from the ImageNet dataset, resulting in significant performance improvements. Notably, Res34-PAN performed best across both datasets. On the DDTI dataset, the model achieved an IoU of 0.8778, an F1-score of 0.9349, an accuracy of 0.9872, a specificity of 0.9928, and a precision of 0.9342. To further validate the model design's rationale, we conducted a systematic ablation experiment. We selected commonly used network structures for encoders and used them as the encoders for nine segmentation models, comparing performance with and without the ImageNet pre-trained weights. The results further confirmed the effectiveness of the proposed method in the thyroid nodule segmentation task.

Fine-needle aspiration (FNA) is the gold-standard diagnostic procedure for the initial evaluation of thyroid nodules because it is minimally invasive, cost-effective, and highly accurate for distinguishing benign from malignant lesions. However, conventional smear preparation is frequently affected by operator-dependent variability, uneven cell distribution, blood contamination, and air-drying artifacts, all of which may compromise specimen adequacy and diagnostic performance. Liquid-based cytology (LBC) has emerged as an effective alternative, offering standardized specimen processing, improved cellular preservation, and enhanced slide quality. Among the available LBC platforms, the ThinPrep system is one of the most extensively validated technologies for thyroid cytopathology and is increasingly integrated into routine clinical practice and molecular diagnostic workflows. This protocol provides a comprehensive, standardized workflow for processing thyroid FNA specimens using the ThinPrep liquid-based cytology system. The protocol describes specimen collection, transportation, fixation, processing with the ThinPrep 5000 processor, slide preparation, staining, and quality control procedures to ensure optimal cytomorphological evaluation. Recommendations for acceptable specimen preservation, including transport in PreservCyt solution or alternative fixation methods, are also provided. In addition, the protocol includes procedures for recovering the residual cellular material preserved in the ThinPrep vial for paraffin block processing, which allows for complementary studies, e.g., immunohistochemistry, fluorescence in situ hybridization (FISH), DNA and RNA extraction, and next-generation sequencing (NGS), thus expanding the diagnostic utility of thyroid cytology beyond conventional microscopic examination. When properly executed, this protocol generates high-quality monolayer cytology preparations suitable for standardized interpretation according to The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) while preserving residual material for downstream molecular analyses. Compared with conventional smear cytology, the ThinPrep system improves specimen reproducibility, reduces background artifacts and inadequate samples, facilitates laboratory standardization, and supports precision medicine applications. This protocol provides a practical and reproducible approach for both clinical diagnostic laboratories and translational research. © 2026 The Author(s). Current Protocols published by Wiley Periodicals LLC. Basic Protocol: Simultaneous preparation of ThinPrep cytology slides and cell blocks from thyroid fine-needle aspiration specimens using liquid-based cytology Alternate Protocol: Alternative ThinPrep FNA processing using CytoLyt and PreservCyt.

This is a protocol for a Cochrane review (diagnostic). The objectives are as follows: To assess the diagnostic accuracy of US, CT, and MRI - each considered separately - in identifying malignant thyroid nodules, using histopathological examination as the reference standard. Each modality will be evaluated within its established role in the preoperative diagnostic pathway. This includes estimating the sensitivity and specificity of each imaging modality for differentiating malignant from benign thyroid nodules (the ability to correctly identify cancerous and non-cancerous nodules). A clearer understanding of each modality's accuracy is intended to support patient and clinician decisions - for example, whether further imaging is warranted when ultrasound is inconclusive, and whether surgery can be avoided in nodules reliably classified as benign. Secondary objectives Comparative accuracy: To compare the diagnostic performance of US, CT, and MRI for detecting thyroid malignancy, restricted to direct comparisons within studies that evaluated two or more of these tests in the same participants. Indirect comparisons across separate studies will not be performed for inference, as differing clinical roles, case-mix, and selection make them vulnerable to confounding by indication; any such observation will be reported descriptively and regarded as exploratory. Sources of heterogeneity: To investigate potential sources of variation (heterogeneity) in diagnostic accuracy. Specifically, we will examine whether factors such as nodule size, use of contrast media (for CT or MRI), variations in imaging protocols (e.g. ultrasound risk stratification systems, MRI sequence parameters), study design (prospective vs. retrospective), and patient characteristics (e.g. endemic goitre populations, baseline prevalence of malignancy) significantly affect sensitivity or specificity.