Ultrasound-guided Fine-needle Aspiration Research Articles

Consistency analysis of ultrasound-guided fine-needle aspiration and histopathology results in thyroid nodules

Introduction: Thyroid nodules are considered as a frequent clinical problem. The great majority of thyroid nodules are benign; however malignancy probability is predicted to be 5-10%. The inconsistency between ultrasonography and fine-needle aspiration (FNA) findings is one of important problems in management of thyroid nodules. Objectives: This study was designed to investigate the consistency of ultrasonography, FNA and histopathology findings in thyroid nodules. Patients and Methods: In this descriptive study, 93 patients who were candidate for thyroid surgery were included. Data on ultrasound-guided fine-needle aspiration of the thyroid and histopathologic results before and after surgery were considered. The Cramér’s V and Fisher’s exact tests were conducted for this study. Results: In patients whose ultrasound (US) results were low-suspicious, we observed a significant correlation between FNA and pathology results (Cramér’s V = 0.574, P = 0.037). However, relationship between FNA and pathology was not accessible among patients whose ultrasonography results were intermediate suspicious or high-suspicious due to insufficient number of samples. Conclusion: In our study, a significant correlation was observed between FNA and pathology results in low-suspicious nodules.

Abstract No. 13 Real-time cytological evaluation of thyroid fine needle aspiration (FNA): is it worthwhile?

Between 12% and 15% of ultrasound-guided thyroid fine-needle aspirations (FNA) are non-diagnostic. Sub-optimal sample quality and slide preparation are major factors. Intra-procedural, real time evaluation reduces non-diagnostic rates; however, this is an added cost and resource. The purpose of this study was to compare our non-diagnostic rates with or without real time evaluation in order to decide if the difference justified the added cost in time and resources.

Thyroid Cancer Screening Using Tumor-Associated DN T Cells as Immunogenomic Markers.

BackgroundThyroid nodules are an extremely common entity, and surgery is considered the ultimate diagnostic strategy in those with unclear malignant potential. Unfortunately, strategies aiming to predict the risk of malignancy have inadequate specificity. Our group recently found that the microenvironment of thyroid cancer is characterized by an enhanced immune invasion and activated immune response mediated by double-negative T lymphocytes (DN T) (CD3+CD4-CD8-), which are believed to enable or promote tumorigenesis. In the present work, we try to use the DN T cells’ proportion in thyroid fine-needle aspiration (FNA) material as a predictor of the risk of malignancy.MethodsWe recruited 127 patients and obtained ultrasound-guided FNA samples from subjects with cytology-positive or suspicious for malignancy and from those with benign nodular goiter associated with compressive symptoms (such as dysphagia, shortness of breath, or hoarseness), Hashimoto thyroiditis, and Graves’ disease. Out of 127, we investigated 46 FNA samples of patients who underwent total thyroidectomy and for which postoperative histological diagnosis by the academic pathologists was available. We specifically measured the number of cells expressing CD3+CD4-CD8- (DN T) as a function of total CD3+ cells in FNA samples using flow cytometry. We correlated their FNA DN T-cell proportions with the pathological findings.ResultsThe DN T cells were significantly more abundant in lymphocytic infiltrates of thyroid cancer cases compared to benign nodule controls (p < 0.0001). When the DN T-cell population exceeded a threshold of 9.14%, of total CD3+ cells, the negative likelihood ratio of being cancer-free was 0.034 (96.6% sensitivity, 95% CI, 0.915–1.000, p < 0.0001). DN T cells at <9.14% were not found in any subject with benign disease (specificity 100%). The high specificity of the test is promising, since it abolishes a false-positive diagnosis and in turn unnecessary surgical procedures.ConclusionThe present study proposes DN T cells’ proportion as a preoperative diagnostic signature for thyroid cancer that with integration of RNA transcriptomics can provide a simplified technology based on the PCR assay for the ease of operation.

Nonfunctional pancreatic neuroendocrine tumours misdiagnosed as autoimmune pancreatitis: A case report and review of literature

BACKGROUNDNonfunctional pancreatic neuroendocrine tumours are difficult to diagnose in the early stage of disease due to a lack of clinical symptoms, but they can rarely manifest as autoimmune pancreatitis. Autoimmune pancreatitis is an uncommon disease that may cause recurrent acute pancreatitis and is therefore often regarded as a special type of chronic pancreatitis.CASE SUMMARYWe report a case of a 42-year-old female who had nonspecific upper abdominal pain for 4 years and radiological abnormalities of the pancreas that mimicked autoimmune pancreatitis. The symptoms and pancreatic imaging did not improve following 1 year of steroid therapy. Finally, pancreatic biopsy was performed through endoscopic ultrasonography-guided fine-needle aspiration biopsy, and nonfunctional pancreatic neuroendocrine tumours were ultimately diagnosed. Pancreatectomy has resolved her symptoms.CONCLUSIONTherefore, the differentiation of nonfunctional pancreatic neuroendocrine tumours from autoimmune pancreatitis is very important, although it is rare. We propose that endoscopic ultrasonography-guided fine-needle aspiration biopsy should be performed if imaging characteristics are equivocal or the diagnosis is in question.

Fast Classification of Thyroid Nodules with Ultrasound Guided-Fine Needle Biopsy Samples and Machine Learning

A rapid classification method was developed for the malignant and benign thyroid nodules with ultrasound guided-fine needle aspiration biopsy (FNAB) samples. With probe electrospray ionization mass spectrometry, the mass-scan data of FNAB samples were used as datasets for machine learning. The patients were marked as malignant (98 patients), benign (110 patients) or undetermined (42 patients) by experienced doctors in terms of ultrasound, the B-Raf (BRAF) gene, and cytopathology inspections. Pairwise coupling was performed on 163 ions to generate 3630 ion ratios as new features for classifier training. With the new features, the performance of the multilayer perception (MLP) classifier is much better than that with the 163 ions as features directly. After training, the accuracy of the MLP classifier is as high as 92.0%. The accuracy of the single-blind test is 82.4%, which proved the good generalization ability of the MLP classifier. The overall concordance is 73.0% between prediction and six-month follow-up for patients in the undetermined group. Especially, the classifier showed high accuracy for the undetermined patients with suspicious for papillary carcinoma diagnosis (90.9%). In summary, the machine learning method based on FNAB samples has potential for real clinical applications.

Mediastinal Nodal Staging Performance of Combined Endobronchial and Esophageal Endosonography in Lung Cancer Cases: A Systematic Review and Meta-Analysis.

By searching lliteratures till January 5, 2022, we evaluated the role of the mediastinal nodal staging of endobronchial ultrasound-guided fine-needle aspiration (EBUS) and endoscopic ultrasound-guided fine-needle aspiration (EUS) in lung cancer. A total of 20 studies with 2,961 patients were included in this study. The pooled sensitivity, specificity, PLR, and NLR for EBUS were 0.79, 0.97, 27.29, and 0.25, respectively. EUS showed staging performance similar to EBUS. The staging performance was significantly improved when combining EBUS + EUS.

The value of 99mTc-MIBI scan in the detection of malignancy potential of hypermetabolic thyroid incidentalomas of 18F-FDG PET/CT

Aim of the studyIncreasingly use of PET/CT leads to discovery of incidental findings. Hypermetabolic thyroid nodules are one of the unexpected lesions in PET/CT imaging with an increased risk of thyroid cancers. Our study aims to determine the malignant potential of incidentally detected 18F-FDG avid thyroid nodules by using Tc-99m MIBI imaging. Materials and methodsPET/CT scans were performed for nonthyroidal purposes and were evaluated for the presence of hypermetabolic thyroid nodules. Tc-99m MIBI scans and ultrasonography-guided fine needle aspiration biopsies were subsequently performed for all patients. ResultsPrimary thyroid malignancies were identified in 25% of patients with increased focal FDG uptake at definitive diagnosis. Among the patients with FDG avid thyroid nodules, Tc-99m MIBI scan showed true-positive results in all thyroid carcinomas (n:7) with a 36.3% (4/11) false-positivity rate. In three patients with indeterminate cytology results, Tc-99m MIBI scan findings were also negative. The sensitivity, specificity, positive predictive value of Tc-99m MIBI scan in predicting the malignancy of FDG-positive thyroid nodules were 100%, 77%, 63.6%, respectively. ConclusionThe implementation of 99mTc-MIBI scan performed by dual phase and SPECT/CT modality might be a helpful cost-effective approach in addition to FNAB in patients with 18F-FDG-positive thyroid nodules and indeterminate cytology to improve the patients’ prognosis and reduce unnecessary thyroid operations with associated use of FNAB.

The Impact of Rapid On-Site Evaluation on the Quality and Diagnostic Value of Thyroid Nodule Fine-Needle Aspirations.

Background: Ultrasound-guided fine-needle aspiration (FNA) is the preferred method to evaluate the dignity of thyroid nodules. Nevertheless, the often-reported high nondiagnostic rate burdens affected patients and the health care system. Rapid on-site evaluation (ROSE) constitutes an addition to the thyroid FNA procedure, with various studies showing its beneficial effect on the Bethesda I nondiagnostic rate. We aimed to assess whether ROSE may reduce the rate of Bethesda categories III and V. Additionally, we examined the influence of ROSE on specimen quality. Methods: We performed a retrospective cohort study, comparing Bethesda categorization and specimen quality in specimens subject to ROSE compared with those not subject to ROSE. We also evaluated aspects of specimen quality that differed according to the use of ROSE. We subcategorized Bethesda I into insufficient cellularity or artifacts, and Bethesda categories III and V into cellular without artifacts, sparsely cellular, or artifacts. Results: We evaluated 5030 thyroid FNAs. ROSE was performed in 1304 (25.9%) cases, and ROSE was not utilized for 3726 (74.1%) specimens. The rate of Bethesda I nondiagnostic and Bethesda III categories was reduced in specimens subject to ROSE (4.3%, 56/1304) compared with non-ROSE (39.9%, 1487/3726, p < 0.001). The rate of both benign Bethesda II and malignant Bethesda VI diagnoses was 91.6% (1194/1270) in ROSE specimens compared with 56.6% (1999/3530) in non-ROSE (p < 0.001). This was reflected by a significant improvement in diagnostic accuracy with ROSE (areas under the curve [AUC]non-ROSE = 0.811, AUCROSE = 0.895, p = 0.004). The overall rate of specimens flawed by sparse cellularity in Bethesda categories III and V was 0.1% (1/1304) in ROSE specimens compared with 1.2% (45/3726) in non-ROSE (p < 0.001). The overall artifact rate was 0.3% (4/1304) for ROSE specimens and 2.5% (92/3726) for non-ROSE (p < 0.001). Conclusions: ROSE significantly increased diagnostic accuracy by improving FNA specimens quantitatively and qualitatively. We suggest considering ROSE as standard of care for thyroid FNAs.

Ampullary Neuroendocrine Neoplasm: Clinicopathological Characteristics and Novel Endoscopic Entity

Background: Neuroendocrine neoplasms of the ampulla of Vater (ampullary NEN) have features of both gastrointestinal and pancreato-biliary (PB) NEN. However, the limited number of studies examining ampullary NEN makes it difficult to clarify their unique characteristics. This study aimed to elucidate the clinical characteristics of ampullary NEN. Methods: We enrolled 162 patients with PB-NEN diagnosed at Kyushu University Hospital between 2011 and 2020. Clinical features, pathological diagnoses, treatments, and prognoses were retrospectively analyzed. We also compared ampullary NEN with pancreatic NEN (PanNEN). Results: We analyzed 10 ampullary NEN cases and 149 PanNEN cases. The ampullary NEN cases consisted of 4 cases of neuroendocrine tumor Grade 1 (NET G1), 1 NET G2 (Grade 2), and 5 neuroendocrine carcinomas (NECs). The incidences of NEC and cholangitis were significantly higher in ampullary NEN than in PanNEN. All ampullary NETs had a submucosal tumor-like appearance, as identified by endoscopic ultrasound-guided fine needle aspiration. We treated small NET G1 (<10 mm) with endoscopic papillectomy and large NET G1 with pancreaticoduodenectomy. There were no cases of recurrence after resection. All ampullary NECs presented with the characteristic endoscopic finding of a “crater sign” similar to deep-mining ulcers seen in gastric malignant lymphoma. Four cases underwent surgical resection, and 1 case was unresectable. Two patients who underwent multidisciplinary treatment were maintained without recurrence for over 2 years. Conclusions: Endoscopic findings showed identifiable distinctions between ampullary NETs and NECs.

EBUS-TBNA: A 2-Year Experience from a Tertiary Pathology Centre with Cyto-Histological Correlation

Introduction: Endoscopic ultrasound-guided fine-needle aspiration and endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) methods are expanding due to their minimally invasive nature using real-time image guidance, a low incidence of complications, and a diagnostic rate ranging from 89% to 93.6%. The application of EBUS-TBNA was able to downgrade 40% of patients with suspicion of stage III lung cancer on PET scan. Methods: In the present study, we analysed the diagnostic outcomes of EBUS cytology at the Laboratory of Pathology, Fimlab Laboratories, Tampere, Finland, during a 2-year period. In our analysis, 1176 EBUS samples from 486 patients that underwent EBUS-TBNA in Tampere University Hospital (TAYS) from January 2017 to December 2018 were reviewed. The mean age of the patients was 64 ± 14 (SD) years with the age range from 19 to 90 years. Cytospin cytology slides stained by Papanicolaou stain and cell blocks were performed in each sample. Further immunohistochemical stains were ordered by the pathologist on demand. Results: The diagnostic yield was 89.5% and the diagnostic accuracy was 77.9% in the present study. The sensitivity for malignancy in EBUS-TBNA diagnosis was 95.69% and the negative predictive value 96.75%. Cyto-histopathological diagnosis correlation was 54.7% (95% confidence interval [CI]: 40.6%–66.4%, p < 0.001), and the histological accuracy compared to the exact original diagnosis was 43.5%. When considering the histological diagnoses that support the original diagnosis, the correlation was 61.3% (95% CI: 48.4%–71.5%, p < 0.001) and the accuracy was 72.2%. The overall inadequate rate of EBUS-TBNA samples in this study was 10.5%. Conclusions: The EBUS-TBNA diagnostic yield, diagnostic accuracy, and inadequate rate in the present study are in line with previously published studies.

Elaboration and Validation of a Nomogram Based on Axillary Ultrasound and Tumor Clinicopathological Features to Predict Axillary Lymph Node Metastasis in Patients With Breast Cancer.

BackgroundThis study aimed at constructing a nomogram to predict axillary lymph node metastasis (ALNM) based on axillary ultrasound and tumor clinicopathological features.MethodsA retrospective analysis of 281 patients with pathologically confirmed breast cancer was performed between January 2015 and March 2018. All patients were randomly divided into a training cohort (n = 197) and a validation cohort (n = 84). Univariate and multivariable logistic regression analyses were performed to identify the clinically important predictors of ALNM when developin1 g the nomogram. The area under the curve (AUC), calibration plots, and decision curve analysis (DCA) were used to assess the discrimination, calibration, and clinical utility of the nomogram.ResultsIn univariate and multivariate analyses, lymphovascular invasion (LVI), axillary lymph node (ALN) cortex thickness, and an obliterated ALN fatty hilum were identified as independent predictors and integrated to develop a nomogram for predicting ALNM. The nomogram showed favorable sensitivity for ALNM with AUCs of 0.87 (95% confidence interval (CI), 0.81–0.92) and 0.84 (95% CI, 0.73–0.92) in the training and validation cohorts, respectively. The calibration plots of the nomogram showed good agreement between the nomogram prediction and actual ALNM diagnosis (P > 0.05). Decision curve analysis (DCA) revealed the net benefit of the nomogram.ConclusionsThis study developed a nomogram based on three daily available clinical parameters, with good accuracy and clinical utility, which may help the radiologist in decision-making for ultrasound-guided fine needle aspiration cytology/biopsy (US-FNAC/B) according to the nomogram score.

Diagnostic efficacy of physical examination, preoperative ultrasound, and/or computed tomography in detecting lymph node metastasis: A single-center retrospective analysis of patients with squamous cell carcinoma of the head and neck

The aim of this study was to compare the diagnostic efficacy of physical examination, ultrasound, and computed tomography for the detection of lymph node metastasis (LNM). Patients with pathologically confirmed squamous cell carcinoma of the head and neck and who were not planned for surgery underwent preoperative physical examinations, ultrasound (US) scans, and computed tomography (CT) scans to detect LNM. Ultrasound-guided fine-needle aspiration cytopathology was the reference standard for calculating the diagnostic parameters. US examination yielded fewer false-negative LNM (P = 0.0125) and higher sensitivity (P= 0.0313), specificity (P= 0.0078), and Youden index (P< 0.0001) than CT examination. The likelihood of detecting LNM with US, CT, and the combination of US and CT as measured in diagnostic confidence/lymph node was >0.718, >0.768, and >0.552, respectively. The combination of US and CT had the least risk of underdiagnosis and higher measures of diagnostic efficacy and predicting outcomes than all other index tests. US has greater diagnostic efficacy than CT in detecting LNM for patients who have no clinical indication for surgery or decline surgical treatment. The combination of US and CT examinations may be a reliable noninvasive method.

Bleeding of a Gastrointestinal Stromal Tumor of the Stomach Because of Tearing of the Overlying Epithelium by Endoscopic Ultrasonography-Guided Fine Needle Aspiration.

Bleeding of a Gastrointestinal Stromal Tumor of the Stomach Because of Tearing of the Overlying Epithelium by Endoscopic Ultrasonography-Guided Fine Needle Aspiration.

Case of needle tract seeding during preoperative neoadjuvant chemotherapy for resectable pancreatic cancer

Reports of needle tract seeding (NTS) as a complication of endoscopic ultrasound‐guided fine‐needle aspiration (EUS‐FNA) have been increasing. To date, most of the reported cases of NTS have been diagnosed during the postoperative follow‐up period. Herein, we report a case of NTS that occurred during preoperative neoadjuvant chemotherapy after EUS‐FNA for resectable pancreatic cancer. The patient underwent transgastric EUS‐FNA for a pancreatic tail tumor. He was diagnosed as having resectable pancreatic cancer and received preoperative neoadjuvant chemotherapy. After completion of the chemotherapy, computed tomography showed a thick‐walled cyst‐like structure appearing between the pancreas and the gastric wall. Combined resection revealed adenocarcinoma invasion into the cyst‐like structure. Based on the clinical course, and surgical and pathological findings, the condition was diagnosed as NTS. It is thus crucial that after EUS‐FNA, a detailed review of the imaging findings be conducted in the preoperative period. If adhesions between the stomach and the pancreas are observed after transgastric EUS‐FNA, combined resection of the gastric wall should be considered.

Positive predictive values of MRI and ultrasound-guided fine-needle aspiration biopsy for isolated parotid swelling in adults: A STARD comparative analysis

ObjectivesTo evaluate positive predictive values (PPVs) of magnetic resonance imaging (MRI) and ultrasound-guided fine-needle aspiration biopsy (UFNAB) in patients with isolated parotid swelling. Material and methodsObservational study following the STARD guideline, based on a cohort of 212 patients from 18 to 93years of age, with isolated parotid swelling (malignant: 16.9%; benign: 83.1%), consecutively operated on between 2015 and 2020, after work-up including MRI and UFNAB in an otorhinolaryngology department of a university hospital. The main endpoint was PPV for diagnosis of benign tumor, malignant tumor and the most frequent etiology. Secondary endpoints were correlations between PPVs and clinical factors for malignancy, and the impact on PPV of various situations: dynamic analysis on MRI; diagnostic disagreement between MRI and UFNAB; and UFNAB PPV according to MRI diagnosis. ResultsPPVs for MRI and UFNAB were respectively 45.4% and 88.8% for malignant tumor, 89.6% and 46.9% for benign tumor, and 88.1% and 85.2% for pleomorphic adenoma (the most frequent etiology). Tumor fixation and history of head and neck radiation therapy PPVs were the only one higher than the MRI one for malignant tumor. MRI PPV did not differ between groups with or without dynamic analysis. PPV for malignant tumor, benign tumor and pleomorphic adenoma on MRI and UFNAB was respectively 42.8% and 33.3%, 42.8% and 100%, and 36.3% and 50% in case of diagnostic discordance. When MRI suggested malignant tumor, UFNAB PPV was 51.8% for malignant tumor, 67.7% for benign tumor, and 37.5% for pleomorphic adenoma; when MRI suggested benign tumor, it was 32.2% for malignant tumor, 91.5% for benign tumor, and 88.5% for pleomorphic adenoma; and, when MRI suggested pleomorphic adenoma, it was 23.5% for malignant tumor, 93.9% for benign tumor, and 92% for pleomorphic adenoma. ConclusionSystematic association of UFNAB to MRI did not fundamentally improve diagnostic accuracy. UFNAB appeared most valuable in case of history of radiation therapy, in case of tumor fixation, and when MRI diagnosis was uncertain and/or suggested malignant tumor and/or the apparent diffusion coefficient (ADC) was low. The contribution of UFNAB when MRI suggested benign tumor or especially pleomorphic adenoma was more limited.

Development of “Mathematical Technology for Cytopathology,” an Image Analysis Algorithm for Pancreatic Cancer

Pancreatic ductal adenocarcinoma (PDAC) is a leading cause of cancer-related death worldwide. The accuracy of a PDAC diagnosis based on endoscopic ultrasonography-guided fine-needle aspiration cytology can be strengthened by performing a rapid on-site evaluation (ROSE). However, ROSE can only be performed in a limited number of facilities, due to a relative lack of available resources or cytologists with sufficient training. Therefore, we developed the Mathematical Technology for Cytopathology (MTC) algorithm, which does not require teaching data or large-scale computing. We applied the MTC algorithm to support the cytological diagnosis of pancreatic cancer tissues, by converting medical images into structured data, which rendered them suitable for artificial intelligence (AI) analysis. Using this approach, we successfully clarified ambiguous cell boundaries by solving a reaction–diffusion system and quantitating the cell nucleus status. A diffusion coefficient (D) of 150 showed the highest accuracy (i.e., 74%), based on a univariate analysis. A multivariate analysis was performed using 120 combinations of evaluation indices, and the highest accuracies for each D value studied (50, 100, and 150) were all ≥70%. Thus, our findings indicate that MTC can help distinguish between adenocarcinoma and benign pancreatic tissues, and imply its potential for facilitating rapid progress in clinical diagnostic applications.

A deep learning-based segmentation system for rapid onsite cytologic pathology evaluation of pancreatic masses: A retrospective, multicenter, diagnostic study.

SummaryBackgroundWe aimed to develop a deep learning-based segmentation system for rapid on-site cytopathology evaluation (ROSE) to improve the diagnostic efficiency of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) biopsy.MethodsA retrospective, multicenter, diagnostic study was conducted using 5345 cytopathological slide images from 194 patients who underwent EUS-FNA. These patients were from Nanjing Drum Tower Hospital (109 patients), Wuxi People's Hospital (30 patients), Wuxi Second People's Hospital (25 patients), and The Second Affiliated Hospital of Soochow University (30 patients). A deep convolutional neural network (DCNN) system was developed to segment cell clusters and identify cancer cell clusters with cytopathological slide images. Internal testing, external testing, subgroup analysis, and human–machine competition were used to evaluate the performance of the system.FindingsThe DCNN system segmented stained cells from the background in cytopathological slides with an F1-score of 0·929 and 0·899–0·938 in internal and external testing, respectively. For cancer identification, the DCNN system identified images containing cancer clusters with AUCs of 0·958 and 0·948–0·976 in internal and external testing, respectively. The generalizable and robust performance of the DCNN system was validated in sensitivity analysis (AUC > 0·900) and was superior to that of trained endoscopists and comparable to cytopathologists on our testing datasets.InterpretationThe DCNN system is feasible and robust for identifying sample adequacy and pancreatic cancer cell clusters. Prospective studies are warranted to evaluate the clinical significance of the system.FundingJiangsu Natural Science Foundation; Nanjing Medical Science and Technology Development Funding; National Natural Science Foundation of China.

Deep convolutional neural network for classification of thyroid nodules on ultrasound: Comparison of the diagnostic performance with that of radiologists

PurposeThis study aimed to train and validate deep learning (DL) models for differentiating malignant from benign thyroid nodules on US images and compare their performance with that of radiologists. MethodsImages of thyroid nodules in patients who underwent US-guided fine-needle aspiration biopsy at our institution between January 2010 and March 2020 were retrospectively reviewed. Four radiologists independently classified the images. Images of thyroid nodules were trained using three different image classification DL models (VGG16, VGG19, and ResNet). The diagnostic performances of the DL models were calculated for the internal and external datasets and compared with the diagnoses of the four radiologists. Pairwise comparisons of the AUCs between the radiologists and DL models were made using bootstrap-based tests. ResultsIn total, 15,409 images from 7,321 patients (mean age, 60 ± 13 years; malignant nodules, 20.7%) were randomly grouped into training (n = 12,327) and validation (n = 3,082) sets. Independent internal (n = 432; 197 patients) and external (n = 168; 59 patients) test sets were also acquired. The DL models demonstrated a higher diagnostic performance than the radiologists in the internal test set (AUC, 0.83 – 0.86 vs. 0.71 – 0.76, P < 0.05), but not in the external test set. The VGG16 model demonstrated the highest diagnostic performance in internal (AUC, 0.86; sensitivity, 91.8%; specificity, 73.2%) and external (AUC: 0.83; sensitivity: 78.6%; specificity: 76.8%) test sets. However, no statistical differences were found in the AUCs among the DL models. ConclusionsThe DL models demonstrated comparable diagnostic performance to radiologists in distinguishing benign from malignant thyroid nodules on US images and may play a potential role in augmenting radiologists’ diagnosis of thyroid nodules.

Is contrast-enhanced endoscopic ultrasound-guided fine needle biopsy better than conventional fine needle biopsy? A retrospective study in a medical center

BackgroundContrast-enhanced endoscopic ultrasound-guided fine needle aspiration (CE-EUS-FNA) could help clinicians to precisely locate and puncture lesions, but its effect on the diagnostic yield improvement is controversial. We designed this study to observe the additional benefit of using contrast in EUS-guided tissue sampling while performing fine needle biopsy (FNB) instead of FNA, as FNB results in a higher diagnostic accuracy.MethodPatients who underwent EUS-FNB performed by a single medical team from January 2019 to March 2021 were included in this study. We analyzed the cytopathological diagnostic accuracy rate and number of needle passes between groups who underwent FNB with and without contrast.ResultWe divided 133 patients who were diagnosed with a malignancy into two groups according to whether they underwent CE-EUS-FNB (n = 48) or conventional EUS-FNB (n = 85). The CE-EUS-FNB group had an equal diagnostic accuracy rate with fewer needle passes compared with the conventional EUS-FNB group. There was no significant trend change in the success cytopathological diagnostic rate for experienced endoscopists for EUS-FNA.ConclusionCE-EUS-FNB had fewer needle passes but no additional benefit for diagnostic yield improvement. There was no difficult threshold for CE-EUS-FNB for endoscopists who were well trained in conventional FNA.

Pathologist-performed ultrasound-guided fine-needle aspirations of the thyroid: A single institution observational study.

Ultrasound-guided fine-needle aspiration biopsies (USFNAs) are increasingly performed by pathologists. This study was designed to assess the diagnostic yield and characterization of thyroid nodules biopsied at a teaching hospital setting in which both attending physicians and trainees are involved in the performance of USFNAs. A retrospective study of pathologist-performed USFNAs of thyroid cases was performed over a period of 9 years at a tertiary medical center. Data collected included patient characteristics and The Bethesda System diagnostic categories. Over the study period, 1531 USFNAs of thyroid nodules were performed in the pathology-based clinic, with 1209 lesions in females and 322 in males. Ninety-three percent of samples were sufficient for diagnosis (n=1420). The majority of nodules biopsied were benign (65.4%, n=1002). Overall, 3.1% of nodules biopsied were diagnostic of malignancy (n=47). The number of USFNAs over the years showed a rapid increase initially, with a coronavirus disease 2019-related decrease in 2020. The authors report their experience with thyroid USFNA over nearly a decade. The most common diagnosis was benign and the second most common was Bethesda category III. Lesions that were diagnostic of malignancy were relatively uncommon. Over the study period, the results showed that at a large tertiary care center in which USFNAs were performed by trainees as well as attending physicians, the diagnostic yield was good with a majority of thyroid nodules biopsied associated with a definitive diagnosis.