Objective: FNA is a simple, safe, cost-effective and accurate diagnostic tool for the initial screening of patients with thyroid nodules. The aims of this study were to determine the diagnostic utility of FNAC performed in our institution, assess the cytomorphologic features that contribute to diagnostic errors and propose improvement measures.Methods: A total of 2781 FNACs were included in the study, and 1122 cases were compared with their histological diagnoses. We retrospectively reexamined our discordant (both false-negative and false-positive) cases and performed a systematic review of previous studies on causes of misdiagnoses.Results: When DC V and DC VI were both considered cytologic-positive, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and diagnostic accuracy were 98.3, 30.9, 94.9, 58.3 and 93.5%, respectively. If DC VI was considered cytologic-positive, the sensitivity, specificity, PPV, NPV and diagnostic accuracy of FNAC were 98.0, 84.0, 99.4, 58.3, and 97.5% respectively. The main cause of false-negative diagnoses was sampling error (13/15, 86.7%), while interpretation error led to the majority of the false-positive diagnoses (38/47, 80.9%). Overlapping cytological features in adenomatous hyperplasia, thyroiditis and cystic lesions were the major factors contributing to interpretation errors, while the size and number of nodules may have led to false-negative diagnoses because of heterogeneity and unsampled areas.Conclusions: The sensitivity and PPV of thyroid FNAC in our institution were higher than those in the published data, while the specificity and NPV were lower. Regarding the FNA category DC V, a frozen section analysis during diagnostic lobectomy is necessary. Multiple passes should be performed in various parts of a large nodule or from different nodules to reduce the risk of false-negative findings. Cytopathologists should strengthen their criteria for the identification of adenomatous hyperplasia, thyroiditis and cystic lesions to avoid false-positive diagnoses. NIFTP has little effect on diagnostic accuracy and the distribution of diagnostic errors.

Cytopathological examination is one of the main examinations for pleural effusion, and especially for many patients with advanced cancer, pleural effusion is the only accessible specimen for establishing a pathological diagnosis. The lack of cytopathologists and the high cost of gene detection present opportunities for the application of deep learning. In this retrospective analysis, data representing 1321 consecutive cases of pleural effusion were collected. We trained and evaluated our deep learning model based on several tasks, including the diagnosis of benign and malignant pleural effusion, the identification of the primary location of common metastatic cancer from pleural effusion, and the prediction of genetic alterations associated with targeted therapy. We achieved good results in identifying benign and malignant pleural effusions (0.932 AUC (area under the ROC curve)) and the primary location of common metastatic cancer (0.910 AUC). In addition, we analyzed ten genes related to targeted therapy in specimens and used them to train the model regarding four alteration statuses, which also yielded reasonable results (0.869 AUC for ALK fusion, 0.804 AUC for KRAS mutation, 0.644 AUC for EGFR mutation and 0.774 AUC for NONE alteration). Our research shows the feasibility and benefits of deep learning to assist in cytopathological diagnosis in clinical settings.

Background: The International System for Reporting Serous Fluid Cytopathology (TIS) was recently proposed. We retrospectively applied TIS recommendations for reporting the cytological diagnosis of serous effusions and reported our experience.Methods: All the serous effusions from January 2018 to September 2021 were retrieved from the database. Recategorization was performed using the TIS classification, the risk of malignancy (ROM) was calculated for each TIS category. In addition, on the basis of the original TIS classification, we further subdivided the TIS category IV (suspicious for malignancy, SFM) into 2 groups (IVa and IVb) according to cytological characteristics (quality and quantity) to explore the necessity of SFM subclassification. The performance evaluation was carried out using different samples (pleural, peritoneal and pericardial effusions) and preparation methods (conventional smears, liquid-based preparations and cell blocks).Results: A total of 3633 cases were studied: 17 (0.5%) were diagnosed as ‘nondiagnostic’ (I, ND), 1100 (30.3%) as ‘negative for malignancy’ (II, NFM), 101 (2.8%) as ‘atypia of undetermined significance’ (III, AUS), 677 (18.6%) as ‘suspicious for malignancy’ (IV, SFM), and 1738 (47.8%) as ‘malignant’ (V, MAL). The ROMs for the categories were 38.5%, 28.6%, 52.1%, 99.4% and 100%, respectively. The ROM for SFM was significantly higher than that for AUS (P < 0.001), while the difference between the ROMs for IVa and IVb was insignificant. The sensitivity, negative predictive value (NPV) and diagnostic accuracy of liquid-based preparations were all superior to those of conventional smears and cell blocks in detecting abnormalities. Using the three preparation methods simultaneously had the highest sensitivity, NPV and diagnostic accuracy.Conclusion: Serous effusion cytology has a high specificity and positive predictive value (PPV), and TIS is a user-friendly reporting system. Liquid-based preparations could improve the sensitivity of diagnosis, and it is best to use three different preparation methods simultaneously for serous effusion cytologic examination.

Controversy exists regarding whether patients with low‐risk papillary thyroid microcarcinoma (PTMC) should undergo surgery or active surveillance; the inaccuracy of the preoperative clinical lymph node status assessment is one of the primary factors contributing to the controversy. It is imperative to accurately predict the lymph node status of PTMC before surgery. We selected 208 preoperative fine‐needle aspiration (FNA) liquid‐based preparations of PTMC as our research objects; all of these instances underwent lymph node dissection and, aside from lymph node status, were consistent with low‐risk PTMC. We separated them into two groups according to whether the postoperative pathology showed central lymph node metastases. The deep learning model was expected to predict, based on the preoperative thyroid FNA liquid‐based preparation, whether PTMC was accompanied by central lymph node metastases. Our deep learning model attained a sensitivity, specificity, positive prediction value (PPV), negative prediction value (NPV), and accuracy of 78.9% (15/19), 73.9% (17/23), 71.4% (15/21), 81.0% (17/21), and 76.2% (32/42), respectively. The area under the receiver operating characteristic curve (value was 0.8503. The predictive performance of the deep learning model was superior to that of the traditional clinical evaluation, and further analysis revealed the cell morphologies that played key roles in model prediction. Our study suggests that the deep learning model based on preoperative thyroid FNA liquid‐based preparation is a reliable strategy for predicting central lymph node metastases in thyroid micropapillary carcinoma, and its performance surpasses that of traditional clinical examination.

ObjectiveTo evaluate the diagnostic utility of The Bethesda System for Reporting Thyroid Cytology (TBSRTC) at Peking University Cancer Hospital, the incidence of noninvasive follicular thyroid neoplasms with papillary-like nuclear features (NIFTP), and the impact of reclassification on cytopathological outcomes. MethodsWe performed a retrospective study of thyroid fine-needle aspiration (FNA) cases between April 2014 and March 2019. The FNA results were classified according to TBSRTC. Post-surgery histological findings were followed up. ResultsA total of 2,781 thyroid FNAs were performed. The incidences of the 6 diagnostic categories (DCs I−VI) were 14.8%, 17.1%, 15.8%, 2.3%, 11.6% and 38.5%, respectively. A total of 1,122 patients (40.3%) had corresponding histological results. NIFTP accounted for 0.4% of papillary thyroid carcinoma (PTC) cases, and there was no significant difference in the risk of malignancy (ROM) for each TBSRTC DC between “NIFTP=carcinoma (Ca)” and “NIFTP≠Ca”. When “NIFTP=Ca”, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of TBSRTC were 98.0%, 84.0%, 99.4%, 58.3%, and 97.5%, respectively. When “NIFTP≠Ca”, the sensitivity, specificity, PPV, NPV and accuracy of the TBSRTC were 98.1%, 81.5%, 99.3%, 61.1%, and 97.5%, respectively. ConclusionsTBSRTC is effective in the preoperative diagnosis of thyroid nodules in Peking University Cancer Hospital. The impact of the reclassification of NIFTP on cytopathological outcomes is limited because of its low incidence, and the revised ROMs are not suitable for Asian patients.