Standardized Classification of Lung Adenocarcinoma Subtypes and Improvement of Grading Assessment Through Deep Learning.

The histopathologic distinction of lung adenocarcinoma (LADC) subtypes is subject to high interobserver variability, which can compromise the optimal assessment of patient prognosis. Therefore, this study developed convolutional neural networks capable of distinguishing LADC subtypes and predicting disease-specific survival, according to the recently established LADC tumor grades. Consensus LADC histopathologic images were obtained from 17 expert pulmonary pathologists and one pathologist in training. Two deep learning models (AI-1 and AI-2) were trained to predict eight different LADC classes. Furthermore, the trained models were tested on an independent cohort of 133 patients. The models achieved high precision, recall, and F1 scores exceeding 0.90 for most of the LADC classes. Clear stratification of the three LADC grades was reached in predicting the disease-specific survival by the two models, with both Kaplan-Meier curves showing significance (P = 0.0017 and 0.0003). Moreover, both trained models showed high stability in the segmentation of each pair of predicted grades with low variation in the hazard ratio across 200 bootstrapped samples. These findings indicate that the trained convolutional neural networks improve the diagnostic accuracy of the pathologist and refine LADC grade assessment. Thus, the trained models are promising tools that may assist in the routine evaluation of LADC subtypes and grades in clinical practice.

Impact of the COVID-19 pandemic on cytology practice: An international survey in the Asia-Pacific region.

BACKGROUND: The purpose of the current study was to examine the impact of coronavirus disease 2019 (COVID-19) on various aspects of cytology practice in the Asia-Pacific region. METHODS: An online questionnaire was distributed to cytopathology laboratories in 24 Asia-Pacific countries to explore the impact of restrictive measures on access to health care, use of general and personal protective equipment (PPE), and changes in cytology workflow and workload from February to April 2020. RESULTS: A total of 167 cytopathology laboratories from 24 countries responded to the survey; the majority reported that restrictive measures that limited the accessibility of health care services had been implemented in their cities and/or countries (80.8%) and their hospitals (83.8%). The respondents noted that COVID-19 had an impact on the cytologic workflow as well as the workload. Approximately one-half of the participants reported the implementation of new biosafety protocols (54.5%) as well as improvements in laboratory facilities (47.3%). Rearrangement or redeployment of the workforce was reported in 53.3% and 34.1% of laboratories, respectively. The majority of the respondents reported a significant reduction (>10%) in caseload associated with both gynecological (82.0%) and nongynecological specimens (78.4%). Most laboratories reported no significant change in the malignancy rates of both gynecological (67.7%) and nongynecological specimens (58.7%) compared with the same period in 2019. CONCLUSIONS: The results of the survey demonstrated that the COVID-19 pandemic resulted in a significant reduction in the number of cytology specimens examined along with the need to implement new biosafety protocols. These findings underscore the need for the worldwide standardization of biosafety protocols and cytology practice.

Histopathologic Assessment of Capsular Invasion in Follicular Thyroid Neoplasms-an Observer Variation Study.

The assessment of capsular invasion is an essential but challenging step in the diagnosis of encapsulated follicular thyroid neoplasms. Therefore, interobserver agreement in the assessment of capsular invasion in these tumors was investigated among 11 thyroid pathologists by using virtual slides of 20 cases in which the original diagnosis considered the differential diagnosis of definite capsular invasion versus questionable capsular invasion. The assessment of capsular invasion was divided into three categories: (1) non-invasive, (2) questionable invasive, and (3) clear-cut invasive. The interobserver agreements for clear-cut invasive and non-invasive categories were fair (Kappa value = 0.578 and 0.404, respectively), whereas agreement for the questionable invasion was poor (Kappa value = 0.186). Disagreements in the assessment of invasion resulted in variable final pathological diagnoses. For example, the agreement for a diagnosis of malignancy was only fair (Kappa value = 0.545). Moreover, pathologists did not have a uniform approach for rendering a final diagnosis in cases with questionable capsular invasion, though nine of 11 pathologists did use the follicular tumor of uncertain malignant potential diagnosis as proposed by the World Health Organization classification of endocrine organs published in 2017. In conclusion, this study revealed considerable interobserver variation in the evaluation of capsular invasion, especially in follicular neoplasms with questionable invasion.