Deep Learning-Based Classification and Targeted Gene Alteration Prediction from Pleural Effusion Cell Block Whole-Slide Images

SIMPLE SUMMARY: For many patients with advanced cancer, pleural effusion is the only accessible specimen for establishing a pathological diagnosis. Some pleural effusion cell blocks have not undergone adequate morphological, immunohistochemical, or genetic analysis due to problems with the specimen itself or cost. Deep learning is a potential way to solve the above problems. In this study, on the basis of scanning whole slide images of pleural effusion cell blocks, we investigated the identification of benign and malignant pleural effusion, the determination of the primary site of pleural effusion common metastatic carcinoma, and the alteration of common targeted genes using a weakly supervised deep learning model. We achieved good results in these tasks. Although deep learning cannot be the gold standard for diagnosis, it can be a useful tool to aid in cytology diagnosis. ABSTRACT: 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.