Deep Convolutional Neural Network Inception-v3 Model for Differential Diagnosing of Lymph Node in Cytological Images: a Pilot Study

Background: In this study, we exploited the Inception-v3 deep convolutional neural network (DCNN) model to differentiate cervical lymphadenopathy using cytological images.

Methods: A dataset of 80 cases was collected through the fine-needle aspiration (FNA) of enlarged cervical lymph nodes, which consisted of 20 cases of reactive lymphoid hyperplasia, 24 cases of non-Hodgkin's lymphoma (NHL), 16 cases of squamous cell carcinoma (SCC), and 20 cases of adenocarcinoma. The images were cropped into fragmented images and divided into a training dataset and a test dataset. Inception-v3 was trained to make differential diagnoses and then tested. The features of misdiagnosed images were further analysed to discover the features that may influence the diagnostic efficiency of such a DCNN.

Results: A total of 742 original images were derived from the cases, from which a total of 7,934 fragmented images were cropped. The classification accuracies for the original images of reactive lymphoid hyperplasia, NHL, SCC and adenocarcinoma were 88.46%, 80.77%, 89.29% and 100%, respectively. The total accuracy on the test dataset was 89.62%. Three fragmented images of reactive lymphoid hyperplasia and three fragmented images of SCC were misclassified as NHL. Three fragmented images of NHL were misclassified as reactive lymphoid hyperplasia, one was misclassified as SCC, and one was misclassified as adenocarcinoma.

Conclusions: In summary, after training with a large dataset, the Inception-v3 DCNN model showed great potential in facilitating the diagnosis of cervical lymphadenopathy using cytological images. Analysis of the misdiagnosed cases revealed that NHL was the most challenging cytology type for DCNN to differentiate.