Deep learning models for image classification of lymphoma: a pilot study in canine

The aim of this study was to distinguish canine lymphoma from other diseases, particularly reactive lymphoid hyperplasia (RLH), based on fine needle aspiration (FNA) images. We developed four deep learning models based on Vision Transformer (ViT) and Inception-v3, which were pre-trained image classification models. The two models out of four were ViT and Inception-v3, and the remained were the two types of combination, i.e., ensemble learning models, of ViT and Inception-v3; the mean of class probabilities of ViT and Inception-v3 (Ensemble model A; MEAN) and the maximum probabilities of ViT and Inception-v3 (Ensemble model B; MAX). A total of 2,290 FNA images of canine lymphoma and 871 FNA images of RLH were analyzed. The FNA images were obtained from the twenty-five slides of fourteen lymphoma cases and eight slides of seven RLH cases in two hospitals. Three types of training and test datasets were prepared from the above image datasets for fair evaluation of the models. Three deep learning-based image classification models (Inception-v3 and the two ensemble models) attained high performance of >80% accuracy, recall and area under the curve (AUC) values for all three datasets. ViT did not archive high performance, except the precision (>0.85). This study is an example of showing potentials of deep learning models through image classification problem in canine lymphoma.