Fig. 6. Application of the model to external datasets. External datasets A and B are IHC samples from different laboratories. The size of the external datasets was rescaled to match the scale of the original dataset. The test was performed using the previously trained model without retraining after preprocessing. The F1 scores for each dataset are presented separately. Scatter plots indicate F1 score on individual images. Bars indicate mean±SEM.
© Exp Neurobiol
