Skip to main content
. 2021 Jun 10;12:3541. doi: 10.1038/s41467-021-23913-3

Fig. 5. Development of the efficient AI-based Digital Pathology Image Quality Control (DPIQC) system.

Fig. 5

a Screening satisfactory and unsatisfactory digital smears produced by three different sedimentation methods. b Division of the whole digital smears into patches with a size of 6000 × 6000 pixels. c Feature extraction of the patches. d Trained XGBoost model for predicting whether digital smear was qualified. The representative images in (ac) display the construction process of the DPIQC system.