Fig. 3 | Scientific Reports

Fig. 3

From: Evaluation of an acne lesion detection and severity grading model for Chinese population in online and offline healthcare scenarios

Fig. 3

The visualized results for the evaluation on online healthcare scenario. (a) shows some typical examples about identifying correct and incorrect cases by AcneDGNet on AcnePA1. The green, red, blue and yellow bounding boxes in the figure represent comedones, papules, pustules, and nodules/cysts, respectively. GT stands for Ground truth, which is the gold standard for acne grade. AI stands for Artificial Intelligence, which is the results predicted by our model.(b) is the visualized results of AcneDGNet on two follow-up images for a patient from AcnePA2. (c) is the confusion matrix for detecting the changing trends in the grade of acne severity. (d) is the counting error statistics of acne lesions for AcneDGNet regarding changing trends in the grade of acne on AcnePA2.

Back to article page