Skip to main content
. 2022 Jul 31;2022:5212128. doi: 10.1155/2022/5212128

Figure 5.

Figure 5

(a) The sensitivity of the four doctors in the first round was lower than that of AI. In the second and third round, the sensitivity of doctors was significantly improved. Among them, the sensitivity of doctor B exceeded that of AI in the third round. Without the assistance of AI, the sensitivity of doctor C decreased in the fourth round, and the same thing happened for doctor D in the fifth round. (b) In the first round, the specificity of three doctors was higher than that of AI and only doctor B's was lower than that of AI. In the second and third round, the specificity of doctors did not change much. Besides doctor A, the specificity of three doctors was higher than that of AI. In the fourth and fifth round, there was no significant trend change in specificity. The reason was that doctors had a great grasp of the diagnosis for the late glaucoma. Whether AI was involved or not, this phenomenon did not change much.