Fig. 2: Comparisons of reliability. | npj Artificial Intelligence

Fig. 2: Comparisons of reliability.

From: Human-AI teaming in healthcare: 1 + 1 > 2?

Fig. 2

Observed HMT reliability was higher than human reliability (A), but not higher than machine reliability (B) or the reliability of the better of humans and machines (C), and was lower than ideal HMT reliability (D). Condition-level data are used. Weights = the number of clinicians * the number of diagnostic cases.

Back to article page