highly significantp < 0.001
In the normal cases, GPT showed no significant change ( p = 0.547), while Claude’s improvement was highly significant, with a 64.08% increase in correct diagnoses ( p < 0.001; Figure 3 ).
In the normal cases, GPT showed no significant change ( p = 0.547), while Claude’s improvement was highly significant, with a 64.08% increase in correct diagnoses ( p < 0.001; Figure 3 ).