Fig. 1
From: Assessing the performance of ChatGPT-4 and ChatGPT-4o in lung cancer diagnoses
Workflow and results comparing the accuracy, confidence, time, and cost of ChatGPT-4, ChatGPT-4o and physicians in diagnosing lung cancer. ChatGPT-4o served as the control. Accuracy comparisons used Pearson’s chi-squared test, while confidence, time, and cost were compared with independent t-tests. ***: statistically significant; ns: not significant