To demonstrate that the privacy-preserving training of AI models for chest radiograph diagnosis is possible with high accuracy and fairness compared to non-private training, high quality clinical chest radiographs were retrospectively collected and manually labeled.