Dear Editor,

We would like to thank the readers for their interest in our article¹. Our responses to their questions regarding our manuscript² are provided below.

Our study is a preliminary study examining the usability of an artificial intelligence (AI) model for the diagnosis and treatment of acute poststreptococcal glomerulonephritis (APSGN) in children. The most recent data of the last 11 patients who were followed up and treated in our department between September 2023 and March 2024 were shared with AI. The results obtained with this limited number of patients are intended to shed light on large-scale studies in the future.

The 12 questions directed at AI regarding APSGN are presented in Table 1. The AI answers were evaluated by two pediatric nephrologists with 5 and 16 years of experience in the field. To make the evaluation transparent, the answers given by AI are shared with the reader in Supplemental File 1. These answers are reproduced in Supplemental File 1. Thus, the answers provided by AI can be easily reviewed by anyone.

In our study, patient data were retrospectively analyzed using the AI model after the treatment processes of our patients were completed. No support was received from the AI in the diagnosis and treatment management of the patients. When the clinical and laboratory data of the patients at the time of initial presentation were summarized to AI, the concordance of the diagnosis and treatment recommendations with the physicians was examined. As an example for readers, the clinical and laboratory data of the first three patients referred to AI, as well as AI’s responses in terms of diagnosis and treatment, are presented in Table 3. Data on the other patients are also shared with the reader as Supplemental File 2. For maximum clarity, these data are presented in Supplemental File 2.

In conclusion, the current study highlights the potential of AI for the diagnosis and treatment of APSGN in children. The findings from our limited patient data may pave the way for more comprehensive future research. In addition, the integration of AI into clinical decision-making processes could significantly improve patient care. Feedback from our readers is invaluable in advancing our work in this field. Thank you for your interest.