Are Questionnaire Forms Necessary in the Age of Artificial Intelligence? A Comparative Study of AI Tools in LUTS Assessment

Background: This study aimed to evaluate the accuracy and clinical applicability of various artificial intelligence (AI) tools (ChatGPT vs. Gemini vs. Grok) in interpreting patient history, calculating the International Prostate Symptom Score (IPSS). Methods: Between November 2024 and January 2025, 32 male patients aged 45–70 years presenting with LUTS to a urology outpatient clinic were enrolled. Patients with a history of cancer, elevated PSA, or previous urological intervention were excluded. Each patient's history was documented using standard IPSS questions in an open-ended format, which was then inputted into four AI tools. Each AI tool calculated the IPSS score, assessed symptom severity, and suggested treatment options. Results: The median age was 62.5 years, with a median PSA of 1.08 ng/dL and the median prostate volume of 38 cc. All AI tools calculated IPSS scores with statistically comparable accuracy to patient-reported scores (p < 0.005). Grok provided specific treatment recommendations for all patients (100%), followed by ChatGPT (78.1%). Grok also demonstrated the highest agreement with physician treatment plans (κ = 0.709, p < 0.001), while ChatGPT and Gemini showed lower compatibility (p > 0.05). Supplementary analysis revealed Grok’s unique approach in calculating PSA density and identifying potential malignancy risk. Conclusions: AI tools can accurately interpret patient history and assess LUTS severity. Grok showed the highest compatibility with urologist recommendations, while ChatGPT offered the most nuanced therapeutic suggestions. These findings suggest that AI tools may soon become valuable clinical decision support systems.