Table 1_Accuracy of AI chatbots in answering frequently asked questions on cervical cancer.docx

ObjectiveTo compare the accuracy of Deepseek and ChatGPT in answering frequently asked questions (FAQs) about cervical cancer.
MethodsTo compile a list of FAQs concerning cervical cancer, a comprehensive search was conducted on social media and community platforms. The answer keys for all the selected questions were created on the basis of the guidelines of the National Comprehensive Cancer Network (NCCN), the International Federation of Gynecology and Obstetrics (FIGO), and the World Health Organization (WHO) for cervical cancer. The answers given by Deepseek-R1 and ChatGPT O1 were scored according to the Global Quality Score (GQS).
ResultsA total of 74 FAQs covered a diverse range of topics related to cervical cancer, including diagnosis (n = 16), risk factors and epidemiology (n = 19), treatment (n = 20), and prevention (n = 19). When all the answers provided by DeepSeek to the FAQs about cervical cancer according to the GQS were evaluated, 68 answers were rated as score five, 4 answers were rated as score four, and 2 answers were rated as score three. For ChatGPT’s responses to the same set of FAQs, 67 answers were classified as score five, 6 answers were classified as score four, and 1 answer was classified as score three. There was no statistically significant difference between the two groups (p > 0.05).
ConclusionBoth DeepSeek and ChatGPT demonstrated accurate and satisfactory responses to FAQs about cervical cancer when evaluated according to the GQS. However, in regard to treatment issues, a cautious attitude should be maintained. Compared to ChatGPT, DeepSeek stands out for its free availability, which makes it more accessible in resource-limited scenarios to the public.