An AI-Powered Medical Chatbot Integrating Large Language Models for Accurate Clinical Histories and Differential Diagnosis

سال انتشار: 1404
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 60

متن کامل این مقاله منتشر نشده است و فقط به صورت چکیده یا چکیده مبسوط در پایگاه موجود می باشد.
توضیح: معمولا کلیه مقالاتی که کمتر از ۵ صفحه باشند در پایگاه سیویلیکا اصل مقاله (فول تکست) محسوب نمی شوند و فقط کاربران عضو بدون کسر اعتبار می توانند فایل آنها را دریافت نمایند.

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

AIMS02_354

تاریخ نمایه سازی: 29 تیر 1404

چکیده مقاله:

Background and Aims: The integration of artificial intelligence (AI) into medicine has revolutionized healthcare, offering innovative solutions to challenges in diagnosis, treatment, and patient care. Despite the growing adoption of AI, many medical chatbots lack the ability to provide accurate, structured clinical histories and differential diagnoses. TebBot addresses this gap by combining Large Language Models (LLMs) and knowledge graphs to deliver precise, patient-specific diagnostic suggestions. Methods: TebBot was developed using a dataset of ۲۵۰,۰۰۰ cases, covering ۳۲۴ symptoms and ۷۰۰ unique diseases. The chatbot employs a Large Language Model (LLM) trained to extract relevant clinical history from patients through a structured dialogue. The collected data is stored in a knowledge graph database, which enables logistic classification to generate differential diagnoses with probability scores. By dynamically identifying discriminative symptoms, the chatbot refines its diagnostic suggestions, achieving high accuracy. Results: TebBot demonstrates the ability to engage users in structured medical conversations, collect detailed clinical histories, and generate differential diagnoses with high accuracy. By maintaining structured medical records and supporting physician decision-making, TebBot enhances diagnostic precision and facilitates continuity of care. Conclusion: This study demonstrates the potential of combining LLMs, knowledge graphs, and probabilistic classification to create intelligent medical assistants that improve diagnostic accuracy and patient engagement. Future developments will focus on expanding TebBot’s capabilities to include personalized health recommendations, treatment guidance, and chronic disease risk assessment, ultimately supporting preventive healthcare and empowering users with actionable insights for better disease management

کلیدواژه ها:

نویسندگان

Mahdi Kalani

Technical Department, Tabiban Roshanandish Houshmand Company, Iran, Isfahan

Fateme Mahdikhoshouei

Product Design Department, Tabiban Roshanandish Houshmand Company, Iran, Isfahan

Atefeh Sanaeifar

Research and Development Department, Tabiban Roshanandish Houshmand Company, Iran, Isfahan