基于大模型的医疗对话“声-智”一体化应用实践  

Application practice of"Sound-Intelligence"integration in medical dialogue based on large model

作　　者：张旭[1] 牛宇翔 刘卿 王力华 ZHANG Xu;NIU Yuxiang;LIU Qing;WANG Lihua(Information Center,Beijing Friendship Hospital,Capital Medical University,Beijing 100050,China)

机构地区：[1]首都医科大学附属北京友谊医院信息中心,北京100050

出　　处：《中国数字医学》2024年第8期14-17,共4页China Digital Medicine

摘　　要：基于医院一体化大模型应用架构,应用文字和语音两个模态,开展门诊电子病历生成和放射报告生成等两个医疗场景的应用实践,阐述系统建设与应用过程中的难点与解决方案;开展面向临床医生的问卷调查,分析医生对大模型技术用于医疗业务的接受度。随着大模型向多模态、拟人化、高响应等方面持续进化,在医疗领域的应用也将进一步深化,促进医疗质量、诊疗效率的提升。Based on the application architecture of hospital integrated large model,text and voice modalities were applied to carry out the application practice of two medical scenarios,such as the generation of outpatient EMR(electronic medical records)and the generation of radiological reports.The difficulties and solutions in the process of system construction and application were described,and a questionnaire survey for clinicians was carried out to analyze the acceptance of large model technology for medical services.As the large model continues to evolve towards multi-modal,human-like and high-response aspects,the application in the medical field will be further deepened,which will also promote the improvement of medical quality and diagnosis and treatment efficiency.

关 键 词：大模型 医院场景 病历生成 案例实践 

分 类 号：R197.3[医药卫生—卫生事业管理] R319[医药卫生—公共卫生与预防医学]