AN AUTOMATED DATABASE-EXPERT SYSTEM FOR ACUTE STROKE

急性中风的自动化数据库专家系统

• 批准号: 3477910
• 负责人: STANLEY TUHRIM
• 金额: $ 11.36万
• 依托单位: MOUNT SINAI SCHOOL OF MEDICINE OF CUNY
• 依托单位国家: 美国
• 财政年份: 1991
• 资助国家: 美国
• 起止时间: 1991-03-01 至 1996-02-29
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/3477910
• 关键词: acute disease /disorder; artificial intelligence; cerebrovascular disorder diagnosis; computer assisted diagnosis; computer assisted medical decision making; computer assisted patient care; computer program /software; computer system design /evaluation; patient care management; physicians; prognosis; sign /symptom; stroke; transient ischemic attack

The purpose of the work proposed in this application is to demonstrate the feasibility of linking a computerized medical record (CMR) with medical problem-solving systems to provide useful advice to clinicians in an acute care setting. The application describe a five year plan to develop and implement a prototype CMR and a set of decision-support systems that aid physicians in the diagnosis and management of cerebrovascular disease. The decision-support systems to be developed in this study include a system that will localize stroke lesions on the basis of the signs and symptoms present on admission, a program that will determine what kind of stroke a person has had, and several programs that will aid in the management of TIA and stroke patients. The information used to create and test these systems will come from a variety of sources, including the NINCDS Stroke Data Bank and the Mount Sinai Stroke Data Bank. Each system will be evaluated extensively, using actual patient data, before being incorporated into the comprehensive system. The information used by these programs in making their determinations will be garnered by physicians during the initial assessment of patients admitted to an acute stroke unit. A computer interface will be developed which will allow the physicians to input this information by completing a structured neurologic assessment form represented onscreen at a terminal. From this information, the computer system will generate a conventional admitting note as well as represent the information entered by the physician in a manner that can be processed by the computer, thus providing an efficient way for the physician to record data and concomitantly input information needed for problem-solving systems to function automatically. The problem-solving systems will then use this information to render their "opinions" without requiring additional interaction with the physician.

本应用程序中提出的工作的目的是演示