Medical Data Classification Technology
- 详细技术说明
- None
- *Abstract
-
International Classification of Disease (ICD) is a standardized system of codes that detail symptoms, diagnoses, and medical procedures and are widely used by many countries in the world for epidemiology, health management, and clinical purposes. The current version (ICD-10) has more than 71,000 procedure codes and more than 69,000 diagnosis codes. In the health care industry, ICD codes are determined by physicians or other health professionals and recorded to computer systems by medical coders. Due to the complexity of the codes and the large number of incidents to be coded, it is not easy for even well-trained professionals to assign appropriate codes to many cases with high accuracy. Inaccurate coding not only causes billing errors, but introduces inaccurate data that may be later used for research. Professionals at Nationwide Children’s Hospital and Ohio University have built several overall strategies for generating accurate and streamlined classification algorithms for ICD codes. These strategies were successfully tested against the MIMIC III dataset, a real-world hospital dataset containing tens of thousands of cases with ICD codes and medical reports. Ultimately, this technology has high potential to be translated into useful and practical tools that can benefit health care industries and medical research.
- *Inquiry
- Andrew CorrisThe Research Institute at Nationwide Children’s HospitalOffice of Technology Commercialization700 Children’s DriveColumbus, OH 43205T: (614) 355-1604 F: (614) 722-2716Andrew.Corris@nationwidechildrens.org
- *IP Issue Date
- None
- *IP Type
- Other Patent
- 国家
- Not Available
- 申请号码
- None
- 国家/地区
- 美国
