Disease automatic coding method and system

Abstract

The invention discloses an automatic disease coding method and system. The method comprises the following steps: input data is received; the input data comprises original diagnosis data and medical history file data, the input original diagnosis data and medical history file data are subjected to preprocessing operation, whether a result can be obtained is determined after searching operation is conducted in standard classification and coding of ICD-10; if no, the original diagnosis data is subjected to word segmenting operation, associating and converting operation and matching tree searchingoperation; an optimal result is screened out from matching tree processing results; if the result is not optimal, the result is split; each single disease name is subjected to the word segmenting operation, the associating and converting operation and the matching tree searching operation; the optimal result is screened out from matching tree processing results or the medical history file data, coding accuracy of the result is assessed, and a coding result and an accuracy assessment result are output. A problem that standard diagnosis coding systems adopted by all medical institutions are notunified can be effectively solved.

Description

technical field [0001] The present invention relates to the field of medical information technology and artificial intelligence technology, and in particular to a disease automatic coding method and system. Background technique [0002] With the continuous deepening and deepening of the country's layout of public medical care and medical reform, the all-round improvement of medical services has put forward urgent requirements for medical informationization. In terms of medical insurance, public health services, and hospital informatization management, medical informatization has developed from the process automation management that completes the entire process of recording to Internet + health care; under the current background of big data and artificial intelligence, medical informatization gradually touches the core of medical care. The core business is to provide auxiliary decision-making and guidance reference for the diagnosis and treatment process. In medical business...

AI Technical Summary

Problems solved by technology

This work is done manually, so it is subject to manpower, the quality of coders, the interpretation of standards and the grasp of the huge content of the coding dictionary, as well as the working years and experience, so the number of manual code matching is limited every day, and there are cases of misjudgment , and the degree of coding thickness and judgment standards also vary from person to person. In the comparison and exchange of data in different hospitals, these problems will be highlighted and magnified. has a difference

[0005] (2) The actual diagnosis of the doctor's personalized and custom input is not fully compatible with the standard diagnosis coding library

[0006] Because the diagnoses contained in the diagnosis coding database cannot completely cover the diagnoses actually used by doctors in their daily work, and for the same diagnosis, due to the particularity of Chinese, the expression of the same disease concept is extremely diverse and there is no uniform standard medical treatment in the industry. The standardization of terminology, the wording and structure of the diagnostic names used by different doctors are also different, varied, and the aliases used are very different; in addition, the update of the standard diagnostic dictionary takes a long time (national standard ICD10, last updated in 2009 8 years have passed since this update in 2017), which cannot cover the latest diseases and their names in time; and for the classification of specific diagnoses, the level of detail of the standard codes cannot meet the actual clinical requirements of doctors; Or when a doctor writes a diagnosis, he will add some additional detailed information when writing the diagnosis, and the standard diagnostic code library can not reflect the distinction and classification of the additional information; moreover, when the doctor writes the diagnosis, he will also use some Only abbreviations or abbreviations familiar to the industry, and these are not included in the standard diagnostic coding library

Fundamentally speaking, doctors record disease information in detail from the perspective of diagnosis and treatment, rather than for classification and coding, so the incompatibility is deeply rooted and inevitable. When filling out the diagnosis, there are many errors in using program selection or handwritten coding

[0007] (3) The version of the standard diagnostic coding library is not uniform

[0010] (5) The defect of the existing technology is that the coding is not precise enough. For diseases with conjunctions in the middle, the existing technology does not refine the treatment, and often does not consider this problem, resulting in insufficient fineness of the disease coding results

Images