Medical text structured recognition processing method based on OCR

Abstract

The invention discloses a medical text structured recognition processing method based on OCR, and the method comprises the steps: collecting a medical text image of a patient, marking textbox position information and text information of the image as a data set, and optimizing the quality of a real scene image through an image correction module; identifying a textbox position and text information in the image through an OCR technology, and performing classification according to text content; a corresponding structured post-processing module is established for specific medical documents such as laboratory test reports, hospital admission checks and image reports so as to extract key information. According to the method, the OCR is utilized, the structured data can be effectively extracted from the medical text image, an important reference basis is provided for subsequent disease diagnosis of doctors, and the efficiency of seeing a doctor for a patient is improved.

Images