Intelligent diagnosis and evaluation method based on mammary gland structured report

Abstract

The invention discloses an intelligent diagnosis and evaluation method based on a mammary gland structured report, which comprises the following steps of: receiving a mammary gland structured report text on a server side and a user terminal of a radiation information system by using a selective multi-mode long-short-term neural network model; and calculating the similarity between the image data of the medical image and the key information of the input text obtained by word segmentation coding, evaluating the diagnosis quality through threshold comparison, and sending a prompt to a user when a problem exists. According to the intelligent diagnosis and evaluation method based on the mammary gland structured report, image focus position information can be extracted and matched with written content in the structured report, the report and the written information lower than a threshold value are screened, an image diagnosis doctor is reminded of reexamination, therefore, the effects of monitoring and evaluating the diagnosis quality in the report are achieved, and finally the effects of reducing the workload of imaging department doctors, improving the diagnosis quality and reducing the misdiagnosis rate are achieved.

Description

technical field [0001] The invention relates to the field of medical services, in particular to an intelligent diagnosis and evaluation method based on breast structured reports. Background technique [0002] At present, the medical industry is gradually moving in the direction of intelligence and automation. Medical institutions use radiology information systems to realize the business processes of patient appointments, examinations, reports, and film delivery. Among them, in the reporting link, in order to improve the content integrity, terminology accuracy and report practicability of the radiological imaging report, a structured report, a report format with unified terms and forms, was developed through expert consensus. However, when writing a structured report, the content of the report still needs to be input by the doctor based on the imaging findings and findings. Due to the doctor's personal level and experience, it is inevitable that missed or misdiagnosed cases w...

AI Technical Summary

Problems solved by technology

However, when writing a structured report, the content of the report still needs to be input by the doctor based on the imaging findings and findings. Due to the doctor's personal level and experience, it is inevitable that missed or misdiagnosed cases will occur, which may delay the patient's condition.

Images