Visualization system for prediction of gynecological neoplastic disease risks

Abstract

The invention discloses a visualization system for prediction of gynecological neoplastic disease risks. The visualization system comprises a risk prediction model construction module for acquiring gynecological related disease variables from gynecological cases, performing correlation analysis with affected gynecological events and constructing a gynecological risk prediction model based on screened risk factors; a gynecological probability prediction module, which receives an onset risk prediction request, calls a related historical disease data queue and acquires a gynecological onset probability prediction result based on the gynecological risk prediction model; and a health report generation module, which is used for generating a visual report according to a gynecological onset probability prediction result, acquired physiological index information of a user, risk factors related to the user and the contribution rate of the risk factors. According to the system, the probability ofsuffering from gynecological diseases is predicted according to the physiological indexes of a testee, indexes that determine a probability value can be visually known, and meanwhile, a data relationship among the indexes is visually displayed through cooperative visualization of a diagnosis report result.

Description

technical field [0001] The invention belongs to the technical field of medical big data processing, and in particular relates to a visualization system for gynecological tumor disease risk prediction. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] Gynecological tumor diseases include ovarian cancer, cervical cancer, breast cancer, endometrial cancer, etc. Currently, there are several clinical indicators for gynecological risk prediction, such as the prediction of ovarian cancer, cancer antigen CA125 is the most commonly used to monitor ovarian cancer Tumor marker, CA125 exists in epithelial gynecological tissues and serum of patients. It is mainly used for auxiliary diagnosis of malignant serous ovarian cancer and epithelial ovarian cancer. It is also an indicator for observing the curative effect after ovarian cancer surgery and chemoth...

AI Technical Summary

Problems solved by technology

[0006] To sum up, very few gynecological risk prediction models have been established at present, and the indicators used in the establishment of the models are mainly obtained by means of clinical experience and existing public literature. Some models are not universal; at the same time, the above-mentioned diagnosis of gynecology based on markers, or the method of image recognition diagnosis by extracting relevant pathological images are all diagnosis of gynecology. In the process of physical examination, the potential incidence of disease cannot be found only through the above methods, or the abnormality of a certain physical examination index cannot effectively diagnose whether there is a probability of incidence

[0007] Secondly, for the diagnosis results, most of them are currently available in the form of paper physical examination reports for users to view, or in the form of spreadsheets to present various indicators of the diagnosis process in tables; data are presented in the form of tables, static data The form reduces the user's physical examination, lacks intuition, and data visualization can realize dynamic display; in addition, it cannot visually display the data relationship between various indicators, and the visualization technology of prediction results has not been promoted

[0008] In addition, in the insurance field, premiums are determined for customers based on age and gender factors, and the pricing is single and not individualized; moreover, in the traditional insurance industry, the risk model for evaluating the health of users is based on the traditional experience of the industry , neither can accurately obtain and evaluate customer health information, nor can it be updated in real time, and false information cannot be ruled out. In the design of customer-oriented insurance products, there are problems such as single pricing and simple underwriting.

Images