Patient-characteristic adaptive treatment scheme recommendation system and method thereof

Abstract

The invention brings forward a patient-characteristic adaptive treatment scheme recommendation system based on medical big data analysis. The system includes a data layer, an engine layer and an application layer, integrates a plurality of treatment recommendation models aiming at multi-scale characteristics, and filters in combination with medical guidelines. When a user inputs a certain characteristic value, the system can give an appropriate treatment scheme recommendation; and with the gradual accumulation of characteristics, the accuracy of recommendation is getting higher and higher. Theinvention also provides a patient-characteristic adaptive treatment scheme recommendation method based on the above system.

Description

technical field [0001] The invention relates to a clinical assistant decision-making system and method. The invention provides personalized and precise treatment plan recommendations for clinicians according to patients' conditions. Background technique [0002] With the continuous advancement of medical informatization and the continuous development of artificial intelligence technology, driven by the national hierarchical diagnosis and treatment policy, the clinical assistant decision-making system has entered many community hospitals to help doctors and provide doctors with auxiliary suggestions. The treatment plan recommendation in the clinical assistant decision-making system generally adopts a rule-based reasoning mechanism, but the establishment of the current treatment plan recommendation rules is often based on a single clinical guideline recommendation, and the guidelines are given for common situations in the group. To come up with a treatment plan, in order to ac...

AI Technical Summary

Problems solved by technology

[0003] The collected feature values ​​are generally based on guidelines, and common ones include basic information, inspections, disease history, medication history, etc. These options are aimed at the common situation of the group, rather than personalized customization. There are redundant options for individuals, so It is more complicated to operate and will reduce efficiency

[0004] The treatment recommendation model based on medical guidelines is often less robust, because the model is recommended based on all features, and does not support only some feature input, so when some features are uncertain or unknown, it will seriously affect the accuracy of the recommendation

Images