Methods and systems for automated data-integration

The automation of data integration between healthcare IT applications using machine learning algorithms addresses the inefficiencies of manual processes, enhancing speed, reliability, and data quality.

WO2025073858A1 Publication Date: 2025-04-10KONINKLIJKE PHILIPS NV
3 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2024/077885
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-12-05
Filing Date
2024-10-03
Publication Date
2025-04-10

AI Technical Summary

Technical Problem

The manual process of data integration between healthcare IT applications is inefficient, requiring significant manual effort and being prone to errors due to the diversity of HIT applications and heterogeneity of hospital information systems.

Method used

A method and system that utilize trained machine learning algorithms to automate the data integration process by mapping data between a source and target data schema, including data extraction, normalization, schema identification, rule derivation, transformation, validation, and user feedback integration.

Benefits of technology

The solution significantly reduces the time and effort required for data integration, increases the speed and reliability of installations, and improves data quality by minimizing errors and ensuring accurate data sharing between medical systems.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A method (300) for mapping data between a source data schema and a target data schema, comprising: extracting (320) and normalizing data to generate a normalized data set (462); analyzing (330), using a first trained machine learning algorithm (463), the normalized data set to identify a source data schema; deriving (340), using a second trained machine learning algorithm (464), mapping rules configured to transform data from the identified source data schema to the target schema; transforming (350), using the derived mapping rules, at least some data of the identified source data schema to data of the target schema; analyzing (360), using a schema validation algorithm (466), the transformed data to identify any mapping anomalies; and providing (370), to a user via a user interface, one or more of the identified source data schema, at least some of the data transformed to the target schema, and any identified mapping anomalies.
Need to check novelty before this filing date? Find Prior Art