Medical Display Apparatus Using Patient-Specific Reference Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional medical information display systems struggle to provide accurate and interpretable assessments of patient exacerbation, as they often rely on reference values that are not suitable for patients with existing conditions and lack precision in comparing time series and reference information.
Innovation Solution
A medical information display apparatus that obtains and displays time series data, identifies patient states, and determines corresponding reference data using relevant data of the same type from another patient, allowing for effective comparison and interpretation of patient conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reference values from general population or healthy people are used to assess patient abnormalities, then assessment can be performed, but the precision and reliability of the assessment deteriorates because the reference values are not suitable for patients with existing conditions
Solution Approach 1:
The patent changes the reference parameter from general population values to patient-specific baseline values derived from their own historical data. The system dynamically adjusts reference thresholds based on individual patient characteristics, disease severity, and temporal patterns, transforming a static one-size-fits-all approach into a customized adaptive reference system that accurately reflects each patient's unique condition trajectory
Solution Approach 2:
The patent creates a digital copy of the patient's own historical data as a reference baseline for comparison. By generating virtual reference values from the patient's accumulated time-series data through machine learning models, the system enables self-comparison where the patient's past state serves as the reference for evaluating current abnormalities, eliminating the need for external population-based references
2Loss of information
If reference information and time series information are displayed in divided regions as mutually-different types of diagnosis/treatment information, then both types of information can be presented, but comparison between the two types becomes difficult
Solution Approach 1:
The patent merges reference information and time series information into a unified visual display framework. Both types of information are presented together in the same display region with corresponding temporal and categorical alignment, allowing medical providers to simultaneously view and compare reference baselines against actual time-series measurements without switching between separate screens or sections
Solution Approach 2:
The patent adds a temporal dimension to the display by showing reference information alongside time-series data across the same time axis. This dimensional integration allows comparison not only across different information types but also across time, enabling providers to see how current measurements deviate from reference values at corresponding temporal points
3Adaptability or versatility
If trained models are used to determine reference values or abnormal values for each patient, then patient-specific assessment can be achieved, but the precision level cannot reach 100% due to model limitations
Solution Approach 1:
The patent implements feedback loops where the system continuously monitors actual patient outcomes and compares them against model-predicted reference values. This feedback is used to iteratively refine and retrain the machine learning models, allowing them to progressively improve their accuracy by learning from real-world performance data and correcting systematic errors in their predictions
Solution Approach 2:
The patent transforms static reference values into dynamic, adaptively updating references. The system continuously updates reference baselines as new patient data becomes available, allowing the reference values to evolve over time alongside the patient's condition. This dynamic adaptation enables the system to capture changes in patient baseline characteristics while maintaining high determination precision
Data Source
AI summary
A medical information display apparatus according to an embodiment includes processing circuitry. The processing circuitry is configured: to obtain time series data related to a predetermined patient; to receive a display time period being a range of data to be displayed from among the time series data; to identify a first state of the patient corresponding to the display time period, on the basis of the time series data; to determine reference data corresponding to a second state different from the first state, on the basis of one selected from between relevant data that is of the same type as the time series data and is relevant to another patient and the time series data; and to cause a display to display time series data corresponding to the display time period and the reference data, so as to be kept in correspondence with each other.