Generalized Biomarker Modeling for Cohort Selection Across Sparse Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying patients with specific biomarkers in medical records are inefficient due to the need for individualized models for each biomarker, which is not feasible given the variety of biomarkers and limited data availability, and are hindered by the presence of unstructured data and handwritten notes.
Innovation Solution
A generalized biomarker model is developed using machine learning techniques to identify patients based on features extracted from structured and unstructured medical records, allowing identification of patients tested for specific biomarkers regardless of the availability of data for those biomarkers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If disease-specific biomarker panels are used for each disease, then diagnostic accuracy for that specific disease is improved, but the complexity of developing and maintaining multiple separate panels increases
Solution Approach 1:
The patent applies universality by creating a single multi-disease biomarker panel that can diagnose multiple different diseases simultaneously. The panel includes biomarkers for cardiovascular disease, cancer, chronic kidney disease, and other conditions, allowing one panel to perform functions that previously required multiple separate panels. This reduces overall system complexity while maintaining diagnostic accuracy for each specific disease.
Solution Approach 2:
The patent merges previously separate disease-specific biomarker panels into a single integrated multi-disease panel. By combining biomarkers for different diseases into one panel, the invention eliminates the need to develop, validate, and maintain multiple separate panels, thereby reducing complexity while preserving the diagnostic capabilities for each individual disease.
2Adaptability or versatility
If multiple separate disease-specific panels are developed, then comprehensive disease coverage is achieved, but the cost and time required for development and validation increase
Solution Approach 1:
The patent combines multiple disease-specific panels into a single multi-disease panel, thereby achieving comprehensive disease coverage without the need to develop and validate separate panels for each disease. This merging approach significantly reduces the total development time and resources required while maintaining the ability to detect multiple disease types.
Solution Approach 2:
By designing a universal panel that can detect multiple disease types simultaneously, the invention achieves broad disease coverage through a single development effort. This multi-functional approach eliminates the repetitive development and validation processes that would be required for multiple separate panels, thereby reducing time loss.
3Measurement precision
If disease-specific panels are used, then diagnostic precision for individual diseases is improved, but interoperability between different diagnostic systems decreases
Solution Approach 1:
The patent creates a universal multi-disease panel that maintains high diagnostic precision for individual diseases while improving interoperability. By using a common panel structure and biomarker set across multiple disease detections, the system becomes more interoperable and can be implemented across different diagnostic platforms without requiring disease-specific customization.
Solution Approach 2:
The patent segments the diagnostic capability into individual disease-specific biomarker measurements within a unified panel structure. This allows each disease to be detected with high precision using its specific biomarkers, while the overall panel maintains interoperability through its standardized structure that can be implemented across different diagnostic systems.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A model-assisted system for identifying candidates for a cohort based on a biomarker may include at least one processor. The processor may be programmed to access a database from which information associated with a population of individuals can be derived; provide, to a generalized biomarker model, a first biomarker associated with a cohort, the generalized biomarker model being trained based on one or more second biomarkers using the information, wherein the first biomarker is different from the one or more second biomarkers; receive, from the generalized biomarker model, a first output indicating a first group of the population of individuals exceeding a first likelihood threshold of having been tested for the first biomarker; and determine, based on the first output, whether an individual from among the first group of the population of individuals is a candidate for the cohort.