Mutation Prioritization for Personalized Therapy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current personalized diagnostics face challenges in organizing unstructured mutation-disease association and cancer-specific targeted therapy information into a structured format for automated analysis, hindering the recommendation of effective therapy options for clinicians.
Innovation Solution
A method and device for mutation prioritization that acquire patient mutation information, map it with a disease knowledgebase, generate a frequency table based on categories and classes, and prioritize mutations using a prioritization scheme, assisting in deciding treatment options.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If unstructured mutation-disease association and therapy information is used directly, then data completeness is maintained, but automated analysis capability deteriorates
Solution Approach 1:
The patent segments unstructured medical data into structured categories including mutation information, disease information, therapy information, and evidence information. Each category contains specific data elements that can be independently processed and analyzed, enabling automated analysis while preserving complete data
Solution Approach 2:
The patent introduces a knowledgebase as an intermediary layer between raw mutation data and clinical analysis. The knowledgebase stores pre-processed relationships between mutations, diseases, and therapies, enabling automated querying and analysis without losing underlying data completeness
2Quantity of substance
If all mutation data is analyzed without prioritization, then comprehensive coverage is achieved, but clinical decision-making efficiency deteriorates
Solution Approach 1:
The patent performs preliminary prioritization of mutations based on multiple criteria (frequency, clinical significance, evidence strength) before clinical analysis. This pre-ranking allows clinicians to focus on the most relevant mutations first, maintaining comprehensive coverage while significantly improving decision-making efficiency
Solution Approach 2:
The patent applies different quality criteria to different mutations based on their specific characteristics. Each mutation is evaluated and prioritized according to its own relevance, frequency, and clinical significance rather than treating all mutations uniformly, enabling efficient clinical focus on high-priority variants
3Reliability
If multiple knowledge sources are integrated, then evidence comprehensiveness is improved, but data processing complexity increases
Solution Approach 1:
The patent creates a universal data structure and knowledgebase framework that can accommodate multiple knowledge sources (literature, clinical trials, databases) through standardized schemas. This multi-functional framework handles diverse data types uniformly, improving evidence comprehensiveness while managing processing complexity through standardization
Data Source
AI summary
Provided are methods and devices for mutation prioritization, which are helpful in personalized therapy of a patient. Also, provided are methods and devices for generating a disease knowledgebase. Information present in various categories of knowledge sources with respect to a particular association of <Disease, Gene, Mutation> set may be identified. The identified information is ranked with respect to the disease knowledgebase to find out the most relevant ones for the treatment of a particular Disease/Gene/Mutation of a patient, thereby enabling medical experts to personalize a therapy to be administered to a patient.