Biological Data Normalization via Linear Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for consolidating biological test results from different laboratories are limited, as they do not account for varying normal ranges and units, making it difficult to compare and analyze results across different laboratories, leading to challenges in monitoring patient health and therapeutic efficacy.
Innovation Solution
A method that normalizes biological analysis data by transforming initial values into a common range using linear correlation coefficients, allowing for a unified graphical representation of results across different laboratories, enabling easy comparison and monitoring of patient health.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If biological test results from different laboratories are consolidated using conventional methods, then data collection is simplified, but the results cannot be accurately compared due to varying normal ranges and units
Solution Approach 1:
The patent transforms biological test results by changing their parameter representation through linear correlation coefficients. Each laboratory's results are converted using a transformation formula that maps their specific normal range and units to a standardized reference frame, enabling accurate comparison while maintaining the original data's informational content
Solution Approach 2:
The patent introduces an intermediary normalization layer between raw laboratory results and final comparison. A reference laboratory's normal range serves as the mediator, and linear correlation coefficients act as the mathematical intermediary that bridges different measurement systems, allowing results from any laboratory to be expressed in terms of the reference framework
2Measurement precision
If a unified reference frame is implemented for all laboratories, then result comparability is improved, but the complexity of data processing increases
Solution Approach 1:
The patent performs preliminary calculations of linear correlation coefficients during the data collection phase rather than during analysis. By pre-computing the transformation parameters when establishing the unified reference frame, the actual comparison process becomes simpler, as it only requires applying the pre-determined coefficients to new results
Solution Approach 2:
The patent transforms complex multi-parameter laboratory data into a simplified standardized format using linear correlation. The transformation changes the parameter representation from laboratory-specific units and ranges to a universal reference scale, reducing processing complexity while preserving comparability
3Stability of the object's composition
If linear correlation coefficients are used for normalization, then data homogeneity is improved, but calculation complexity increases
Solution Approach 1:
The patent applies linear correlation coefficient transformation to change the parameters of biological test results into a homogeneous standardized format. This parameter change achieves data homogeneity by expressing all results in terms of a common reference range, while the linearity of the transformation keeps calculations relatively simple compared to non-linear normalization methods
Data Source
AI summary
A method for tracking a biological parameter of a patient by processing one or more initially-measured values A(n) that correspond to results of one or more biological analyses of the patient's biological parameter from one or more laboratories, may include receiving data from one or more laboratories that includes at least one initial value A(n) and metadata indicating a corresponding normal range. The data is recorded and each of the initial values A(n) are transformed into a normalized value Anorm(n) through the use of a mathematical model and computer processing. A progression over time of the given biological parameter may be generated and displayed using a method of electronic graphical representation, wherein the normalized values Anorm(n) may be presented in combination with at least one element of data that relates to a common normalized normal range.