Predicting a future property using reagents by measuring properties at points in time
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A method for measuring specific substances and extracting objective information from the measurements that is especially applicable in immunoassays but can also be used in other assays which rely on affinity between binding partners. The invention overcomes problems and difficulties caused by disturbing or interfering phenomena by exploiting a mathematical algorithm which predicts the target value from a series of measurements performed during the reaction. The mathematical method of linear prediction is used for this purpose.
Core Innovation
The invention predicts a future property of a system comprising one or more reagents, where the reagents undergo a process. The approach measures a plurality of properties of the system at a plurality of points in time over a period of time to generate measurements for each property. The future property is predicted by generating a combination of the measured properties using predetermined weights for each measured property for each point in time.
Predicting comprises summing the plurality of measured properties multiplied by the predetermined weights at each point in time to generate a respective scalar product at each point in time, followed by summing the scalar products. The predetermined weights are calculated by analyzing a population of the system with one or more reagents undergoing the process, using a known future property at the plurality of points in time and measurements of the properties from the population.
The disclosed implementation is described for reagent-based systems such as immunoassays used for diagnosis of analytes based on affinity between binding partners, including ligand/anti-ligand and antibody/antigen interactions. In the example, C-reactive protein is measured in an immunoassay using absorbance at predetermined wavelengths, including absorbance at 653 nm and absorbance at 942 nm over an earlier prediction interval, to predict a red absorbance value at a later target time.
Claims Coverage
The independent claim is directed to predicting a future property by combining multiple measured properties obtained at multiple time points using predetermined weights, where the weights are calculated from a population using known future property values. The claim includes two inventive features.
Weighted scalar-product prediction from multi-time property measurements
Measuring a plurality of properties of a system at a plurality of points in time to generate measurements for each property over a period of time; and predicting a future property by generating a combination of the plurality of measured properties by combining the properties using predetermined weights for each measured property for each point in time, wherein predicting comprises summing the plurality of measured properties multiplied by the predetermined weights at each point in time to generate a respective scalar product at each point in time, followed by summing the scalar products.
Population-based calculation of predetermined weights from known future property
Calculating the predetermined weights by analyzing a population of the system of the one or more reagents undergoing the process with a known said future property at said plurality of points in time and measurements of said properties from said population.
Overall, the claim coverage centers on predicting a future property by weighted scalar products formed from multiple measured properties at multiple time points, and determining the predetermined weights by population analysis using known future property values.
Stated Advantages
Reducing mean square errors.
Reducing coefficients of variance.
Roughly halving the time versus waiting for later measurement.
Documented Applications
Immunoassays for diagnosis of analytes, including cases based on affinity between binding partners (ligand/anti-ligand and antibody/antigen), using measurements at predetermined wavelengths to predict a later target property.
C-reactive protein (CRP) immunoassay data example using absorbance measurements, including absorbance at 653 nm and absorbance at 942 nm, to predict a red absorbance value at a later time point.
Interested in licensing this patent?