Systems and methods for determining the probability of a pregnancy at a selected point in time
AssigneeCelmatix, Inc.
InventorsMichael Elashoff, Hrishikesh Karvir, Piraye Yurttas Beim
Life of this patent
Estimating the term…
Abstract
The present invention generally relates to systems and methods for determining the probability of a pregnancy at a selected point in time. Systems and methods of the invention employ an algorithm that has been trained on a reference set of data from a plurality of women for whom at least one of fertility-associated phenotypic traits, fertility-associated medical interventions, or pregnancy outcomes are known, in which the algorithm accounts for any woman who ceases pregnancy attempts prior to reaching a live birth outcome.
Claims31
A system for determining a probability of achieving a pregnancy at a selected point in time, the system comprising:
- a central processing unit (CPU);
- and
- storage coupled to said CPU for storing instructions that when executed by the CPU cause the CPU to:
- accept as input, data representative of a plurality of fertility-associated phenotypic traits of a female subject;
- analyze the input data using a prognosis predictor generated by:
- obtaining training data from a plurality of women across a plurality of reproductive cycles, wherein the training data corresponds to fertility-associated characteristics selected from the group consisting of fertility-associated phenotypic traits, fertility-associated interventions, and pregnancy outcomes, and wherein the plurality of women includes a dropout population, who ceased pregnancy attempts after one or more reproductive cycles and prior to live birth, and a non-dropout population, who continued pregnancy attempts in one or more reproductive cycles until live birth,
- determining one or more correlations between at least one of the characteristics and a known pregnancy outcome at a plurality of time points across the reproductive cycles;
- weighting a correlation of the at least one characteristic associated with the non-dropout population based upon a value of that at least one characteristic within the dropout population;
- and
- training the prognosis predictor with said training data to provide outputs indicative of a probability of achieving pregnancy at selected time points across the plurality of reproductive cycles;
- and
- provide a probability of achieving a pregnancy at a selected point in time as a result of using the prognosis predictor on the input data.
A method for determining a probability of achieving a pregnancy at a selected point in time, the method comprising:
- accepting as input, data representative of a plurality of fertility-associated phenotypic traits of a female subject;
- analyzing the input data using a prognosis predictor correlated with infertility across a plurality of reproductive cycles, wherein the prognosis predictor was generated by:
- obtaining training data from a plurality of women across a plurality of reproductive cycles, wherein the training data corresponds to fertility-associated characteristics selected from the group consisting of fertility-associated phenotypic traits, fertility-associated interventions, and pregnancy outcomes, and wherein the plurality of women includes a dropout population, who ceased pregnancy attempts after one or more reproductive cycles and prior to live birth, and a non-dropout population, who continued pregnancy attempts in one or more reproductive cycles until live birth,
- determining one or more correlations between at least one of the characteristics and a known pregnancy outcome at a plurality of time points across the reproductive cycles;
- weighting a correlation of the at least one characteristic associated with the non-dropout population based upon a value of that at least one characteristic within the dropout population;
- and
- training the prognosis predictor with said training data to provide outputs indicative of a probability of achieving pregnancy at selected time points across the plurality of reproductive cycles;
- and
- providing a probability of achieving a pregnancy at a selected point in time as a result of using the prognosis predictor on the input data.
Dependencies are inferred from the claim text.