Back to results
US 2024/0389950 A1Published application
Systems and methods for processing retinal signal data and identifying conditions
AssigneeDiamentis Inc.
InventorClaude HARITON
Life of this patent
Estimating the term…
Abstract
There is disclosed a method and system for predicting a likelihood that a patient is subject to one or more conditions. Retinal signal data corresponding to the patient may be received. Retinal signal features may be extracted from the retinal signal data. The retinal signal features may be applied to a mathematical model. The mathematical model may correspond to a condition. A predicted probability for the condition may be output by the mathematical model. The predicted probability may be displayed on an interface.
Claims20
A method for predicting a probability that a human subject is subject to a condition, the method executable by at least one processor of a computer system, the method comprising:
- receiving retinal signal data corresponding to the human subject, wherein the retinal signal data was collected by an electroretinogram (ERG) system comprising a light stimulator that emits light, at least one electrode for detecting electrical signals, a photodetector for measuring an intensity of the emitted light from the light stimulator, and a spectrometer for measuring the spectrum of the emitted light from the light stimulator, wherein the retinal signal data is associated with a light stimulation of the human subject's retina and includes at least one parameter, recorded by the photodetector or the spectrometer, which is voltage independent and/or time independent, and wherein the at least one parameter comprises the intensity of the emitted light, the spectrum of the emitted light, or an impedance component of a receiving circuit;
- determining, based on the retinal signal data, one or more retinal signal features;
- applying the one or more retinal signal features to a mathematical model, wherein the mathematical model corresponds to the condition, wherein the mathematical model comprises a machine learning algorithm trained using a sample dataset of retinal signal data from subjects with the condition, and wherein the sample dataset comprises at least one parameter which is voltage independent and/or time independent;
- outputting, by the mathematical model, a predicted probability for the condition;
- and
- displaying an interface comprising the predicted probability.
3 more elements
A method for determining a distance between a human subject and a biosignature of a condition, the method executable by at least one processor of a computer system, the method comprising:
- receiving retinal signal data corresponding to the human subject, wherein the retinal signal data was collected by an electroretinogram (ERG) system comprising a light stimulator that emits light, at least one electrode for detecting electrical signals, a photodetector for measuring an intensity of the emitted light from the light stimulator, and a spectrometer for measuring the spectrum of the emitted light from the light stimulator, wherein the retinal signal data is associated with a light stimulation of the human subject's retina and includes at least one parameter, recorded by the photodetector or the spectrometer, which is voltage independent and/or time independent, and wherein the at least one parameter comprises the intensity of the emitted light, the spectrum of the emitted light, or an impedance component of a receiving circuit;
- determining, based on the retinal signal data, one or more retinal signal features;
- determining a distance between the one or more retinal signal features and the biosignature of the condition, wherein the biosignature was determined using a sample dataset of retinal signal data from subjects with the condition, wherein the sample dataset comprises at least one parameter which is voltage independent and/or time independent;
- and
- displaying an interface indicating the distance between the one or more retinal signal features and the biosignature of the condition.
2 more elements
A system for predicting a probability that a human subject is subject to a condition, the system comprising:
- an electroretinogram (ERG) system comprising:
- a light stimulator that emits light;
- a photodetector for measuring an intensity of the emitted light from the light stimulator;
- a spectrometer for measuring the spectrum of the emitted light from the light stimulator;
- and
- one or more electrodes for collecting retinal signal data corresponding to the human subject, and
- a computer system comprising at least one processor and memory storing a plurality of executable instructions which, when executed by the at least one processor, cause the computer system to:
- receive the retinal signal data, wherein the retinal signal data is associated with a light stimulation of the human subject's retina and includes at least one parameter, recorded by the photodetector or the spectrometer, which is voltage independent and/or time independent, and wherein the at least one parameter comprises the intensity of the emitted light, the spectrum of the emitted light, or an impedance of a receiving circuit;
- determine, based on the retinal signal data, one or more retinal signal features;
- apply the one or more retinal signal features to a mathematical model, wherein the mathematical model corresponds to the condition, wherein the mathematical model comprises a machine learning algorithm trained using a sample dataset of retinal signal data from subjects with the condition, and wherein the sample dataset comprises at least one parameter which is voltage independent and/or time independent;
- output, by the mathematical model, a predicted probability for the condition;
- and
- display an interface comprising the predicted probability.
10 more elements
Dependencies are inferred from the claim text.