Prediction of IOL power
Described are implementations of systems and methods for an improved machine learning-based system that incorporates pre-operative and intraoperative measurements captured during surgery, as well as additional patient-specific data, to provide an individualized, highly accurate post-operative manifest refraction prediction. According to some embodiments, a determination engine generates a predictive feature set of one or more predictors associated with diagnostic measurements of one or more eyes and performs a recursive selection operation using one or more combinations within the predictive feature set and one or more models to produce a most predictive subset, the most predictive subset having a highest prediction accuracy among other predictive subsets for post-operative manifest refraction. The determination engine generates a determination model by refining and retraining the one or more models of the recursive selection operation utilizing the most predictive subset.
1. A method, comprising:
receiving, by a determination engine executed by one or more processors, a first set of diagnostic measurements of one or more eyes prior to undergoing intraocular lens (IOL) implantation;
calculating, by the determination engine using the first set of diagnostic measurements, an intermediate prediction of post-operative manifest refraction;
receiving, by the determination engine, a plurality of cornea apex position and curvature measurements of the one or more eyes;
generating, by the determination engine, a predictive feature subset comprising one or more of the plurality of cornea apex position and curvature measurements, the first set of diagnostic measurements, and the intermediate prediction;
calculating, by the determination engine using the predictive feature subset, a final prediction of post-operative manifest refraction associated with one or more IOL parameters, wherein an IOL associated with the one or more IOL parameters is selected for implantation responsive to the final prediction of post-operative manifest refraction being zero or near zero.
2. The method of claim 1 , wherein calculating the final prediction comprises performing a linear regression algorithm on the predictive feature subset.
3. The method of claim 2 , wherein the predictive feature subset comprises a subset of the one or more of the plurality of cornea apex position and curvature measurements, the first set of diagnostic measurements, and the intermediate prediction having a highest prediction accuracy among other subsets for post-operative manifest refraction for training data comprising diagnostic measurements and measured post-operative outcomes.
4. The method of claim 1 , wherein the IOL is selected from a monofocal IOL, a multifocal IOL, a toric IOL, an extended depth-of-focus IOL, an adjustable IOL, or an accommodative IOL.
5. A system, comprising:
an infrared laser; and
a computing sub-system comprising a memory storing a determination engine and a processor executing the determination engine, wherein the determination engine during execution is configured to:
receive a first set of diagnostic measurements of one or more eyes prior to undergoing intraocular lens (IOL) implantation,
calculate, using the first set of diagnostic measurements, an intermediate prediction of post-operative manifest refraction,
receive, via the infrared laser, a plurality of cornea apex position and curvature measurements of the one or more eyes,
generate a predictive feature subset comprising one or more of the plurality of cornea apex position and curvature measurements, the first set of diagnostic measurements, and the intermediate prediction, and
calculate, using the predictive feature subset, a final prediction of post-operative manifest refraction associated with one or more IOL parameters, wherein an IOL associated with the one or more IOL parameters is selected for implantation responsive to the final prediction of post-operative manifest refraction being zero or near zero.
6. The system of claim 5 , wherein the IOL is selected from a monofocal IOL, a multifocal IOL, a toric IOL, an extended depth-of-focus IOL, an adjustable IOL, or an accommodative IOL.