Systems and methods for intraocular lens selection
Systems and methods for intraocular lens selection include receiving, by one or more computing devices implementing a prediction engine, pre-operative multi-dimensional images of an eye; extracting, by the prediction engine, pre-operative measurements of the eye based on the pre-operative images; estimating, by the prediction engine using a prediction model based on a machine learning strategy, a post-operative position of an intraocular lens based on the extracted pre-operative measurements; selecting a power of the intraocular lens based on the estimated post-operative position of the intraocular lens; and selecting the intraocular lens based on the selected power. In some embodiments, the systems and methods further include receiving post-operative multi-dimensional images of the eye after implantation of the selected intraocular lens, extracting post-operative measurements of the eye, and updating the prediction model based on the pre-operative measurements and the post-operative measurements.
1. A method comprising: receiving, by one or more computing devices implementing a prediction engine, one or more pre-operative multi-dimensional images of an eye, wherein the pre-operative multi-dimensional images of the eye include each of: an angle to angle width describing a width of a line joining each of two angular recesses of a cornea of the eye; an angle to angle depth measured as a perpendicular distance between an intersection point on the line joining each of the two angular recesses of the cornea of the eye and a posterior corneal surface of the eye; an angle between a pupillary axis of the pre-operative eye and a line of sight axis of the eye when the eye is fixated on a fixation point; and an estimated position of a lens equator determined as an equator line between each of two intersection points of an anterior lens radius and a posterior lens radius;
extracting, by the prediction engine, one or more pre-operative measurements of the eye based on the one or more pre-operative images of the eye; estimating, by the prediction engine using a first prediction model based on a machine learning strategy, a post-operative position of an intraocular lens based on the one or more extracted pre-operative measurements of the eye; selecting a power of the intraocular lens based on at least the estimated post-operative position of the intraocular lens;
selecting the intraocular lens based on at least the selected power; receiving one or more post-operative multi-dimensional images of the eye after implantation of the selected intraocular lens;
extracting one or more post-operative measurements of the eye; and updating the first prediction model based on the one or more pre-operative measurements and the one or more post-operative measurements.
2. The method of claim 1 , wherein the first prediction model comprises a neural network.
3. The method of claim 1 , wherein the post-operative position of the intraocular lens is an anterior corneal depth (ACD) of the intraocular lens or a position of an equator of the intraocular lens.
4. The method of claim 1 , wherein selecting the intraocular lens is further based on one or more of the estimated post-operative position of the intraocular lens or a diameter of a pre-operative lens.
5. The method of claim 1 , wherein the one or more pre-operative measurements include one or more of a group consisting of:
a diameter of a pre-operative lens of the eye;
a pre-operative anterior chamber depth of the eye, and
a refractive power of a cornea of the eye.
6. The method of claim 1 , wherein selecting the power comprises recommending, by the prediction engine using a second prediction model, a recommended power for the intraocular lens based on at least the estimated post-operative position.
7. The method of claim 6 , further comprising updating the second prediction model based on the recommended power and the selected power.
8. The method of claim 1 , further comprising estimating, by the prediction engine using a second prediction model, a post-operative manifest refraction spherical equivalent (MRSE) based on at least the estimated post-operative position and the selected power.
9. The method of claim 8 , further comprising:
measuring an actual power-operative MRSE; and
updating the second prediction model based on the estimated post-operative MRSE and the actual power-operative MRSE.
10. The method of claim 1 , wherein the one or more pre-operative multi-dimensional images are each received from a diagnostic device selected from a group consisting of an optical coherence tomography (OCT) device, a rotating Scheimpflug camera, and a magnetic resonance imaging (MRI) device.
11. The method of claim 1 , further comprising planning one or more procedures for implantation of the intraocular lens.
12. The method of claim 1 , further comprising displaying one of the one or more pre-operative multi-dimensional images annotated with one of the one or more extracted pre-operative measurements.
13. A prediction engine comprising: one or more processors; wherein the prediction engine is configured to: receive one or more pre-operative multi-dimensional images of an eye obtained by a diagnostic device, wherein the pre-operative multi-dimensional images of the eye include each of: an angle to angle width describing a width of a line joining each of two angular recesses of a cornea of the eye; an angle to angle depth measured as a perpendicular distance between an intersection point on the line joining each of the two angular recesses of the cornea of the eye and a posterior corneal surface of the eye; an angle between a pupillary axis of the pre-operative eye and a line of sight axis of the eye when the eye is fixated on a fixation point; and an estimated position of a lens equator determined as an equator line between each of two intersection points of an anterior lens radius and a posterior lens radius; extract one or more pre-operative measurements of the eye based on the one or more pre-operative images of the eye; estimate, using a first prediction model based on a machine learning strategy, a post-operative position of an intraocular lens based on the one or more extracted pre-operative measurements of the eye; recommend a power of the intraocular lens based on at least the estimated post-operative position of the intraocular lens; provide the recommended power to a user to facilitate selection of the intraocular lens for implantation; receive one or more post-operative multi-dimensional images of the eye after implantation of the selected intraocular lens; extract one or more post-operative measurements of the eye; and update the first prediction model based on the one or more pre-operative measurements and the one or more post-operative measurements.
14. The prediction engine of claim 13 , wherein the post-operative position of the intraocular lens is an anterior corneal depth (ACD) of the intraocular lens or a position of an equator of the intraocular lens.
15. The prediction engine of claim 13 , wherein the one or more pre-operative measurements include one or more of a group consisting of:
a diameter of a pre-operative lens of the eye;
a pre-operative anterior chamber depth of the eye, and
a refractive power of a cornea of the eye.
16. The prediction engine of claim 13 , further comprising estimating, by the prediction engine using a second prediction model, a post-operative manifest refraction spherical equivalent (MRSE) based on at least the estimated post-operative position and a selected power of the intraocular lens.
17. A non-transitory machine-readable medium comprising a plurality of machine-readable instructions which when executed by one or more processors are adapted to cause the one or more processors to perform a method comprising: receiving one or more pre-operative multi-dimensional images of an eye, wherein the pre-operative multi-dimensional images of the eye include each of: an angle to angle width describing a width of a line joining each of two angular recesses of a cornea of the eye; an angle to angle depth measured as a perpendicular distance between an intersection point on the line joining each of the two angular recesses of the cornea of the eye and a posterior corneal surface of the eye; an angle between a pupillary axis of the pre-operative eye and a line of sight axis of the eye when the eye is fixated on a fixation point; and an estimated position of a lens equator determined as an equator line between each of two intersection points of an anterior lens radius and a posterior lens radius; extracting one or more pre-operative measurements of the eye based on the one or more pre-operative images of the eye; estimating, using a prediction model based on a machine learning strategy, a post-operative position of an intraocular lens based on the one or more extracted pre-operative measurements of the eye; recommending a power of the intraocular lens based on at least the estimated post-operative position of the intraocular lens; providing the recommended power to a user to facilitate selection of the intraocular lens for implantation; receiving one or more post-operative multi-dimensional images of the eye after implantation of the selected intraocular lens; extracting one or more post-operative measurements of the eye; and updating the prediction model based on the one or more pre-operative measurements and the one or more post-operative measurements.