Artificial lens diopter key parameter acquisition system and method based on data distillation

Abstract

This invention relates to a system and method for acquiring key parameters of intraocular lens (IOL) refractive power based on data distillation, belonging to the field of medical IOL technology. This invention collects anterior and posterior corneal surface curvature data and outputs a CSV format data matrix. Based on the obtained CSV data matrix, it undergoes two data distillation processes, sequentially completing the conversion from radius of curvature to refractive power, data sorting, difference calculation, outlier removal, and average value calculation, ultimately obtaining the anterior corneal tangential curvature Kf and the posterior corneal tangential curvature Kb, and then calculating the IOL refractive power. This invention achieves the objective and standardized acquisition of the anterior corneal tangential curvature Kf and the posterior corneal tangential curvature Kb parameters, effectively reducing parameter selection bias for patients with special corneal conditions, providing accurate input for the ZZ IOL FORMULA, improving the accuracy of IOL refractive power calculation, and is suitable for patients with various corneal morphologies, especially suitable for special corneal populations such as those after refractive surgery.

Description

Technical Field

[0001] This invention belongs to the field of medical intraocular lens technology, and in particular to a system and method for obtaining key parameters of intraocular lens refractive power based on data distillation. Background Technology

[0002] In cataract surgery, accurate calculation of the intraocular lens (IOL) refractive power is crucial to ensuring patients achieve clear vision post-surgery. ZZ IOL FORMULA, a high-precision IOL refractive power calculation method, significantly improves accuracy by incorporating the tangential curvature of the posterior corneal surface and calculating the entire refraction path of light segmentally, making it suitable for all types of patients.

[0003] In the ZZ IOL FORMULA calculation process, the anterior corneal tangential curvature (Kf) and posterior corneal tangential curvature (Kb) are core input parameters, and their accuracy directly determines the reliability of the final refractive power calculation result. In existing technologies, Kf and Kb are obtained by physicians using a corneal refractive power map provided by a three-dimensional anterior segment analysis system. Within a common area of ​​3mm centered on the pupil and corneal apex, the physician subjectively judges and selects the curvature data corresponding to the central point of the area with the minimum refractive power.

[0004] However, this method of obtaining parameters based on physicians' subjective experience has significant drawbacks: on the one hand, different physicians have different judgment criteria, leading to a lack of objectivity and standardization in parameter selection, which easily introduces human error; on the other hand, for patients with special corneal morphologies such as those after corneal refractive surgery or corneal transplantation, their corneal refractive distribution is irregular, making subjective judgment by physicians extremely difficult, easily leading to the selection of incorrect data, resulting in significant deviations in Kf and Kb, which in turn affects the accuracy of intraocular lens refractive power calculation and cannot meet the surgical needs of patients with special corneal morphologies. Moreover, some special corneal morphologies require data beyond 3mm, still requiring judgment by experienced physicians.

[0005] Therefore, there is an urgent need for an objective and standardized method for obtaining Kf and Kb parameters to overcome the limitations of physicians' subjective experience, reduce parameter selection bias in special corneal patients, provide stable and accurate input for ZZ IOL FORMULA, and further improve the accuracy and reliability of intraocular lens refractive power calculation. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a system and method for acquiring key parameters of intraocular lens refractive power based on data distillation. Through standardized data acquisition and a two-stage distillation algorithm, the objective and accurate acquisition of the parameters of anterior corneal tangential curvature Kf and posterior corneal tangential curvature Kb is achieved. This effectively solves the problem of bias caused by the reliance on subjective experience in traditional methods. It is especially suitable for special corneal patients and provides reliable parameter support for ZZ IOL FORMULA, which has important clinical application value.

[0007] The technical problem solved by this invention is achieved through the following technical solution: The system for acquiring key parameters of intraocular lens refractive power based on data distillation includes a data acquisition module, a parameter acquisition module, and a refractive power calculation module, which are connected sequentially. The data acquisition module is used to acquire the curvature data of the anterior and posterior corneal surfaces and output a CSV format data matrix. The parameter acquisition module is used to extract the tangential curvature Kf and tangential curvature Kb of the anterior and posterior corneal surfaces from the CSV format data matrix. The refractive power calculation module is based on ZZ IOL FORMULA and calculates the refractive power of the intraocular lens based on the obtained tangential curvature Kf, tangential curvature Kb, corneal thickness, anterior chamber depth, axial length, intraocular lens parameters, and reserved refractive power.

[0008] Furthermore, the artificial lens parameters include the A constant and thickness of the artificial lens.

[0009] Furthermore, if the artificial lens used is a Toric artificial lens, the diopter calculation module also introduces a steep axis to correct the diopter.

[0010] A method for acquiring key parameters of intraocular lens refractive power based on data distillation includes the following steps: Step 1: The data acquisition module collects the curvature data of the anterior and posterior surfaces of the cornea and outputs a data matrix in CSV format; Step 2: The parameter acquisition module performs the first data distillation on the CSV format data to obtain the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb. Step 3: The parameter acquisition module performs a second data distillation on the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb to obtain the anterior corneal tangential curvature Kf and the posterior corneal tangential curvature Kb. Step 4: The refractive power calculation module is based on ZZ IOL FORMULA and calculates the refractive power of the intraocular lens based on the obtained anterior corneal tangential curvature Kf, posterior corneal tangential curvature Kb, corneal thickness, anterior chamber depth, axial length, intraocular lens parameters, and reserved refractive power.

[0011] Moreover, the specific implementation method of step 1 is as follows: the data acquisition module scans the cornea through the three-dimensional anterior segment analysis system, divides the anterior surface of the cornea into circumferences on an average basis, takes sampling points on an average basis for each circumference, and obtains the curvature data of the anterior surface; similarly, the curvature data of the posterior surface of the cornea is acquired, and the curvature data of the anterior and posterior surfaces are output as CSV format data matrices respectively.

[0012] Furthermore, the specific implementation method of step 2 is as follows: the parameter acquisition module processes each curvature data in the CSV format data matrix and converts the radius of curvature into the corresponding diopter: D=(n - 1) / R Where D is the diopter, n is the refractive index of the corresponding medium, and R is the radius of curvature, thus obtaining the anterior corneal surface diopter dataset Ff and the posterior corneal surface diopter dataset Fb.

[0013] Furthermore, step 3 includes the following steps: Step 3.1: The parameter acquisition module sorts the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb by numerical value, respectively. Step 3.2: Calculate the difference in refractive power between adjacent data in the sorted dataset. Data with large differences are directly removed and not included in subsequent calculations. Step 3.3: Remove outlier data; Step 3.4: Calculate the average value of the remaining Ff data after removing outliers, and use it as the tangential curvature Kf of the anterior corneal surface; calculate the average value of the remaining Fb data after removing outliers, and use it as the tangential curvature Kb of the posterior corneal surface.

[0014] Moreover, the specific implementation method of step 3.3 is as follows: data with a difference greater than a set threshold between the anterior corneal refractive power dataset Ff and the visual axis refractive power dataset Fb are removed, where the visual axis refractive power is the refractive power corresponding to the median of the sorted dataset.

[0015] The advantages and positive effects of this invention are: 1. This invention automatically collects curvature data from 3001 sampling points using a three-dimensional anterior segment analysis system, and obtains the anterior corneal tangential curvature Kf and posterior corneal tangential curvature Kb by combining a standardized data distillation algorithm. This completely eliminates the subjective judgment of physicians, avoids human error, and achieves standardization and uniformity in parameter acquisition.

[0016] 2. This invention uses two data distillation processes. First, the radius of curvature is converted into diopter. Then, through sorting, difference calculation, and abnormal data removal, extreme values ​​and interference data are effectively filtered out to ensure that the anterior corneal tangential curvature Kf and posterior corneal tangential curvature Kb can accurately reflect the true refractive characteristics of key areas of the cornea. It is especially suitable for patients with special corneal morphology after corneal refractive surgery and significantly reduces parameter deviation.

[0017] 3. The sampling area of ​​this invention covers the common area of ​​3mm centered on the pupil and corneal apex required by ZZ IOL FORMULA. The data processing process conforms to the optical principles of the eye, can be perfectly adapted to ZZ IOL FORMULA, and is suitable for patients with various corneal morphologies, providing a reliable guarantee for the accurate calculation of intraocular lens refractive power. Attached Figure Description

[0018] Figure 1 This is a block diagram of the system for acquiring key refractive parameters of intraocular lenses based on data distillation, as described in this invention. Figure 2 This is a flowchart of the acquisition method for the key parameters acquisition system of intraocular lens refractive power based on data distillation according to the present invention. Detailed Implementation

[0019] The present invention will be further described in detail below with reference to the accompanying drawings.

[0020] A system for acquiring key parameters of intraocular lens refractive power based on data distillation, such as Figure 1 As shown, the system includes a data acquisition module, a parameter acquisition module, and a refractive power calculation module, which are connected sequentially. The data acquisition module is used to acquire the curvature data of the anterior and posterior corneal surfaces and output a CSV format data matrix. The parameter acquisition module is used to extract the anterior corneal tangential curvature Kf and the posterior corneal tangential curvature Kb from the CSV format data matrix. The refractive power calculation module is based on ZZ IOL FORMULA and calculates the refractive power of the intraocular lens (IOL) based on the obtained anterior corneal tangential curvature Kf, posterior corneal tangential curvature Kb, corneal thickness (CT), anterior chamber depth (ACT), axial length (AL), IOL parameters (A constant, thickness), and reserved refractive power.

[0021] The parameters of the intraocular lens include the A constant and thickness. When the Toric intraocular lens is used, the diopter calculation module also introduces a steep axis to correct the diopter, adapting to the astigmatism correction requirements.

[0022] A method for acquiring key parameters of intraocular lens refractive power based on data distillation, such as... Figure 2 As shown, it includes the following steps: Step 1: The data acquisition module collects the curvature data of the anterior and posterior surfaces of the cornea and outputs a data matrix in CSV format.

[0023] The data acquisition module scans the cornea using a 3D anterior segment analysis system, dividing the anterior corneal surface into 31 circumferences on average. An average of 99 sampling points are taken from each circumference, resulting in a total of 3001 points of curvature data. Similarly, curvature data from 3001 points on the posterior corneal surface is acquired. The anterior and posterior surface curvature data are output as CSV format data matrices. The 31 circumferences are centered on the corneal apex, with a radius covering a 3mm common area centered on both the pupil and the corneal apex, ensuring that the sampling data covers the critical areas required for ZZIOL FORMULA.

[0024] The three-dimensional anterior segment analysis system supports the output of three-dimensional curvature data of the anterior and posterior surfaces of the cornea.

[0025] Step 2: The parameter acquisition module performs the first data distillation on the CSV format data to obtain the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb.

[0026] The parameter acquisition module processes each curvature data point in the CSV format data matrix, converting the radius of curvature into the corresponding diopter: D=(n - 1) / R Where D is the refractive power, n is the refractive index of the corresponding medium, and R is the radius of curvature. When converting the anterior corneal surface, n is taken as the corneal refractive index of 1.376, and when converting the posterior corneal surface, n is taken as the aqueous humor refractive index of 1.336 to ensure that the conversion process conforms to the optical characteristics of the eye, thereby obtaining the anterior corneal surface refractive power dataset Ff and the posterior corneal surface refractive power dataset Fb.

[0027] Step 3: The parameter acquisition module performs a second data distillation on the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb to obtain the anterior corneal tangential curvature Kf and the posterior corneal tangential curvature Kb.

[0028] Step 3 includes the following steps: Step 3.1: The parameter acquisition module sorts the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb by numerical value to facilitate subsequent difference calculation and anomaly identification. Step 3.2: Calculate the difference in refractive power between adjacent data in the sorted dataset, and determine the stability of the data by the distribution of the difference; Step 3.3: Remove outlier data; The median was calculated from 36 values, from smallest to largest, extracted from the anterior surface refractive power data.

[0029] Each of the 36 values ​​on the front surface is compared with its median, and data with a difference exceeding 1.50D are discarded. This is to avoid extreme outliers affecting the accuracy of the parameters.

[0030] Further selection involves comparing the difference between the 15th refractive power data (from smallest to largest) and the 15th data (from largest to smallest) in the anterior surface refractive power data set, as well as the mean values ​​of the inner three circles and outer three circles of refractive power in the refractive power matrix. When the difference is ≥4.0 and the mean value of the outer three circles of refractive power is higher, Kf = the mean of the effective anterior surface data and the mean value of the inner eight circles of refractive power. Otherwise, Kf is directly set to the mean value of the inner eight circles of refractive power.

[0031] The back surface is similar, except that among the 36 values ​​on the back surface, the difference standard is 0.15 D for elimination; the difference standard for the 15th data from smallest to largest and the 15th data from largest to smallest is 1.0 D, to obtain Kb.

[0032] Step 4: The refractive power calculation module is based on ZZ IOL FORMULA and calculates the refractive power of the intraocular lens based on the obtained anterior corneal tangential curvature Kf, posterior corneal tangential curvature Kb, corneal thickness, anterior chamber depth, axial length, intraocular lens parameters, and reserved refractive power.

[0033] Example 1: Based on the above-mentioned system and method for obtaining key parameters of intraocular lens refractive power based on data distillation, the effectiveness of the present invention was verified by using post-corneal refractive surgery patients as research subjects, obtaining the anterior corneal tangential curvature Kf and posterior corneal tangential curvature Kb using the method of the present invention, and using them to calculate the refractive power of the intraocular lens in ZZ IOL FORMULA.

[0034] Step 1, Data Collection: The patient's cornea was scanned using a three-dimensional anterior segment analysis system. The anterior surface of the cornea was divided into 31 circumferences (radii covering an area of ​​0-9 mm) centered on the corneal apex. 99 sampling points were taken from each circumference, resulting in a total of 3001 points of curvature radius data, which were output as a CSV format anterior surface data matrix. Similarly, 3001 points of curvature radius data were collected from the posterior surface of the cornea, and the output was a CSV format posterior surface data matrix.

[0035] Step 2, First data distillation: For each radius of curvature Rf in the anterior surface data matrix, the refractive power is calculated using the formula Df=(1.376-1) / Rf, resulting in the anterior surface refractive power dataset Ff; for each radius of curvature Rb in the posterior surface data matrix, the refractive power is calculated using the formula Db=(1.336-1.376) / Rb, resulting in the posterior surface refractive power dataset Fb.

[0036] Step 3: Second data distillation: Step 3.1: Sort the anterior surface refractive power dataset Ff in ascending order of values ​​to obtain the sorted Ff=[33.2D,33.3D,..., 35.8D]; Sort the posterior surface refractive power dataset Fb in ascending order of values ​​to obtain the sorted Fb=[5.9D,6.0D,..., 7.2D].

[0037] Step 3.2: Calculate the difference between adjacent data in the anterior surface refractive power dataset Ff. The difference is between 0.05 and 0.1D, indicating good data stability. Calculate the difference between adjacent data in the posterior surface refractive power dataset Fb. The difference is between 0.03 and 0.08D, indicating good data stability.

[0038] Step 3.3: The median of the anterior surface refractive power dataset Ff is 34.5D. Data with a difference greater than 1.50D from 34.5D are removed (no outliers). The median of the posterior surface refractive power dataset Fb is 6.5D. Data with a difference greater than 1.0D from 6.5D are removed (no outliers).

[0039] Step 3.4: Calculate the average value of the anterior surface refractive power dataset Ff, which is 34.48D, i.e., the tangential curvature of the anterior corneal surface Kf = 34.48D; calculate the average value of the posterior surface refractive power dataset Fb, which is 6.47D, i.e., the tangential curvature of the posterior corneal surface Kb = 6.47D.

[0040] The anterior corneal tangential curvature Kf=34.48D, the posterior corneal tangential curvature Kb=6.47D, and the acquired CT=408μm, ACT=2.83mm, AL=29.05mm, intraocular lens A constant=118.9, and reserved refractive power=0.0001D were input into ZZ IOLFORMULA to calculate the intraocular lens refractive power as 18.12D. The actual postoperative refraction result was 18.10D, with an error of only 0.02D, which is significantly better than the calculation error of traditional subjective parameter selection (usually greater than 0.5D).

[0041] Example 2 Based on the above-described system and method for obtaining key parameters of intraocular lens refractive power using data distillation, the effectiveness of the method of the present invention was verified by using ordinary cataract patients as research subjects.

[0042] Step 1: Data Collection The patient's cornea was scanned using a three-dimensional anterior segment analysis system. The anterior surface of the cornea was divided into 31 circumferences centered on the corneal vertex. 99 sampling points were taken from each circumference, resulting in a total of 3001 points of curvature radius data. The data was output as a CSV format anterior surface data matrix. Similarly, the curvature radius data of 3001 points on the posterior surface of the cornea was collected, and the data was output as a CSV format posterior surface data matrix.

[0043] Step 2: First data distillation: For each radius of curvature Rf in the anterior surface data matrix, the refractive power is calculated using the formula Df=(1.376-1) / Rf, resulting in the anterior surface refractive power dataset Ff; for each radius of curvature Rb in the posterior surface data matrix, the refractive power is calculated using the formula Db=(1.336-1.376) / Rb, resulting in the posterior surface refractive power dataset Fb.

[0044] Step 3: Second data distillation: Step 3.1: The anterior surface refractive error dataset Ff is sorted as [37.2D, 37.3D,..., 39.5D], and the posterior surface refractive error dataset Fb is sorted as [6.3D, 6.4D,..., 7.5D]. Step 3.2: Calculate the differences between adjacent data points; all differences are within a reasonable range.

[0045] Step 3.3: The median of the anterior surface refractive error dataset Ff is 38.3D, and one outlier with a difference of 1.6D is removed; the median of the posterior surface refractive error dataset Fb is 6.9D, and there are no outliers.

[0046] Step 3.4: The average value of the remaining data in the anterior surface refractive power dataset Ff is 38.25D (anterior corneal anterior surface tangential curvature Kf=38.25D), and the average value of the posterior surface refractive power dataset Fb is 6.88D (posterior corneal posterior surface tangential curvature Kb=6.88D).

[0047] Inputting the above parameters into ZZ IOL FORMULA, the calculated refractive power of the intraocular lens is 22.35D. The actual postoperative refraction result is 22.33D, with an error of 0.02D, which meets the clinical precision requirements.

[0048] It should be emphasized that the embodiments described in this invention are illustrative rather than limiting. Therefore, this invention includes, but is not limited to, the embodiments described in the specific implementation. Any other implementations derived by those skilled in the art based on the technical solutions of this invention are also within the scope of protection of this invention.

Claims

1. A system for acquiring key parameters of intraocular lens refractive power based on data distillation, characterized in that: It includes a data acquisition module, a parameter acquisition module, and a refractive power calculation module, which are connected sequentially. The data acquisition module is used to collect the curvature data of the anterior and posterior corneal surfaces and output a CSV format data matrix. The parameter acquisition module is used to extract the anterior corneal tangential curvature Kf and the posterior corneal tangential curvature Kb from the CSV format data matrix. The refractive power calculation module is based on ZZ IOL FORMULA and calculates the refractive power of the intraocular lens (IOL) based on the obtained anterior corneal tangential curvature Kf, posterior corneal tangential curvature Kb, corneal thickness, anterior chamber depth, axial length, IOL parameters, and reserved refractive power.

2. The system for acquiring key parameters of intraocular lens refractive power based on data distillation according to claim 1, characterized in that: The artificial lens parameters include the A constant and thickness of the artificial lens.

3. The system for acquiring key parameters of intraocular lens refractive power based on data distillation according to claim 1, characterized in that: If the artificial lens used is a Toric artificial lens, the refractive power calculation module also introduces a steep axis to correct the refractive power.

4. A method for acquiring key parameters of intraocular lens refractive power based on data distillation as described in any one of claims 1 to 3, characterized in that: Includes the following steps: Step 1: The data acquisition module collects the curvature data of the anterior and posterior surfaces of the cornea and outputs a data matrix in CSV format; Step 2: The parameter acquisition module performs the first data distillation on the CSV format data to obtain the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb. Step 3: The parameter acquisition module performs a second data distillation on the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb to obtain the anterior corneal tangential curvature Kf and the posterior corneal tangential curvature Kb. Step 4: The refractive power calculation module is based on ZZ IOL FORMULA and calculates the refractive power of the intraocular lens based on the obtained anterior corneal tangential curvature Kf, posterior corneal tangential curvature Kb, corneal thickness, anterior chamber depth, axial length, intraocular lens parameters, and reserved refractive power.

5. The method for acquiring key parameters of intraocular lens refractive power based on data distillation as described in claim 4, characterized in that: The specific implementation method of step 1 is as follows: the data acquisition module scans the cornea through the three-dimensional anterior segment analysis system, divides the anterior surface of the cornea into circumferences on an average basis, takes sampling points on an average basis for each circumference, and obtains the curvature data of the anterior surface. Similarly, collect the curvature data of the posterior corneal surface and output the curvature data of the anterior and posterior surfaces as CSV format data matrices.

6. The method for acquiring key parameters of intraocular lens refractive power based on data distillation as described in claim 4, characterized in that: The specific implementation method of step 2 is as follows: the parameter acquisition module processes each curvature data in the CSV format data matrix and converts the radius of curvature into the corresponding diopter. D=(n - 1) / R Where D is the diopter, n is the refractive index of the corresponding medium, and R is the radius of curvature, thus obtaining the anterior corneal surface diopter dataset Ff and the posterior corneal surface diopter dataset Fb.

7. The method for acquiring key parameters of intraocular lens refractive power based on data distillation as described in claim 4, characterized in that: Step 3 includes the following steps: Step 3.1: The parameter acquisition module sorts the anterior corneal refractive power dataset Ff and the posterior corneal refractive power dataset Fb by numerical value, respectively. Step 3.2: Calculate the difference in refractive power between adjacent data in the sorted dataset. Data with large differences are directly removed and not included in subsequent calculations. Step 3.3: Remove outlier data; Step 3.4: Calculate the average value of the remaining Ff data after removing outliers, and use it as the tangential curvature Kf of the anterior corneal surface; calculate the average value of the remaining Fb data after removing outliers, and use it as the tangential curvature Kb of the posterior corneal surface.

8. The method for acquiring key parameters of intraocular lens refractive power based on data distillation as described in claim 7, characterized in that: The specific implementation method of step 3.3 is as follows: data with a difference greater than a set threshold between the anterior corneal refractive power dataset Ff and the visual axis refractive power dataset Fb are removed, where the visual axis refractive power is the refractive power corresponding to the median of the sorted dataset.

Images