System for predicting post-cataract surgery vision in patients with corneal opacity based on corneal topography parameters

Abstract

The application discloses a system for predicting postoperative vision of a cataract of a corneal opacity patient based on corneal topographic parameters, and the system determines the prediction value of multi-dimensional characteristics of the corneal opacity on postoperative vision of the cataract, provides a new quantitative tool for preoperative evaluation of postoperative vision of the cataract of the corneal opacity patient, and enriches the mechanism research of vision impairment caused by the corneal opacity.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and specifically to a system for predicting postoperative visual acuity in patients with corneal opacity after cataract surgery based on corneal topography parameters. Background Technology

[0002] Cataracts, the leading cause of blindness worldwide, have entered an era of precision refractive surgery. Phacoemulsification combined with intraocular lens implantation has become the preferred surgical method in clinical practice due to its advantages of minimal invasiveness, rapid recovery, and definite efficacy. With the continuous maturation of surgical techniques, postoperative visual recovery no longer solely depends on the precision of the surgical procedure; the preoperative corneal condition's impact on postoperative visual quality is receiving increasing attention. Corneal opacity is a common corneal lesion, essentially caused by disordered arrangement of collagen fibers or accumulation of deposits in the corneal stroma. It can disrupt corneal optical homogeneity, alter refractive states, and lead to abnormal light refraction. Even if cataract surgery successfully removes the opaque lens, corneal opacity can still become a bottleneck for postoperative visual recovery. In clinical practice, postoperative visual prognosis in patients with corneal opacity exhibits significant individual differences. Traditional preoperative assessments often rely on the doctor's subjective experience and simple visual acuity and astigmatism tests, lacking objective quantitative indicators and making it difficult to accurately predict postoperative visual recovery. Corneal topography, as a non-invasive and precise corneal morphology detection technology, uses accompanying corneal optical density software to quantitatively and locally evaluate the density of backscattered light intensity in a 12mm diameter corneal region using grayscale values. This makes corneal optical density an effective indicator for evaluating corneal health and opacity. Simultaneously, it can acquire key optical parameters such as corneal aberrations and spherical aberrations, providing objective evidence for corneal optical quality assessment. However, research on systematically integrating multi-dimensional parameters of corneal topography to analyze their correlation with postoperative visual acuity and constructing predictive models for patients with corneal opacity and cataracts remains lacking. Summary of the Invention

[0003] To address the shortcomings and deficiencies of existing technologies, this invention provides a system for predicting postoperative visual acuity in patients with corneal opacity after cataract surgery based on corneal topography parameters. The system analyzes the correlation between corneal topography parameters (corneal opacity characteristics, corneal aberrations) and postoperative visual acuity in patients with corneal opacity after simple cataract surgery. It employs multivariate stepwise regression to screen key influencing factors and constructs a postoperative visual acuity prediction model. This provides a scientific tool and theoretical support for accurate preoperative prognosis assessment and postoperative intervention strategies in clinical practice.

[0004] The technical solution adopted in this invention is: a system for predicting postoperative visual acuity in patients with corneal opacity after cataract surgery based on corneal topography parameters, including an information acquisition and data extraction module, an image processing module, and a visual acuity prediction module. The information acquisition and data extraction module extracts raw corneal optical density data (including corneal optical density map and raw corneal optical density data matrix) and corneal optical parameters based on the corneal topography map; The image processing module reproduces the corneal optical density map based on the original data matrix of the Scheimpflug corneal topography imaging, manually delineates the opacity area, and calculates the average optical density value and opacity area ratio of four regions with diameters of 0-2 mm, 2-4 mm, 4-6 mm, and 6-7 mm centered on the corneal apex. The vision prediction module uses a prediction model to predict the postoperative vision of patients with corneal opacity after cataract surgery. The prediction model is: y=-0.108+0.002a+0.196h+0.005s×d; Where y is the best corrected visual acuity after surgery, a is the age, h is the total higher-order aberration of the cornea, s is the proportion of the 0-2mm region of opacity, and d is the average corneal optical density of the 0-2mm region.

[0005] The corneal opacity features described are derived from the analysis of the original data matrix of the corneal light density map.

[0006] The average corneal optical density of the region refers to the average optical density values ​​of four different regions with diameters of 0-2 mm, 2-4 mm, 4-6 mm, and 6-7 mm centered on the corneal apex.

[0007] The aforementioned regional opacity area ratio refers to the opacity area ratio of four different regions with diameters of 0-2 mm, 2-4 mm, 4-6 mm, and 6-7 mm centered on the corneal apex.

[0008] The corneal optical parameters mentioned are the total higher-order aberrations of the cornea.

[0009] The method for extracting raw corneal optical density data and corneal optical parameters is as follows: Extract the raw optical density image data matrix, and reproduce the image using MATLAB code; based on the reproduced image, mark the cloudy areas, and calculate the cloudy characteristic parameters: the average optical density value and cloudy area ratio of four regions 0-2mm, 2-4mm, 4-6mm, and 6-7mm, centered on the corneal vertex; extract the total higher-order aberrations of the cornea within a 4mm range measured by corneal topography.

[0010] The beneficial effects of this invention are: This invention provides a system for predicting postoperative visual acuity in patients with corneal opacity based on corneal topography parameters, clarifies the predictive value of multidimensional characteristics of corneal opacity for postoperative visual acuity in cataract surgery, provides a new quantitative tool for clinical preoperative assessment of postoperative visual acuity in patients with corneal opacity, and enriches the research on the mechanism of visual impairment caused by corneal opacity. Attached Figure Description

[0011] Figure 1 Raw corneal optical density image measured by Pentacam.

[0012] Figure 2 This is the original numerical matrix of corneal optical density.

[0013] Figure 3 This is the interface for annotating the opacity of the original numerical matrix of corneal optical density after it has been encoded using a MATLAB program.

[0014] Figure 4 The images are the output images after processing by the MATLAB program; (a) is the original data matrix of the light density image extracted from the pentacam software, and the image reproduced after being encoded by the MATLAB program; (b) is the image with the turbid areas marked by the same operator based on the reproduced image.

[0015] Figure 5 For Pentacam Cataract Pre-OP images: extract the Total Corneal HOA (4mm).

[0016] Figure 6 This is the original corneal optical density image of Sample 1.

[0017] Figure 7 This is the original numerical matrix of corneal optical density for sample 1.

[0018] Figure 8 Processing of corneal optical density image for sample 1.

[0019] Figure 9 This is the original corneal optical density image of Sample 2.

[0020] Figure 10 This is the original numerical matrix of corneal optical density for sample 2.

[0021] Figure 11 Processing of corneal optical density image for sample 2.

[0022] Figure 12 This is the original corneal optical density image of sample 3.

[0023] Figure 13 This is the original numerical matrix of corneal optical density for sample 3.

[0024] Figure 14 Processing of corneal optical density image for sample 3. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative effort are all within the scope of protection of the present invention.

[0026] Data extraction and processing Corneal topography Measurements were taken using Pentacam (OCULUS, Germany) equipment, and the extracted data included: ① corneal densitogram: original corneal densitogram and original data matrix; ② corneal optical parameters: total higher-order aberrations (Total HOA, μm).

[0027] Image processing Based on the raw corneal optical density data matrix, the corneal optical density map was reproduced using MATLAB R2024b software. To eliminate interference from arcus senilis and eyelid eyelash obstruction, a diameter range of 7 mm was considered clinically significant. The opacified areas were manually delineated, and the corneal opacity characteristics were calculated using built-in code across four regions, totaling eight parameters: the cornea was divided into four regions centered on the corneal apex: 0-2 mm, 2-4 mm, 4-6 mm, and 6-7 mm. The average optical density value and opacity area ratio of each region were output.

[0028] result This study ultimately retrospectively included 265 eyes from 265 individuals, with a mean age of 70.99 ± 9.36 years. Specific demographic and ocular baseline characteristics of the sample are shown in Table 1, where quantitative variables are described as mean ± standard deviation.

[0029] Table 1. Baseline characteristics of the eye

[0030] HOA, Higher-Order Aberration, is the "Total Corneal HOA (4mm)" from the pentacam report, which refers to the total corneal higher-order aberrations within a 4mm diameter range centered on the corneal apex; GSU, Grayscale Unit; CDVA, Corrected Distance Visual Acuity; LogMAR, Logarithm of the Minimum Angle of Resolution.

[0031] Correlation analysis Spearman's rank correlation analysis was used to explore the correlation between various clinical indicators, corneal topography parameters, and postoperative visual acuity. Relevant indicators were screened, and the results are shown in Table 2. The results showed that postoperative visual acuity was significantly correlated with gender and age. After controlling for the influence of age using analysis of covariance (ANCOVA), the results showed that gender had no significant independent effect on postoperative visual acuity (F=3.441, p=0.065), while the main effect of age was significant (F=8.718, p=0.003).

[0032] Among corneal parameters, postoperative visual acuity was strongly correlated with HOA (r=0.558); significantly correlated with the average optical density of each corneal region (p<0.001); and significantly correlated with the proportion of opacity area in the central pupillary region (0-2mm, 2-4mm). Based on these findings, and considering regional characteristics, a weighted index for corneal opacity burden in each region was defined as "weighted value = regional average optical density × regional opacity area proportion," and its correlation was further analyzed. The results showed that postoperative visual acuity was correlated with the weighted values ​​of the 0-2 mm, 2-4 mm, and 4-6 mm regions, and the correlation coefficients showed a decreasing trend from the center to the periphery. The correlation was most significant in the central 0-2 mm region (r=0.393). This result supports the clinical understanding that "central corneal opacity has a greater impact on visual acuity than peripheral corneal opacity."

[0033] Table 2 Correlation analysis of postoperative visual acuity

[0034] Building a prediction model This study first used univariate linear regression analysis to preliminarily screen variables that may be related to postoperative best-corrected visual acuity (HOA), as shown in Table 3. The results showed that postoperative visual acuity was statistically significantly associated with age and HOA. Among the opacity characteristics, the mean optical density and the proportion of opacity area in the 0-2, 2-4, and 4-6 mm regions all showed linear regression relationships with postoperative visual acuity, while only the mean optical density in the 6-7 mm region showed a linear correlation with postoperative visual acuity.

[0035] To further control for confounding factors and identify independent influencing factors of postoperative visual acuity, a multivariate stepwise regression model was constructed. The inclusion criterion was set at α=0.05, and the exclusion criterion at α=0.10. Statistically significant independent variables were selected through stepwise regression to determine the predictors of the model. Finally, age, HOA, and four-region parameters were substituted into the model for regression.

[0036] Considering that optical density and area together represent the turbidity characteristics and optical properties within a region, four "weighted values" (area ratio × optical density) were introduced as turbidity load indicators for each region to further optimize model performance. The final model results are shown in Tables 4-5, with age (B=0.002), HOA (B=0.196), and weighted 0_2 (B=0.005) included as predictors. The model variance inflation factor (VIF) was around 1.0, less than 10, indicating the absence of multicollinearity; the Durbin-Watson statistic (DW statistic = 2.048) was around 2.0, indicating no autocorrelation in the residuals, satisfying the regression independence assumption. The constructed multiple regression model was statistically significant (F = 71.230, p < 0.001), explaining 44.4% of postoperative visual acuity variation (adjusted R² = 0.444).

[0037] Table 3 Linear univariate regression model for postoperative visual acuity

[0038] B is the unstandardized coefficient; CI is the confidence interval.

[0039] Table 4. Multiple stepwise regression model of postoperative visual acuity

[0040] Table 5 Model Fitting Results

[0041] Define y as postoperative best corrected visual acuity, a as age, h as total higher-order corneal aberration (HOA), s as the proportion of corneal opacity in the 0-2mm region, and d as the average corneal optical density in the 0-2mm region. The prediction model is as follows: y=-0.108+0.002a+0.196h+0.005s×d Taking three samples as an example, the model prediction performance of this study is shown in Table 6-7.

[0042] Table 6 Model Prediction Results

[0043] Table 7 Model Prediction Error

[0044] *The overall metrics are based on this sample and are only used for intuitive reference in demonstrating the model's performance.

[0045] in conclusion In summary, this study clarifies the predictive value of multidimensional characteristics of corneal opacity for postoperative visual acuity in cataract surgery, provides a new quantitative tool for clinical preoperative assessment of postoperative visual prognosis in patients with corneal opacity, and enriches the research on the mechanism of visual impairment caused by corneal opacity.

[0046] The specific embodiments described in this invention are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains can make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding its defined scope. Although the invention has been detailed and described in the accompanying drawings and foregoing description, such descriptions are considered illustrative or exemplary rather than restrictive. It should be understood that changes and modifications can be made by those skilled in the art within the scope of the following claims.

Claims

1. A system for predicting postoperative visual acuity in patients with corneal opacity after cataract surgery based on corneal topography parameters, characterized in that, It includes an information collection and data extraction module, an image processing module, and a vision prediction module. The information acquisition and data extraction module extracts raw corneal optical density data and corneal optical parameters based on the corneal topography map. The raw corneal optical density data includes a corneal optical density map and a raw corneal optical density data matrix. The image processing module reproduces the corneal optical density map based on the raw data matrix of Scheimpflug imaging of corneal topography, and artificially delineates the cloudy area by dividing the corneal diameter region with the corneal vertex as the center, and calculates the corneal cloudiness characteristics of the region. The vision prediction module uses a prediction model to predict the postoperative vision of patients with corneal opacity after cataract surgery. The prediction model is: y = -0.108 + 0.002a + 0.196h + 0.005s × d; Where y is the best corrected visual acuity after surgery, a is the age, h is the total higher-order aberration of the cornea, s is the proportion of corneal opacity in the 0-2mm area, and d is the average corneal optical density in the 0-2mm area.

2. The system for predicting postoperative visual acuity in patients with corneal opacity based on corneal topography parameters according to claim 1, characterized in that, The corneal opacity characteristics are the regional average corneal optical density and the proportion of regional opacity area obtained by analyzing the raw data matrix of the corneal optical density map obtained by Scheimpflug imaging.

3. The system for predicting postoperative visual acuity in patients with corneal opacity based on corneal topography parameters according to claim 2, characterized in that, The average corneal optical density of the region refers to the average optical density values ​​of four different regions with diameters of 0-2 mm, 2-4 mm, 4-6 mm, and 6-7 mm centered on the corneal apex.

4. The system for predicting postoperative visual acuity in patients with corneal opacity based on corneal topography parameters according to claim 2, characterized in that, The aforementioned regional opacity area ratio refers to the opacity area ratio of four different regions with diameters of 0-2 mm, 2-4 mm, 4-6 mm, and 6-7 mm centered on the corneal apex.

5. The system for predicting postoperative visual acuity in patients with corneal opacity based on corneal topography parameters according to claim 1, characterized in that, The corneal optical parameters mentioned are the total higher-order aberrations of the cornea.

6. The system for predicting postoperative visual acuity in patients with corneal opacity based on corneal topography parameters according to claim 1, characterized in that, The method for extracting raw corneal optical density data and corneal optical parameters is as follows: extract the raw optical density image data matrix, reproduce the image after encoding with a MATLAB program; based on the reproduced image, mark the cloudy areas, and calculate the cloudy characteristic parameters: the average optical density value and the proportion of cloudy area in four regions, 0-2mm, 2-4mm, 4-6mm, and 6-7mm, centered on the corneal vertex; Total higher-order aberrations of the cornea are extracted from corneal topography measurements within a 4 mm range.

Images