Comparative Evaluation of Traditional and AI-Based Intraocular Lens Power Calculation Formulas in Highly Myopic Eyes

Comparative Evaluation of Traditional and AI-Based Intraocular Lens Power Calculation Formulas in Highly Myopic Eyes | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Comparative Evaluation of Traditional and AI-Based Intraocular Lens Power Calculation Formulas in Highly Myopic Eyes Xiaopeng Jiang, Jiangjie Wang, Qingmin Jiang, Xiangyu Zhou, Fei Xia, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6777942/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 23 Sep, 2025 Read the published version in BMC Ophthalmology → Version 1 posted 13 You are reading this latest preprint version Abstract Purpose: To assess the accuracy of artificial intelligence (AI)-based intraocular lens (IOL) power calculation formulas compared with traditional methods in highly myopic eyes, and to evaluate their performance across varying axial lengths and corneal curvatures. Methods: This retrospective case series included 115 highly myopic eyes that underwent phacoemulsification with IOL implantation. IOL power was calculated using four conventional formulas (SRK/T, Haigis, Holladay 2, Barrett Universal II) and seven AI-based formulas (Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Ladas Super Formula, Kane, HM-ZL). The outcomes were evaluated using mean error (ME), mean absolute error (MAE), median absolute error (MedAE), and the percentage of eyes within ±0.25 D to ±1.00 D of the prediction error. Subgroup analyses were conducted based on axial length (AL) and corneal curvature (Kmean). Results: AI-based formulas—especially Hill-RBF 3.0, Hoffer QST, and PEARL-DGS—demonstrated significantly higher accuracy than traditional formulas. Hill-RBF 3.0 achieved the lowest MAE (0.50 D) and MedAE (0.33 D) and the highest percentage of eyes within ±0.50 D (67.83%）and ±1.00 D (89.57%). Subgroup analyses showed that AI formulas maintained consistent performance across various AL and Kmean categories. Significant differences were noted between AI-based and traditional formulas, particularly in eyes with extreme biometric values. Conclusion: AI-based formulas provide superior refractive prediction in highly myopic eyes compared with traditional methods, particularly in cases of long axial length or steep corneal curvature. Tailored formula selection based on biometric profiles may enhance refractive outcomes in cataract surgery. artificial intelligence intraocular lens high myopia IOL power formula prediction accuracy Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Background Myopia is a prevalent refractive error that poses a growing public health concern worldwide (1) . High myopia, defined as a refractive error of − 6.00 diopters (D) or more, not only impairs visual acuity but also predisposes individuals to a range of ocular complications, such as retinal detachment and macular degeneration (2) . The global burden of high myopia is expected to reach 938 million by 2050, underscoring the urgent need for effective management strategies (3) . Highly myopic eyes frequently present with complex posterior segment alterations, including peripapillary crescent formation, chorioretinal atrophy, and posterior staphyloma (4) . These anatomical variations can compromise the precision of axial length measurements, thereby reducing the accuracy of intraocular lens (IOL) power calculations and adversely impacting postoperative refractive outcomes (5) . With the rising expectations for optimal postoperative vision, accurate prediction of refractive outcomes has become increasingly critical in cataract surgery. The success of modern cataract procedures depends heavily on precise IOL power selection, which directly affects visual acuity and overall patient satisfaction (6) . Conventional IOL power calculation formulas, including SRK/T, Haigis, and Holladay 2, are based on geometric optics and regression modeling (7) . Although widely applied in clinical settings, their accuracy diminishes in eyes with extreme biometric values—such as markedly long or short axial lengths and unusually flat or steep corneal curvatures—conditions often present in high myopia (8) . Recently, artificial intelligence (AI) has emerged as a promising innovation in ophthalmology, particularly in enhancing the accuracy of IOL power calculations. AI-based formulas are developed based on different principles. For instance, the Ladas Super Formula (LSF) combines the advantages of multiple existing formulas to create a more comprehensive model (9) . Karmona Formula, Hill-RBF 3.0, and the HM-ZL formula purely utilize machine learning algorithms to model the complex relationships between various ocular biometric factors (10–12) . Kane, Pearl-DGS, and Hoffer QST integrate Gaussian optics with AI technology to improve the prediction accuracy of IOL power (12–14) . Notable examples include the Kane, Hill-RBF 3.0, Karmona, and others, which have shown encouraging results in difficult-to-predict eyes such as those with high myopia (15, 16) . Previous studies have systematically compared AI-based intraocular lens formulas with traditional formulas in patients with high myopia, covering a range of axial lengths and corneal curvatures (15, 17) . This study further explores the comparison between more AI-based formulas and traditional formulas, offering more accurate formula selection recommendations based on the individualized ocular characteristics of patients. 2. Methods 2.1 Ethics Statement This retrospective, consecutive case-series study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of the Affiliated Hospital of Shandong Second Medical University (ID: wyfy-2024-ky-432). Written informed consent was obtained from all patients prior to data collection and analysis. 2.2 Patient Selection and Subgrouping: Medical records of patients with high myopia who underwent cataract surgery between January and December 2024 at the Eye Center of the Affiliated Hospital of Shandong Second Medical University were retrospectively reviewed. Inclusion criteria were: age ≥ 18 years, a diagnosis of complicated cataract with high myopia (defined as spherical equivalent ≥ − 6.0 D or axial length ≥ 26.00 mm), complete preoperative and postoperative clinical data, and a postoperative best-corrected visual acuity (BCVA) of at least 20/40. Exclusion criteria included other ocular diseases, history of intraocular or refractive surgery, ocular trauma, missing biometric parameters, corneal astigmatism > − 3.00 D, or intraoperative complications affecting IOL position. If both eyes met the inclusion criteria, only one eye per patient was randomly selected using a random number table. Based on prior studies (8, 18) , eyes were categorized by axial length (AL) into three subgroups: Long (26.00 mm ≤ AL < 28.00 mm) Super-long (28.00 mm ≤ AL < 30.00 mm) Extreme-long (AL ≥ 30.00 mm) Corneal curvature (K mean ) was similarly stratified into: Flat (K mean 46.0 D) 2.3 Biometry and IOL Power Calculation Preoperative ocular biometry was performed using the IOLMaster® 700 (Carl Zeiss Meditec AG, Jena, Germany), which provided measurements for axial length (AL), keratometry (K), anterior chamber depth (ACD), lens thickness (LT), and white-to-white (WTW) distance. IOL power calculations were performed using 11 formulas: four traditional formulas (SRK/T, Haigis, Holladay 2, and Barrett Universal II) and seven AI-based formulas (Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Ladas Super Formula, Kane, and HM-ZL). A keratometric index of 1.3375 was used. A-constants were obtained from the IOLCon database ( www.iolcon.org ) as recommended by Hoffer and Savini (19) . All postoperative refractions were measured one month after surgery under mesopic conditions by certified technicians. 2.4 Surgical procedures All surgeries were performed by an experienced ophthalmic surgeon using phacoemulsification and intraocular lens (IOL) implantation. After topical anesthesia, a clear corneal incision was created. Continuous curvilinear capsulorhexis (CCC) was performed, with the capsulotomy diameter tailored to match the optic zone of the IOL (20) . The crystalline lens was emulsified and aspirated within the capsular bag using an ultrasonic phacoemulsification probe. Subsequently, the IOL was implanted into the capsular bag, and the corneal incision was secured to ensure a watertight seal. 2.5 Statistical Analysis Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 27.0 (IBM Corp., Armonk, NY, USA). The prediction error (PE) was defined as the difference between the actual postoperative refraction and the predicted value. The following metrics were calculated for each IOL formula: mean prediction error (ME), mean absolute error (MAE), and median absolute error (MedAE). The one-sample Wilcoxon signed-rank test was used to determine whether the prediction errors significantly deviated from zero. Comparisons of MAEs among the different formulas were conducted using the Friedman test with Bonferroni correction for multiple comparisons. The Bonferroni-adjusted Cochran's Q test was employed to assess differences in the proportions of eyes falling within predefined prediction error thresholds (± 0.25 D, ± 0.50 D, ± 0.75 D, and ± 1.00 D). A p-value < 0.05 was considered statistically significant. 3. Results 3.1Basic Patient Information A total of 115 eyes from 115 patients were included in the analysis. Demographic characteristics and biometric parameters are summarized in Table 1. The IOL models used included Rayner 920H, Prosert A1UL22, and HOYA XY1-SP. Based on prior literature suggesting refractive stability typically occurs between 2 weeks and 1 month after surgery ( 19 , 21 ) , we utilized refractive outcomes recorded one month postoperatively for analysis. 3.2 Refractive prediction error Table 2 and Figure 1 summarize the refractive prediction errors across all tested IOL formulas. The Karmona formula yielded the lowest mean error (ME, –0.03 D), while Hill-RBF 3.0 demonstrated the lowest mean absolute error (MAE, 0.50 D) and median absolute error (MedAE, 0.33 D). The HM-ZL formula achieved the highest proportion of eyes within ±0.25 D of the predicted value (41.74%). Table 2 and Figure 2 present the percentage of ME for each formula at ±0.25 D, ±0.50 D, ±0.75 D, and ±1.00 D. Hill-RBF 3.0 and Barrett Universal II (BU II) both had the highest percentages within ±0.50 D (67.83%), while BU II led in the ±0.75 D category (80.87%). The Hill-RBF 3.0 formula demonstrated the best performance within ±1.00 D, achieving 89.57% accuracy. Statistical analysis revealed significant differences among formulas in MAE and in the proportions of eyes falling within ±0.50 D, ±0.75 D, and ±1.00 D prediction error ranges (P < 0.05). 3.3 Influence of ocular bioparameters on prediction error 3.3.1 Axial Length Figure 3 illustrates the distribution of prediction errors across the axial length spectrum (26 mm to 35 mm), and Table 3 presents the prediction errors of each formula in different axial length subgroups. In the long axial length group (26.00 mm ≤ AL < 28.00 mm), Hill-RBF 3.0 demonstrated the lowest MAE (0.42 D). In the super-long group (28.00 mm ≤ AL < 30.00 mm), the Hoffer QST formula exhibited the best accuracy with the lowest MAE (0.41 D). In the extreme-long axial length group (AL ≥ 30.00 mm), both Hill-RBF 3.0 and HM-ZL formulas shared the lowest MAE (0.65 D). These findings suggest that formula performance varies by axial length and highlight the robustness of Hill-RBF 3.0 and Hoffer QST in challenging cases. 3.3.2 Mean corneal curvature Figure 4 shows the trend of prediction errors as a function of mean corneal curvature (K mean ) and Table 4 presents the prediction errors of each formula in various K mean subgroups. In the flat cornea group (K mean 46.0 D), Hill-RBF 3.0 again showed superior accuracy with the lowest MAE (0.48 D). Boxplot analyses demonstrated that AI-based formulas produced more concentrated prediction errors with lower MedAE values, supporting their reliability and consistency across varying corneal curvatures. 4. Discussion 4.1 Key Findings Overview In cataract surgery for highly myopic eyes, accurate measurement of ocular biometric parameters and appropriate selection of intraocular lens (IOL) power calculation formulas are critical to minimizing postoperative refractive errors. This study retrospectively compared the performance of 11 formulas, including 4 conventional formulas (Barrett Universal II, SRK/T, Haigis, and Holladay 2) and 7 artificial intelligence (AI)-based formulas (Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Ladas, Kane, and HM-ZL). As anticipated, hyperopic prediction errors were common in this cohort, occurring in approximately 59% of cases. Across various accuracy metrics, AI-based formulas outperformed traditional ones, with Hill-RBF 3.0 exhibiting the most consistent and superior performance overall. 4.2 Performance Comparison: AI vs Traditional Formulas Our findings demonstrate that AI-based formulas consistently outperformed traditional methods in terms of predictive accuracy. Among them, the Hill-RBF 3.0 formula achieved the lowest MAE (0.50 D) and MedAE (0.33 D), along with the highest percentage of eyes within ±0.50 D (67.83%) and ±1.00 D (89.57%) of the predicted refraction. These results are in line with prior studies highlighting the robustness of the Hill-RBF algorithm in myopic populations ( 15 , 17 , 22 , 23 ) . Notably, the Hoffer QST and PEARL-DGS formulas also performed well, particularly in subgroup analyses. While Barrett Universal II (BU II) is a traditional formula, it delivered results comparable to the best-performing AI-based formulas, particularly in the ±0.50 D and ±0.75 D categories, with relatively low mean prediction errors. In contrast, the Holladay 2 formula consistently demonstrated the highest prediction errors across all subgroups, underscoring its limitations in highly myopic eyes—an observation corroborated by previous studies such as Miao et al ( 24 ) . 4.3 Axial Length and Corneal Curvature Subgroup Analysis In the long axial length group (26–28 mm), no statistically significant differences were observed in MAE or the percentage within ±0.50 D among the formulas, indicating comparable accuracy between AI-based and traditional formulas in this range. In the super-long axial length group (28–30 mm), Hoffer QST achieved the lowest MAE (0.41), significantly lower than Holladay 2. BU II, Hill-RBF 3.0, Karmona, and Hoffer QST showed the highest percentage within ±0.50 D (70.97%). While the differences in percentages were statistically significant, no statistically significant differences were observed in pairwise comparisons. Overall, Hoffer QST demonstrated higher predictive accuracy in the super-long axial length range (28–30 mm), while other AI-based formulas showed comparable accuracy to traditional formulas. In the extreme-long axial length group (AL ≥ 30 mm), both Hill-RBF 3.0 and the HM-ZL formula achieved the lowest MAE (0.65 D). Several formulas—including BU II, Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Kane, and HM-ZL—showed significantly lower MAEs than Holladay 2. PEARL-DGS had the highest percentage of eyes within ±0.50 D (61.11%). While overall differences among formulas were statistically significant, individual pairwise comparisons were not, likely due to sample size limitations. These results affirm the reliability of AI-based formulas, particularly Hill-RBF 3.0 and HM-ZL, in highly myopic eyes with extreme axial lengths. Regarding corneal curvature, performance also varied: In the flat corneal curvature group (Kmean < 42.0 D), no statistically significant differences in MAE or the percentage within ±0.50 D among the formulas, indicating comparable performance between AI-based and traditional formulas. In the moderate group (42.0–46.0 D), Hill-RBF 3.0 showed the best accuracy (MAE = 0.50 D). MAE values of BU II, Hill-RBF 3.0, Hoffer QST, and PEARL-DGS were significantly lower than Holladay 2. These results indicate higher predictive accuracy of these formulas in the moderate corneal curvature rang. In the steep corneal curvature group (Kmean > 46.0 D), Hill-RBF 3.0 again provided the lowest MAE (0.48 D), followed closely by Hoffer QST, PEARL-DGS, and HM-ZL. The MAE of BU II, Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Kane, and the HM-ZL formula was significantly lower than that of Holladay 2. These formulas demonstrated relatively higher predictive accuracy in the steep corneal curvature range. AI-based formulas consistently demonstrated more centralized prediction error distributions and lower median absolute errors (MedAE) across curvature subgroups, as shown in boxplots. These findings support their robustness and suggest their superiority over traditional formulas in eyes with extreme corneal profiles. 4.4 Clinical Relevance: Benchmarks and Personalized Formula Selection Prediction errors within ±0.50 D and ±1.00 D are clinically significant, as they directly affect postoperative visual quality and patient satisfaction. Gale et al. proposed benchmark standards for refractive outcomes, suggesting that 62.36% of cases should fall within ±0.50 D and 88.76% within ±1.00 D ( 25 ) . In our study, only Hill-RBF 3.0 met both benchmarks, with 67.83% of cases within ±0.50 D and 89.57% within ±1.00 D. Other AI-based and traditional formulas fell short of these thresholds, particularly in highly myopic eyes. These results highlight the ongoing need to optimize IOL power calculation formulas and support the use of AI algorithms for personalized formula selection tailored to individual biometric profiles. 4.5 Comparison with Previous Research Our findings are consistent with prior research. Mo et al. ( 26 ) reported that Hill-RBF 3.0 performed best in eyes with axial lengths >28 mm and remained unaffected by corneal curvature. Hoffer QST excelled in moderate curvature subgroups. These observations align with our data, which further affirm the reliability of these formulas across biometric extremes. Stopyra et al. ( 17 ) demonstrated the superiority of AI-based formulas in Caucasian eyes with AL >30 mm. Our study expands on this by including a broader range of AI models and conducting detailed subgroup analyses in an Asian high myopia cohort. This offers clinicians more comprehensive guidance for formula selection and extends current evidence across populations. 4.6 Limitations of AI-based Formulas Despite their advantages, AI-based formulas have certain limitations. Some models—such as Kane, Hill-RBF 3.0, and Ladas Super Formula—are restricted by upper limits in axial length input (e.g., 35 mm), limiting their use in eyes with extremely long axial lengths. Additionally, Hill-RBF 3.0 supports only target refractions between –2.50 D and +1.00 D, making it unsuitable for patients who prefer residual myopia postoperatively. Yukiko Kora et al. ( 27 ) found that many highly myopic patients prefer a postoperative refraction of –3.00 D or –5.00 D for near vision tasks. Hence, the inability of certain AI formulas to accommodate such targets may restrict their broader applicability in clinical practice. 4.7 Practical Advantages and Future Potential of AI Formulas Beyond prediction accuracy, AI-based formulas offer additional clinical advantages. For example, the Kane, Hill-RBF 3.0, and Hoffer QST formulas incorporate patient gender into calculations, improving prediction precision. PEARL-DGS accommodates complex cases, including post-laser vision correction, radial keratotomy, and ICL implantation. It allows integration of data from the first eye to refine the IOL selection for the second eye ( 13 ) . The Ladas Super Formula offers convenience by enabling users to set default lens models, A-constants, and target refractions, streamlining the calculation process. These features position AI formulas as potential components of clinical decision-support systems. Such tools could accept biometric data, recommend optimal formulas, estimate prediction error ranges, and provide personalized guidance for complex cases. With continued development and data accumulation, their role in precision cataract surgery is expected to grow. 4.8 Study Limitations and Future Directions This study has several limitations. The relatively small sample size, particularly in extreme biometric subgroups, may reduce statistical power. Variability in IOL types and their associated constants may have influenced formula accuracy. Parameters such as lens thickness and central corneal thickness were not consistently available, limiting the applicability of certain formulas (e.g., Kane, PEARL-DGS). Surgical procedures were performed by different surgeons, though previous studies suggest minimal impact on refraction outcomes ( 28 , 29 ) . Furthermore, the study did not control for consistent refractive targets across patients. Future research should focus on multicenter validation of AI formulas, development of integrated clinical tools, and exploration of personalized refractive targets, especially for patients with high myopia who may prefer residual myopia post-surgery. 5. Conclusion Although most AI-based intraocular lens (IOL) power calculation formulas evaluated in this study demonstrated high predictive accuracy, further refinement is warranted through continued research and clinical validation—particularly in the context of highly myopic eyes. Optimizing these formulas holds the potential to enhance postoperative visual outcomes and elevate patient satisfaction. Importantly, formula selection should be personalized based on individual ocular biometric profiles to ensure optimal refractive accuracy. As AI-based technologies continue to evolve, their integration into clinical practice may significantly improve the precision and efficiency of cataract surgery, especially for patients with challenging biometric characteristics. Abbreviations AI : Artificial intelligence IOL: Intraocular lens ME : Mean prediction error SD = Standard deviation MAE : Mean absolute error MedAE : Median absolute prediction error D : Diopter K: Corneal power AL: Axial length BCVA : Best-corrected visual acuity ACD : Anterior chamber depth LT : Lens thickness, WTW : White-to-White LSF : Ladas Super Formula RBF 3.0 : Hill-radial basis function 3.0 BU Ⅱ :Barrett Universal II Declarations Ethics approval and consent to participate This retrospective consecutive case-series study adhered to the Declaration of Helsinki and was approved by the Institutional Review Committee of Affiliated Hospital of Shandong Second Medical University (ID: wyfy-2024-ky-432). Informed consent was obtained from all patients before data collection and analysis. Consent for publication Not applicable. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding None. Authors' contributions X.J. contributed to the literature search and data analysis. J.W. and X.Z. contributed to the data analysis. Q.J. contributed to sample collection and patient management. F.X. contributed to the design of this research ,literature search and revision of the manuscript. M.G. contributed to clinical diagnosis and treatment. Acknowledgements Not applicable. Authors' information Authors and Affiliations Xiaopeng Jiang, Jiangjie Wang, Qingmin Jiang, Xiangyu Zhou, Fei Xia & Meng Gao Department of Ophthalmology, Affiliated Hospital of Shandong Second Medical References Morgan IG, Ohno-Matsui K, Saw S-M. Myopia. Lancet.379(9827):1739-48. Haarman AE, Enthoven CA, Tideman JWL, Tedja MS, Verhoeven VJ, Klaver CC. The complications of myopia: a review and meta-analysis. Investigative ophthalmology & visual science. 2020;61(4):49-. Holden BA, Fricke TR, Wilson DA, Jong M, Naidoo KS, Sankaridurg P, et al. Global prevalence of myopia and high myopia and temporal trends from 2000 through 2050. Ophthalmology. 2016;123(5):1036-42. Curtin BJ, Karlin DB. Axial length measurements and fundus changes of the myopic eye. Am J Ophthalmol.71(1 Pt 1):42-53. Zaldivar R, Shultz MC, Davidorf JM, Holladay JT. Intraocular lens power calculations in patients with extreme myopia. J Cataract Refract Surg.26(5):668-74. Kane JX, Chang DF. Intraocular Lens Power Formulas, Biometry, and Intraoperative Aberrome try: A Review. Ophthalmology.128(11):e94-e114. Olsen T. Calculation of intraocular lens power: a review. Acta Ophthalmol Scand.85(5):472-85. Melles RB, Holladay JT, Chang WJ. Accuracy of intraocular lens calculation formulas. Ophthalmology. 2018;125(2):169-78. Ladas JG, Siddiqui AA, Devgan U, Jun AS. A 3-D "Super Surface" Combining Modern Intraocular Lens Formulas to Ge nerate a "Super Formula" and Maximize Accuracy. JAMA Ophthalmol.133(12):1431-6. Guo D, He W, Wei L, Song Y, Qi J, Yao Y, et al. The Zhu-Lu formula: a machine learning-based intraocular lens power calculation formula for highly myopic eyes. Eye and Vision. 2023;10(1):26. Carmona González D, Palomino Bautista C. Accuracy of a new intraocular lens power calculation method based on a rtificial intelligence. Eye (Lond).35(2):517-22. Stopyra W, Langenbucher A, Grzybowski A. Intraocular lens power calculation formulas—a systematic review. Ophthalmology and Therapy. 2023;12(6):2881-902. Debellemanière G, Dubois M, Gauvin M, Wallerstein A, Brenner LF, Rampat R, et al. The PEARL-DGS Formula: The Development of an Open-source Machine Learn ing-based Thick IOL Calculation Formula. Am J Ophthalmol.232:58-69. Connell BJ, Kane JX. Comparison of the Kane formula with existing formulas for intraocular lens power selection. BMJ Open Ophthalmol.4(1):e000251. Zhou Y, Dai M, Sun L, Tang X, Zhou L, Tang Z, et al. The accuracy of intraocular lens power calculation formulas based on a rtificial intelligence in highly myopic eyes: a systematic review and network meta-analysis. Front Public Health.11:1279718. Li H, Ye Z, Luo Y, Li Z. Comparing the accuracy of the new-generation intraocular lens power ca lculation formulae in axial myopic eyes: a meta-analysis. Int Ophthalmol.43(2):619-33. Stopyra W, Voytsekhivskyy O, Grzybowski A. Accuracy of 7 Artificial Intelligence-Based Intraocular Lens Power Cal culation Formulas in Extremely Long Caucasian Eyes. Am J Ophthalmol.271:337-46. Lin L, Xu M, Mo E, Huang S, Qi X, Gu S, et al. Accuracy of Newer Generation IOL Power Calculation Formulas in Eyes Wi th High Axial Myopia. J Refract Surg.37(11):754-8. Hoffer KJ, Savini G. Update on intraocular lens power calculation study protocols: the better way to design and report clinical trials. Ophthalmology. 2021;128(11):e115-e20. Sharma B, Abell RG, Arora T, Antony T, Vajpayee RB. Techniques of anterior capsulotomy in cataract surgery. Indian J Ophthalmol.67(4):450-60. Charlesworth E, Alderson AJ, de Juan V, Elliott DB. When is refraction stable following routine cataract surgery? A system atic review and meta-analysis. Ophthalmic Physiol Opt.40(5):531-9. Wei L, Cheng K, He W, Zhu X, Lu Y. Application of total keratometry in ten intraocular lens power calcula tion formulas in highly myopic eyes. Eye Vis (Lond).9(1):21. Mo ER, Chen Z, Feng KE, Zhu Z, Xu J, Zhu C, et al. Accuracy of Modern Intraocular Lens Formulas in Highly Myopic Eyes Imp lanted With Plate-Haptic Intraocular Lenses. Am J Ophthalmol.265:105-16. Miao A, Lin P, Ren S, Xu J, Yang F, Qian D, et al. Influence of Ocular Biometry Parameters on the Predictive Accuracy of IOL Power Formulas in Patients with High Myopia. Ophthalmology and therapy.13(1):435-48. Gale RP, Saldana M, Johnston RL, Zuberbuhler B, McKibbin M. Benchmark standards for refractive outcomes after NHS cataract surgery. Eye (Lond).23(1):149-52. Mo E, Feng K, Li Q, Xu J, Cen J, Li J, et al. Efficacy of corneal curvature on the accuracy of 8 intraocular lens po wer calculation formulas in 302 highly myopic eyes. J Cataract Refract Surg.49(12):1195-200. Kora Y, Yaguchi S, Inatomi M, Ozawa T. Preferred postoperative refraction after cataract surgery for high myo pia. J Cataract Refract Surg.21(1):35-8. Wan KH, Lam TC, Marco C, Chan TC. Accuracy and precision of intraocular lens calculations using the new Hill-RBF version 2.0 in eyes with high axial myopia. Am J Ophthalmol. 2019;205:66-73. Reinstein DZ, Yap TE, Carp GI, Archer TJ, Gobbe M. Reproducibility of manifest refraction between surgeons and optometrists in a clinical refractive surgery practice. Journal of Cataract & Refractive Surgery. 2014;40(3):450-9. Hayashi K, Hayashi H. Optimum target refraction for highly and moderately myopic patients af ter monofocal intraocular lens implantation. J Cataract Refract Surg.33(2):240-6. Tables Table 1: Demographic Information and Ocular Biometric Parameters Variable Mean ± SD Range Gender (Male/Female) 74 males (64.3%) / 41 females (35.7%) / Eyes (Right/Left) 63 right (54.8%) / 52 left (45.2%) / Age (years) 60.0±10.8 25-80 Axial Length (mm) 29.0±2.2 26.01-35.00 Anterior Chamber Depth (mm) 3.4±0.4 2.29-4.63 Kmean (D) 44.0±1.9 38.31-49.3 IOL Power (D) 8.0±5.5 -4-18✳ IOL = Intraocular lens; SD = Standard deviation; K mean =Average keratometry ✳ Some highly myopic patients are targeted for a myopic postoperative refraction after cataract surgery ( 30 ) , resulting in a wider range of IOL power requirements. Table 2 Prediction error results for different IOL formulas Formula ME±SD MAE MedAE %±0.25D %±0.50D %±0.75D %±1.00D BUⅡ 0.17±0.77 0.54 0.38 29.57 67.83 80.87 84.35 SRK/T 0.41±0.85 0.64 0.43 35.65 57.39 72.17 84.48 Haigis 0.50±0.81 0.68 0.48 25.22 53.91 65.22 80.87 Holladay 2 0.72±0.88 0.83 0.66 25.22 41.74 53.04 72.17 Hill-RBF 3.0 0.07±0.74 0.50 0.33 37.39 67.83 80.00 89.57 Karmona -0.03±0.76 0.54 0.40 32.17 63.48 78.26 87.83 Hoffer QST 0.12±0.75 0.52 0.37 38.26 66.96 80.00 86.96 PEARL-DGS 0.12±0.73 0.51 0.33 34.78 65.22 78.26 86.96 Ladas 0.33±0.84 0.63 0.42 24.35 56.52 73.91 83.48 Kane -0.06±0.73 0.53 0.38 33.04 61.74 80.00 86.96 HM-ZL formula 0.28±0.72 0.52 0.33 41.74 63.48 75.65 82.61 BU II = Barrett Universal II, RBF = Radial Basis Function, Ladas = Ladas Super Formula, ME = mean prediction error; MAE = mean absolute error; MedAE = median absolute error, MAE = mean absolute error, MedAE = median absolute error, SD = standard deviation, D = diopter. Table 3 Predict error in various axial length subgroups Formula ME SD MAE MedAE %±0.50 D BU Ⅱ 0.11 0.66 0.45 0.32 72.92% SRK/T 0.10 0.68 0.47 0.29 68.75% Haigis 0.25 0.68 0.47 0.30 68.75% Holladay 2 0.31 0.67 0.51 0.38 64.58% Hill-RBF 3.0 0.10 0.65 0.42 0.26 72.92% Karmona 0.16 0.67 0.44 0.31 68.75% Hoffer QST 0.10 0.64 0.43 0.29 75.00% PEARL-DGS 0.07 0.66 0.43 0.29 70.83% Ladas 0.02 0.66 0.45 0.34 68.75% Kane 0.04 0.65 0.44 0.31 66.67% HM-ZL formula 0.34 0.66 0.48 0.30 68.75% BU Ⅱ 0.06 0.60 0.49 0.40 70.97% SRK/T 0.28 0.60 0.49 0.34 64.52% Haigis 0.44 0.65 0.65 0.62 45.16% Holladay 2 0.68 0.64 0.77 0.77 29.03% Hill-RBF 3.0 -0.09 0.54 0.44 0.36 70.97% Karmona -0.16 0.57 0.46 0.37 70.97% Hoffer QST -0.05 0.52 0.41 0.37 70.97% PEARL-DGS 0.03 0.58 0.45 0.36 61.29% Ladas 0.25 0.64 0.54 0.37 64.52% Kane -0.08 0.54 0.43 0.39 67.74% HM-ZL formula 0.14 0.54 0.43 0.33 67.74% BU Ⅱ 0.36 0.98 0.70 0.39 58.33% SRK/T 0.95 1.00 1.01 0.65 36.11% Haigis 0.89 0.95 0.98 0.68 41.67% Holladay 2 1.30 1.00 1.31 1.00 22.22% Hill-RBF 3.0 0.16 0.97 0.65 0.41 58.33% Karmona -0.16 0.95 0.74 0.55 50.00% Hoffer QST 0.28 1.00 0.74 0.44 52.78% PEARL-DGS 0.26 0.92 0.66 0.33 61.11% Ladas 0.81 0.99 0.96 0.63 33.33% Kane -0.16 0.95 0.73 0.53 50.00% HM-ZL formula 0.31 0.89 0.65 0.41 52.78% Table 4 Predict error in various K mean subgroups Formula ME SD MAE MedAE %±0.50 D Kmean＜42 D （n=16） BU Ⅱ 0.13 0.77 0.54 0.36 68.18% SRK/T 0.33 0.83 0.58 0.40 61.62% Haigis 0.45 0.81 0.62 0.39 61.62% Holladay 2 0.65 0.85 0.73 0.54 47.47% Hill-RBF 3.0 0.03 0.74 0.54 0.35 61.62% Karmona -0.06 0.74 0.53 0.39 63.64% Hoffer QST 0.07 0.74 0.51 0.37 69.70% PEARL-DGS 0.09 0.73 0.49 0.30 69.70% Ladas 0.26 0.84 0.57 0.36 62.63% Kane -0.05 0.71 0.53 0.35 61.62% HM-ZL formula 0.24 0.71 0.52 0.34 65.66% 42.0 D ≤ Kmean ≤ 46.0 （n=80） BU Ⅱ 0.17 0.77 0.54 0.38 67.83% SRK/T 0.41 0.85 0.64 0.43 57.39% Haigis 0.50 0.81 0.68 0.48 53.91% Holladay 2 0.72 0.88 0.83 0.66 41.74% Hill-RBF 3.0 0.07 0.74 0.50 0.33 67.83% Karmona -0.03 0.76 0.54 0.40 63.48% Hoffer QST 0.12 0.75 0.52 0.37 66.96% PEARL-DGS 0.12 0.73 0.51 0.33 65.22% Ladas 0.33 0.84 0.63 0.42 56.52% Kane -0.06 0.73 0.53 0.38 61.74% HM-ZL formula 0.28 0.72 0.52 0.33 63.48% Kmean ＞ 46.0 D （n=19） BU Ⅱ 0.11 0.74 0.52 0.37 70.53% SRK/T 0.30 0.80 0.56 0.35 62.11% Haigis 0.42 0.78 0.62 0.43 56.84% Holladay 2 0.61 0.83 0.73 0.54 45.26% Hill-RBF 3.0 0.01 0.73 0.48 0.33 69.47% Karmona -0.07 0.77 0.54 0.42 62.11% Hoffer QST 0.05 0.72 0.49 0.35 70.53% PEARL-DGS 0.07 0.72 0.49 0.31 68.42% Ladas 0.23 0.79 0.57 0.38 61.05% Kane -0.06 0.72 0.51 0.35 63.16% HM-ZL formula 0.22 0.72 0.49 0.32 66.32% Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 23 Sep, 2025 Read the published version in BMC Ophthalmology → Version 1 posted Editorial decision: Revision requested 23 Jun, 2025 Reviews received at journal 19 Jun, 2025 Reviews received at journal 18 Jun, 2025 Reviews received at journal 15 Jun, 2025 Reviewers agreed at journal 13 Jun, 2025 Reviewers agreed at journal 12 Jun, 2025 Reviewers agreed at journal 12 Jun, 2025 Reviewers agreed at journal 12 Jun, 2025 Reviewers invited by journal 12 Jun, 2025 Editor invited by journal 06 Jun, 2025 Editor assigned by journal 03 Jun, 2025 Submission checks completed at journal 03 Jun, 2025 First submitted to journal 29 May, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6777942","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":471485432,"identity":"205158e2-3502-4ab3-a272-3403f6ae792b","order_by":0,"name":"Xiaopeng Jiang","email":"","orcid":"","institution":"Affiliated Hospital of Shandong Second Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiaopeng","middleName":"","lastName":"Jiang","suffix":""},{"id":471485433,"identity":"c4730382-96a7-42bc-b3a1-850e31c8d8af","order_by":1,"name":"Jiangjie Wang","email":"","orcid":"","institution":"Affiliated Hospital of Shandong Second Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jiangjie","middleName":"","lastName":"Wang","suffix":""},{"id":471485434,"identity":"ec31feca-c783-42fd-a8c5-e4a329669efb","order_by":2,"name":"Qingmin Jiang","email":"","orcid":"","institution":"Affiliated Hospital of Shandong Second Medical University","correspondingAuthor":false,"prefix":"","firstName":"Qingmin","middleName":"","lastName":"Jiang","suffix":""},{"id":471485435,"identity":"dda1bab2-fb03-43b2-8958-ec93c3fff3bf","order_by":3,"name":"Xiangyu Zhou","email":"","orcid":"","institution":"Affiliated Hospital of Shandong Second Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiangyu","middleName":"","lastName":"Zhou","suffix":""},{"id":471485437,"identity":"b994ec47-f1fb-4cac-8b02-5f908e53807f","order_by":4,"name":"Fei Xia","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBUlEQVRIiWNgGAWjYBAC9gbGBoYEBhs5+/bmgw8+AEX4GBiY8WrhOQDWkmZswHMs2XAGA4MEG2EtYOpw4gaJHDNpHqK0sDe3STyoOcy4XSLBTNqm4nAdG3vzYQOGGptonFp4DjYbJBxLZ7bseZBsnXPmsAQb0IUJDMfSchtwaLGXSGx8kMBmzcZwPOHg7dw2oBaJHGOgBw/j1MIj/7DhQMI/Zh6GA4kN0pZEaZFgbHyQ2OYsYXAimUmaEaolAa8WnsRmg8S+NAPJnmPMhj1n0iXbgH4B+g63X3jYjz+T/PHNpr6fvf/jgx8V1vz8wBCT+FBjg1MLDpBAmvJRMApGwSgYBWgAAEx0WT3umZSWAAAAAElFTkSuQmCC","orcid":"","institution":"Affiliated Hospital of Shandong Second Medical University","correspondingAuthor":true,"prefix":"","firstName":"Fei","middleName":"","lastName":"Xia","suffix":""},{"id":471485439,"identity":"df79051a-daea-4cef-a8a2-e2b7ce90da3e","order_by":5,"name":"Meng Gao","email":"","orcid":"","institution":"Affiliated Hospital of Shandong Second Medical University","correspondingAuthor":false,"prefix":"","firstName":"Meng","middleName":"","lastName":"Gao","suffix":""}],"badges":[],"createdAt":"2025-05-29 15:38:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6777942/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6777942/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12886-025-04365-5","type":"published","date":"2025-09-23T15:58:24+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":84810324,"identity":"869199ad-1a25-45d2-9720-3b98a3a7ac2c","added_by":"auto","created_at":"2025-06-17 14:47:43","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":63412,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of the absolute prediction error\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6777942/v1/38c9be1d9464c62ce866d8c2.png"},{"id":84810325,"identity":"144b1c75-2f65-4b05-84fe-e30aa4b33674","added_by":"auto","created_at":"2025-06-17 14:47:43","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":77765,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of the absolute prediction error\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6777942/v1/408538965ec37b41dd28ea65.png"},{"id":84811623,"identity":"e0614175-bd44-4943-ae27-3eda545bc265","added_by":"auto","created_at":"2025-06-17 14:55:43","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":98644,"visible":true,"origin":"","legend":"\u003cp\u003eTrend of Mean Prediction Error with Axial Length\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6777942/v1/1ab38cc4f85c1daca2ef319d.png"},{"id":84810331,"identity":"67b3e9a7-512d-402c-978d-4b6750189334","added_by":"auto","created_at":"2025-06-17 14:47:43","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":96336,"visible":true,"origin":"","legend":"\u003cp\u003eTrend plot of mean prediction error with mean corneal curvature\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6777942/v1/a8326d36f4a1d41c9bfc2684.png"},{"id":92431125,"identity":"47c24263-b235-4747-84d2-a83a5f3e6095","added_by":"auto","created_at":"2025-09-29 16:08:32","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1100166,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6777942/v1/94a1bf0b-6a64-46f4-81b4-aee377a3c25b.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Comparative Evaluation of Traditional and AI-Based Intraocular Lens Power Calculation Formulas in Highly Myopic Eyes","fulltext":[{"header":"1. Background","content":"\u003cp\u003eMyopia is a prevalent refractive error that poses a growing public health concern worldwide\u003csup\u003e(1)\u003c/sup\u003e. High myopia, defined as a refractive error of \u0026minus;\u0026thinsp;6.00 diopters (D) or more, not only impairs visual acuity but also predisposes individuals to a range of ocular complications, such as retinal detachment and macular degeneration\u003csup\u003e(2)\u003c/sup\u003e. The global burden of high myopia is expected to reach 938\u0026nbsp;million by 2050, underscoring the urgent need for effective management strategies\u003csup\u003e(3)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHighly myopic eyes frequently present with complex posterior segment alterations, including peripapillary crescent formation, chorioretinal atrophy, and posterior staphyloma\u003csup\u003e(4)\u003c/sup\u003e. These anatomical variations can compromise the precision of axial length measurements, thereby reducing the accuracy of intraocular lens (IOL) power calculations and adversely impacting postoperative refractive outcomes\u003csup\u003e(5)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWith the rising expectations for optimal postoperative vision, accurate prediction of refractive outcomes has become increasingly critical in cataract surgery. The success of modern cataract procedures depends heavily on precise IOL power selection, which directly affects visual acuity and overall patient satisfaction\u003csup\u003e(6)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eConventional IOL power calculation formulas, including SRK/T, Haigis, and Holladay 2, are based on geometric optics and regression modeling\u003csup\u003e(7)\u003c/sup\u003e. Although widely applied in clinical settings, their accuracy diminishes in eyes with extreme biometric values\u0026mdash;such as markedly long or short axial lengths and unusually flat or steep corneal curvatures\u0026mdash;conditions often present in high myopia\u003csup\u003e(8)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eRecently, artificial intelligence (AI) has emerged as a promising innovation in ophthalmology, particularly in enhancing the accuracy of IOL power calculations. AI-based formulas are developed based on different principles. For instance, the Ladas Super Formula (LSF) combines the advantages of multiple existing formulas to create a more comprehensive model\u003csup\u003e(9)\u003c/sup\u003e. Karmona Formula, Hill-RBF 3.0, and the HM-ZL formula purely utilize machine learning algorithms to model the complex relationships between various ocular biometric factors\u003csup\u003e(10\u0026ndash;12)\u003c/sup\u003e. Kane, Pearl-DGS, and Hoffer QST integrate Gaussian optics with AI technology to improve the prediction accuracy of IOL power\u003csup\u003e(12\u0026ndash;14)\u003c/sup\u003e. Notable examples include the Kane, Hill-RBF 3.0, Karmona, and others, which have shown encouraging results in difficult-to-predict eyes such as those with high myopia\u003csup\u003e(15, 16)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003ePrevious studies have systematically compared AI-based intraocular lens formulas with traditional formulas in patients with high myopia, covering a range of axial lengths and corneal curvatures\u003csup\u003e(15, 17)\u003c/sup\u003e. This study further explores the comparison between more AI-based formulas and traditional formulas, offering more accurate formula selection recommendations based on the individualized ocular characteristics of patients.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003e2.1 Ethics Statement\u003c/h2\u003e\n \u003cp\u003eThis retrospective, consecutive case-series study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of the Affiliated Hospital of Shandong Second Medical University (ID: wyfy-2024-ky-432). Written informed consent was obtained from all patients prior to data collection and analysis.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\n \u003ch2\u003e2.2 Patient Selection and Subgrouping:\u003c/h2\u003e\n \u003cp\u003eMedical records of patients with high myopia who underwent cataract surgery between January and December 2024 at the Eye Center of the Affiliated Hospital of Shandong Second Medical University were retrospectively reviewed.\u003c/p\u003e\n \u003cp\u003eInclusion criteria were: age\u0026thinsp;\u0026ge;\u0026thinsp;18 years, a diagnosis of complicated cataract with high myopia (defined as spherical equivalent \u0026ge; \u0026minus;\u0026thinsp;6.0 D or axial length\u0026thinsp;\u0026ge;\u0026thinsp;26.00 mm), complete preoperative and postoperative clinical data, and a postoperative best-corrected visual acuity (BCVA) of at least 20/40.\u003c/p\u003e\n \u003cp\u003eExclusion criteria included other ocular diseases, history of intraocular or refractive surgery, ocular trauma, missing biometric parameters, corneal astigmatism \u0026gt; \u0026minus;\u0026thinsp;3.00 D, or intraoperative complications affecting IOL position.\u003c/p\u003e\n \u003cp\u003eIf both eyes met the inclusion criteria, only one eye per patient was randomly selected using a random number table.\u003c/p\u003e\n \u003cp\u003eBased on prior studies\u003csup\u003e(8, 18)\u003c/sup\u003e, eyes were categorized by axial length (AL) into three subgroups:\u003c/p\u003e\n \u003cp\u003eLong (26.00 mm\u0026thinsp;\u0026le;\u0026thinsp;AL\u0026thinsp;\u0026lt;\u0026thinsp;28.00 mm)\u003c/p\u003e\n \u003cp\u003eSuper-long (28.00 mm\u0026thinsp;\u0026le;\u0026thinsp;AL\u0026thinsp;\u0026lt;\u0026thinsp;30.00 mm)\u003c/p\u003e\n \u003cp\u003eExtreme-long (AL\u0026thinsp;\u0026ge;\u0026thinsp;30.00 mm)\u003c/p\u003e\n \u003cp\u003eCorneal curvature (K\u003csub\u003emean\u003c/sub\u003e) was similarly stratified into:\u003c/p\u003e\n \u003cp\u003eFlat (K\u003csub\u003emean\u003c/sub\u003e \u0026lt; 42.0 D)\u003c/p\u003e\n \u003cp\u003eModerate (42.0 D\u0026thinsp;\u0026le;\u0026thinsp;K\u003csub\u003emean\u003c/sub\u003e \u0026le; 46.0 D)\u003c/p\u003e\n \u003cp\u003eSteep (K\u003csub\u003emean\u003c/sub\u003e \u0026gt; 46.0 D)\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\n \u003ch2\u003e2.3 Biometry and IOL Power Calculation\u003c/h2\u003e\n \u003cp\u003ePreoperative ocular biometry was performed using the IOLMaster\u0026reg; 700 (Carl Zeiss Meditec AG, Jena, Germany), which provided measurements for axial length (AL), keratometry (K), anterior chamber depth (ACD), lens thickness (LT), and white-to-white (WTW) distance. IOL power calculations were performed using 11 formulas: four traditional formulas (SRK/T, Haigis, Holladay 2, and Barrett Universal II) and seven AI-based formulas (Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Ladas Super Formula, Kane, and HM-ZL). A keratometric index of 1.3375 was used. A-constants were obtained from the IOLCon database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ewww.iolcon.org\u003c/span\u003e\u003c/span\u003e) as recommended by Hoffer and Savini\u003csup\u003e(19)\u003c/sup\u003e. All postoperative refractions were measured one month after surgery under mesopic conditions by certified technicians.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\n \u003ch2\u003e2.4 Surgical procedures\u003c/h2\u003e\n \u003cp\u003eAll surgeries were performed by an experienced ophthalmic surgeon using phacoemulsification and intraocular lens (IOL) implantation. After topical anesthesia, a clear corneal incision was created. Continuous curvilinear capsulorhexis (CCC) was performed, with the capsulotomy diameter tailored to match the optic zone of the IOL\u003csup\u003e(20)\u003c/sup\u003e. The crystalline lens was emulsified and aspirated within the capsular bag using an ultrasonic phacoemulsification probe. Subsequently, the IOL was implanted into the capsular bag, and the corneal incision was secured to ensure a watertight seal.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\n \u003ch2\u003e2.5 Statistical Analysis\u003c/h2\u003e\n \u003cp\u003eStatistical analyses were performed using IBM SPSS Statistics for Windows, Version 27.0 (IBM Corp., Armonk, NY, USA). The prediction error (PE) was defined as the difference between the actual postoperative refraction and the predicted value. The following metrics were calculated for each IOL formula: mean prediction error (ME), mean absolute error (MAE), and median absolute error (MedAE).\u003c/p\u003e\n \u003cp\u003eThe one-sample Wilcoxon signed-rank test was used to determine whether the prediction errors significantly deviated from zero. Comparisons of MAEs among the different formulas were conducted using the Friedman test with Bonferroni correction for multiple comparisons. The Bonferroni-adjusted Cochran\u0026apos;s Q test was employed to assess differences in the proportions of eyes falling within predefined prediction error thresholds (\u0026plusmn;\u0026thinsp;0.25 D, \u0026plusmn;\u0026thinsp;0.50 D, \u0026plusmn;\u0026thinsp;0.75 D, and \u0026plusmn;\u0026thinsp;1.00 D). A p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"3. Results","content":"\u003cp\u003e\u003cem\u003e3.1Basic Patient Information\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eA total of 115 eyes from 115 patients were included in the analysis. Demographic characteristics and biometric parameters are summarized in Table 1. The IOL models used included Rayner 920H, Prosert A1UL22, and HOYA XY1-SP. Based on prior literature suggesting refractive stability typically occurs between 2 weeks and 1 month after surgery\u003csup\u003e(\u003c/sup\u003e\u003ca href=\"#_ENREF_19\" title=\"Hoffer, 2021 #11\"\u003e\u003csup\u003e19\u003c/sup\u003e\u003c/a\u003e\u003csup\u003e,\u0026nbsp;\u003c/sup\u003e\u003ca href=\"#_ENREF_21\" title=\"Charlesworth, #54\"\u003e\u003csup\u003e21\u003c/sup\u003e\u003c/a\u003e\u003csup\u003e)\u003c/sup\u003e, we utilized refractive outcomes recorded one month postoperatively for analysis.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e3.2 Refractive prediction error\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTable 2 and Figure 1 summarize the refractive prediction errors across all tested IOL formulas. The Karmona formula yielded the lowest mean error (ME,\u0026nbsp;\u0026ndash;0.03 D), while Hill-RBF 3.0 demonstrated the lowest mean absolute error (MAE, 0.50 D) and median absolute error (MedAE, 0.33 D). The HM-ZL formula achieved the highest proportion of eyes within\u0026nbsp;\u0026plusmn;0.25 D of the predicted value (41.74%).\u003c/p\u003e\n\u003cp\u003eTable 2 and Figure 2 present the percentage of ME for each formula at \u0026plusmn;0.25 D, \u0026plusmn;0.50 D, \u0026plusmn;0.75 D, and \u0026plusmn;1.00 D. Hill-RBF 3.0 and Barrett Universal II (BU II) both had the highest percentages within\u0026nbsp;\u0026plusmn;0.50 D (67.83%), while BU II led in the\u0026nbsp;\u0026plusmn;0.75 D category (80.87%). The Hill-RBF 3.0 formula demonstrated the best performance within\u0026nbsp;\u0026plusmn;1.00 D, achieving 89.57% accuracy. Statistical analysis revealed significant differences among formulas in MAE and in the proportions of eyes falling within\u0026nbsp;\u0026plusmn;0.50 D,\u0026nbsp;\u0026plusmn;0.75 D, and\u0026nbsp;\u0026plusmn;1.00 D prediction error ranges (P \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e3.3 Influence of ocular bioparameters on prediction error\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e3.3.1 Axial Length\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFigure 3 illustrates the distribution of prediction errors across the axial length spectrum (26 mm to 35 mm), and Table 3 presents the prediction errors of each formula in different axial length subgroups. In the long axial length group (26.00 mm\u0026nbsp;\u0026le;\u0026nbsp;AL \u0026lt; 28.00 mm), Hill-RBF 3.0 demonstrated the lowest MAE (0.42 D). In the super-long group (28.00 mm\u0026nbsp;\u0026le;\u0026nbsp;AL \u0026lt; 30.00 mm), the Hoffer QST formula exhibited the best accuracy with the lowest MAE (0.41 D). In the extreme-long axial length group (AL\u0026nbsp;\u0026ge;\u0026nbsp;30.00 mm), both Hill-RBF 3.0 and HM-ZL formulas shared the lowest MAE (0.65 D). These findings suggest that formula performance varies by axial length and highlight the robustness of Hill-RBF 3.0 and Hoffer QST in challenging cases.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e3.3.2 Mean corneal curvature\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFigure 4 shows the trend of prediction errors as a function of mean corneal curvature (K\u003csub\u003emean\u003c/sub\u003e) and Table 4 presents the prediction errors of each formula in various K\u003csub\u003emean\u003c/sub\u003e subgroups. In the flat cornea group (K\u003csub\u003emean\u003c/sub\u003e \u0026lt; 42.0 D), PEARL-DGS yielded the lowest MAE (0.49 D). In the moderate curvature group (42.0 D\u0026nbsp;\u0026le;\u0026nbsp;K\u003csub\u003emean\u003c/sub\u003e \u0026le;\u0026nbsp;46.0 D), Hill-RBF 3.0 performed best with an MAE of 0.50 D. In the steep cornea group (K\u003csub\u003emean\u003c/sub\u003e \u0026gt; 46.0 D), Hill-RBF 3.0 again showed superior accuracy with the lowest MAE (0.48 D). Boxplot analyses demonstrated that AI-based formulas produced more concentrated prediction errors with lower MedAE values, supporting their reliability and consistency across varying corneal curvatures.\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cp\u003e\u003cem\u003e4.1 Key Findings Overview\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn cataract surgery for highly myopic eyes, accurate measurement of ocular biometric parameters and appropriate selection of intraocular lens (IOL) power calculation formulas are critical to minimizing postoperative refractive errors.\u003c/p\u003e\n\u003cp\u003eThis study retrospectively compared the performance of 11 formulas, including 4 conventional formulas (Barrett Universal II, SRK/T, Haigis, and Holladay 2) and 7 artificial intelligence (AI)-based formulas (Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Ladas, Kane, and HM-ZL). As anticipated, hyperopic prediction errors were common in this cohort, occurring in approximately 59% of cases. Across various accuracy metrics, AI-based formulas outperformed traditional ones, with Hill-RBF 3.0 exhibiting the most consistent and superior performance overall.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e4.2 \u003c/em\u003e\u003cem\u003ePerformance Comparison: AI vs Traditional Formulas\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOur findings demonstrate that AI-based formulas consistently outperformed traditional methods in terms of predictive accuracy. Among them, the Hill-RBF 3.0 formula achieved the lowest MAE (0.50 D) and MedAE (0.33 D), along with the highest percentage of eyes within \u0026plusmn;0.50 D (67.83%) and \u0026plusmn;1.00 D (89.57%) of the predicted refraction. These results are in line with prior studies highlighting the robustness of the Hill-RBF algorithm in myopic populations\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e15\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e17\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e22\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e23\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003e. Notably, the Hoffer QST and PEARL-DGS formulas also performed well, particularly in subgroup analyses.\u003c/p\u003e\n\u003cp\u003eWhile Barrett Universal II (BU II) is a traditional formula, it delivered results comparable to the best-performing AI-based formulas, particularly in the \u0026plusmn;0.50 D and \u0026plusmn;0.75 D categories, with relatively low mean prediction errors.\u003c/p\u003e\n\u003cp\u003eIn contrast, the Holladay 2 formula consistently demonstrated the highest prediction errors across all subgroups, underscoring its limitations in highly myopic eyes\u0026mdash;an observation corroborated by previous studies such as Miao et al\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e24\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e4.3 Axial Length and Corneal Curvature Subgroup Analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn the long axial length group (26\u0026ndash;28 mm), no statistically significant differences were observed in MAE or the percentage within \u0026plusmn;0.50 D among the formulas, indicating comparable accuracy between AI-based and traditional formulas in this range.\u003c/p\u003e\n\u003cp\u003eIn the super-long axial length group (28\u0026ndash;30 mm), Hoffer QST achieved the lowest MAE (0.41), significantly lower than Holladay 2. BU II, Hill-RBF 3.0, Karmona, and Hoffer QST showed the highest percentage within \u0026plusmn;0.50 D (70.97%). While the differences in percentages were statistically significant, no statistically significant differences were observed in pairwise comparisons. Overall, Hoffer QST demonstrated higher predictive accuracy in the super-long axial length range (28\u0026ndash;30 mm), while other AI-based formulas showed comparable accuracy to traditional formulas.\u003c/p\u003e\n\u003cp\u003eIn the extreme-long axial length group (AL \u0026ge; 30 mm), both Hill-RBF 3.0 and the HM-ZL formula achieved the lowest MAE (0.65 D). Several formulas\u0026mdash;including BU II, Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Kane, and HM-ZL\u0026mdash;showed significantly lower MAEs than Holladay 2. PEARL-DGS had the highest percentage of eyes within \u0026plusmn;0.50 D (61.11%). While overall differences among formulas were statistically significant, individual pairwise comparisons were not, likely due to sample size limitations. These results affirm the reliability of AI-based formulas, particularly Hill-RBF 3.0 and HM-ZL, in highly myopic eyes with extreme axial lengths.\u003c/p\u003e\n\u003cp\u003eRegarding corneal curvature, performance also varied:\u003c/p\u003e\n\u003cp\u003eIn the flat corneal curvature group (Kmean \u0026lt; 42.0 D), no statistically significant differences in MAE or the percentage within \u0026plusmn;0.50 D among the formulas, indicating comparable performance between AI-based and traditional formulas.\u003c/p\u003e\n\u003cp\u003eIn the moderate group (42.0\u0026ndash;46.0 D), Hill-RBF 3.0 showed the best accuracy (MAE = 0.50 D). MAE values of BU II, Hill-RBF 3.0, Hoffer QST, and PEARL-DGS were significantly lower than Holladay 2. These results indicate higher predictive accuracy of these formulas in the moderate corneal curvature rang.\u003c/p\u003e\n\u003cp\u003eIn the steep corneal curvature group (Kmean \u0026gt; 46.0 D), Hill-RBF 3.0 again provided the lowest MAE (0.48 D), followed closely by Hoffer QST, PEARL-DGS, and HM-ZL. The MAE of BU II, Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Kane, and the HM-ZL formula was significantly lower than that of Holladay 2. These formulas demonstrated relatively higher predictive accuracy in the steep corneal curvature range.\u003c/p\u003e\n\u003cp\u003eAI-based formulas consistently demonstrated more centralized prediction error distributions and lower median absolute errors (MedAE) across curvature subgroups, as shown in boxplots. These findings support their robustness and suggest their superiority over traditional formulas in eyes with extreme corneal profiles.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e4.4 Clinical Relevance: Benchmarks and Personalized Formula Selection\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003ePrediction errors within \u0026plusmn;0.50 D and \u0026plusmn;1.00 D are clinically significant, as they directly affect postoperative visual quality and patient satisfaction. Gale et al. proposed benchmark standards for refractive outcomes, suggesting that 62.36% of cases should fall within \u0026plusmn;0.50 D and 88.76% within \u0026plusmn;1.00 D\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e25\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003e. \u003c/p\u003e\n\u003cp\u003eIn our study, only Hill-RBF 3.0 met both benchmarks, with 67.83% of cases within \u0026plusmn;0.50 D and 89.57% within \u0026plusmn;1.00 D. Other AI-based and traditional formulas fell short of these thresholds, particularly in highly myopic eyes. These results highlight the ongoing need to optimize IOL power calculation formulas and support the use of AI algorithms for personalized formula selection tailored to individual biometric profiles.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e4.5 Comparison with Previous Research\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOur findings are consistent with prior research. Mo et al.\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e26\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003e reported that Hill-RBF 3.0 performed best in eyes with axial lengths \u0026gt;28 mm and remained unaffected by corneal curvature. Hoffer QST excelled in moderate curvature subgroups. These observations align with our data, which further affirm the reliability of these formulas across biometric extremes.\u003c/p\u003e\n\u003cp\u003eStopyra et al.\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e17\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003edemonstrated the superiority of AI-based formulas in Caucasian eyes with AL \u0026gt;30 mm. Our study expands on this by including a broader range of AI models and conducting detailed subgroup analyses in an Asian high myopia cohort. This offers clinicians more comprehensive guidance for formula selection and extends current evidence across populations.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e4.6 Limitations of AI-based Formulas \u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDespite their advantages, AI-based formulas have certain limitations. Some models\u0026mdash;such as Kane, Hill-RBF 3.0, and Ladas Super Formula\u0026mdash;are restricted by upper limits in axial length input (e.g., 35 mm), limiting their use in eyes with extremely long axial lengths. Additionally, Hill-RBF 3.0 supports only target refractions between \u0026ndash;2.50 D and +1.00 D, making it unsuitable for patients who prefer residual myopia postoperatively.\u003c/p\u003e\n\u003cp\u003eYukiko Kora et al.\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e27\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003efound that many highly myopic patients prefer a postoperative refraction of \u0026ndash;3.00 D or \u0026ndash;5.00 D for near vision tasks. Hence, the inability of certain AI formulas to accommodate such targets may restrict their broader applicability in clinical practice.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e4.7 Practical Advantages and Future Potential of AI Formulas\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eBeyond prediction accuracy, AI-based formulas offer additional clinical advantages. For example, the Kane, Hill-RBF 3.0, and Hoffer QST formulas incorporate patient gender into calculations, improving prediction precision. PEARL-DGS accommodates complex cases, including post-laser vision correction, radial keratotomy, and ICL implantation. It allows integration of data from the first eye to refine the IOL selection for the second eye\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e13\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003e. The Ladas Super Formula offers convenience by enabling users to set default lens models, A-constants, and target refractions, streamlining the calculation process.\u003c/p\u003e\n\u003cp\u003eThese features position AI formulas as potential components of clinical decision-support systems. Such tools could accept biometric data, recommend optimal formulas, estimate prediction error ranges, and provide personalized guidance for complex cases. With continued development and data accumulation, their role in precision cataract surgery is expected to grow.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e4.8 Study Limitations and Future Directions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis study has several limitations. The relatively small sample size, particularly in extreme biometric subgroups, may reduce statistical power. Variability in IOL types and their associated constants may have influenced formula accuracy. Parameters such as lens thickness and central corneal thickness were not consistently available, limiting the applicability of certain formulas (e.g., Kane, PEARL-DGS).\u003c/p\u003e\n\u003cp\u003eSurgical procedures were performed by different surgeons, though previous studies suggest minimal impact on refraction outcomes\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e28\u003c/sup\u003e\u003csup\u003e, \u003c/sup\u003e\u003csup\u003e29\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003e. Furthermore, the study did not control for consistent refractive targets across patients.\u003c/p\u003e\n\u003cp\u003eFuture research should focus on multicenter validation of AI formulas, development of integrated clinical tools, and exploration of personalized refractive targets, especially for patients with high myopia who may prefer residual myopia post-surgery.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eAlthough most AI-based intraocular lens (IOL) power calculation formulas evaluated in this study demonstrated high predictive accuracy, further refinement is warranted through continued research and clinical validation\u0026mdash;particularly in the context of highly myopic eyes. Optimizing these formulas holds the potential to enhance postoperative visual outcomes and elevate patient satisfaction.\u003c/p\u003e \u003cp\u003eImportantly, formula selection should be personalized based on individual ocular biometric profiles to ensure optimal refractive accuracy. As AI-based technologies continue to evolve, their integration into clinical practice may significantly improve the precision and efficiency of cataract surgery, especially for patients with challenging biometric characteristics.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAI\u003c/em\u003e\u003c/strong\u003e: Artificial intelligence\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eIOL:\u0026nbsp;\u003c/em\u003e\u003c/strong\u003eIntraocular lens\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eME\u003c/em\u003e\u003c/strong\u003e: Mean prediction error\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eSD =\u0026nbsp;\u003c/em\u003e\u003c/strong\u003eStandard deviation\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eMAE\u003c/em\u003e\u003c/strong\u003e: Mean absolute error\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eMedAE\u003c/em\u003e\u003c/strong\u003e: Median absolute prediction error\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eD\u003c/em\u003e\u003c/strong\u003e: Diopter\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eK:\u003c/em\u003e\u003c/strong\u003eCorneal power\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAL:\u0026nbsp;\u003c/em\u003e\u003c/strong\u003eAxial length\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eBCVA\u003c/em\u003e\u003c/strong\u003e: Best-corrected visual acuity\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eACD\u003c/em\u003e\u003c/strong\u003e: Anterior chamber depth\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eLT\u003c/em\u003e\u003c/strong\u003e: Lens thickness,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eWTW\u003c/em\u003e\u003c/strong\u003e: White-to-White\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eLSF\u003c/em\u003e\u003c/strong\u003e: Ladas Super Formula\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eRBF 3.0\u003c/em\u003e\u003c/strong\u003e: Hill-radial basis function\u0026nbsp;3.0\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eBU\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cem\u003eⅡ\u003c/em\u003e\u003c/strong\u003e:Barrett Universal II\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cem\u003eEthics approval and consent to participate\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis retrospective consecutive case-series study adhered to the Declaration of Helsinki and was approved by the Institutional Review Committee of Affiliated Hospital of Shandong Second Medical University (ID: wyfy-2024-ky-432). Informed consent was obtained from all patients before data collection and analysis.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eConsent for publication\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNot applicable.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAvailability of data and materials\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCompeting interests\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eThe authors declare that they have no competing interests.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eNone.\u003c/p\u003e\n\u003cp\u003eAuthors\u0026apos; contributions\u003c/p\u003e\n\u003cp\u003eX.J. contributed to the literature search and data analysis. J.W. and X.Z. contributed to the data analysis. Q.J. contributed to sample collection and patient management. F.X. contributed to the design of this research ,literature search and revision of the manuscript. \u0026nbsp; M.G. contributed to clinical diagnosis and treatment.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAcknowledgements\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNot applicable.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAuthors\u0026apos; information\u003cem\u003e\u003cbr\u003e\u0026nbsp;\u003c/em\u003eAuthors and Affiliations\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eXiaopeng Jiang, Jiangjie Wang, Qingmin Jiang, Xiangyu Zhou, Fei Xia \u0026amp; Meng Gao\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDepartment of Ophthalmology, Affiliated Hospital of Shandong Second Medical\u003c/em\u003e\u003c/p\u003e"},{"header":" References","content":"\u003col\u003e\n\u003cli\u003eMorgan IG, Ohno-Matsui K, Saw S-M. Myopia. Lancet.379(9827):1739-48.\u003c/li\u003e\n\u003cli\u003eHaarman AE, Enthoven CA, Tideman JWL, Tedja MS, Verhoeven VJ, Klaver CC. The complications of myopia: a review and meta-analysis. Investigative ophthalmology \u0026amp; visual science. 2020;61(4):49-.\u003c/li\u003e\n\u003cli\u003eHolden BA, Fricke TR, Wilson DA, Jong M, Naidoo KS, Sankaridurg P, et al. Global prevalence of myopia and high myopia and temporal trends from 2000 through 2050. Ophthalmology. 2016;123(5):1036-42.\u003c/li\u003e\n\u003cli\u003eCurtin BJ, Karlin DB. Axial length measurements and fundus changes of the myopic eye. Am J Ophthalmol.71(1 Pt 1):42-53.\u003c/li\u003e\n\u003cli\u003eZaldivar R, Shultz MC, Davidorf JM, Holladay JT. Intraocular lens power calculations in patients with extreme myopia. J Cataract Refract Surg.26(5):668-74.\u003c/li\u003e\n\u003cli\u003eKane JX, Chang DF. Intraocular Lens Power Formulas, Biometry, and Intraoperative Aberrome try: A Review. Ophthalmology.128(11):e94-e114.\u003c/li\u003e\n\u003cli\u003eOlsen T. Calculation of intraocular lens power: a review. Acta Ophthalmol Scand.85(5):472-85.\u003c/li\u003e\n\u003cli\u003eMelles RB, Holladay JT, Chang WJ. Accuracy of intraocular lens calculation formulas. Ophthalmology. 2018;125(2):169-78.\u003c/li\u003e\n\u003cli\u003eLadas JG, Siddiqui AA, Devgan U, Jun AS. A 3-D \u0026quot;Super Surface\u0026quot; Combining Modern Intraocular Lens Formulas to Ge nerate a \u0026quot;Super Formula\u0026quot; and Maximize Accuracy. JAMA Ophthalmol.133(12):1431-6.\u003c/li\u003e\n\u003cli\u003eGuo D, He W, Wei L, Song Y, Qi J, Yao Y, et al. The Zhu-Lu formula: a machine learning-based intraocular lens power calculation formula for highly myopic eyes. Eye and Vision. 2023;10(1):26.\u003c/li\u003e\n\u003cli\u003eCarmona Gonz\u0026aacute;lez D, Palomino Bautista C. Accuracy of a new intraocular lens power calculation method based on a rtificial intelligence. Eye (Lond).35(2):517-22.\u003c/li\u003e\n\u003cli\u003eStopyra W, Langenbucher A, Grzybowski A. Intraocular lens power calculation formulas\u0026mdash;a systematic review. Ophthalmology and Therapy. 2023;12(6):2881-902.\u003c/li\u003e\n\u003cli\u003eDebellemani\u0026egrave;re G, Dubois M, Gauvin M, Wallerstein A, Brenner LF, Rampat R, et al. The PEARL-DGS Formula: The Development of an Open-source Machine Learn ing-based Thick IOL Calculation Formula. Am J Ophthalmol.232:58-69.\u003c/li\u003e\n\u003cli\u003eConnell BJ, Kane JX. Comparison of the Kane formula with existing formulas for intraocular lens power selection. BMJ Open Ophthalmol.4(1):e000251.\u003c/li\u003e\n\u003cli\u003eZhou Y, Dai M, Sun L, Tang X, Zhou L, Tang Z, et al. The accuracy of intraocular lens power calculation formulas based on a rtificial intelligence in highly myopic eyes: a systematic review and network meta-analysis. Front Public Health.11:1279718.\u003c/li\u003e\n\u003cli\u003eLi H, Ye Z, Luo Y, Li Z. Comparing the accuracy of the new-generation intraocular lens power ca lculation formulae in axial myopic eyes: a meta-analysis. Int Ophthalmol.43(2):619-33.\u003c/li\u003e\n\u003cli\u003eStopyra W, Voytsekhivskyy O, Grzybowski A. Accuracy of 7 Artificial Intelligence-Based Intraocular Lens Power Cal culation Formulas in Extremely Long Caucasian Eyes. Am J Ophthalmol.271:337-46.\u003c/li\u003e\n\u003cli\u003eLin L, Xu M, Mo E, Huang S, Qi X, Gu S, et al. Accuracy of Newer Generation IOL Power Calculation Formulas in Eyes Wi th High Axial Myopia. J Refract Surg.37(11):754-8.\u003c/li\u003e\n\u003cli\u003eHoffer KJ, Savini G. Update on intraocular lens power calculation study protocols: the better way to design and report clinical trials. Ophthalmology. 2021;128(11):e115-e20.\u003c/li\u003e\n\u003cli\u003eSharma B, Abell RG, Arora T, Antony T, Vajpayee RB. Techniques of anterior capsulotomy in cataract surgery. Indian J Ophthalmol.67(4):450-60.\u003c/li\u003e\n\u003cli\u003eCharlesworth E, Alderson AJ, de Juan V, Elliott DB. When is refraction stable following routine cataract surgery? A system atic review and meta-analysis. Ophthalmic Physiol Opt.40(5):531-9.\u003c/li\u003e\n\u003cli\u003eWei L, Cheng K, He W, Zhu X, Lu Y. Application of total keratometry in ten intraocular lens power calcula tion formulas in highly myopic eyes. Eye Vis (Lond).9(1):21.\u003c/li\u003e\n\u003cli\u003eMo ER, Chen Z, Feng KE, Zhu Z, Xu J, Zhu C, et al. Accuracy of Modern Intraocular Lens Formulas in Highly Myopic Eyes Imp lanted With Plate-Haptic Intraocular Lenses. Am J Ophthalmol.265:105-16.\u003c/li\u003e\n\u003cli\u003eMiao A, Lin P, Ren S, Xu J, Yang F, Qian D, et al. Influence of Ocular Biometry Parameters on the Predictive Accuracy of IOL Power Formulas in Patients with High Myopia. Ophthalmology and therapy.13(1):435-48.\u003c/li\u003e\n\u003cli\u003eGale RP, Saldana M, Johnston RL, Zuberbuhler B, McKibbin M. Benchmark standards for refractive outcomes after NHS cataract surgery. Eye (Lond).23(1):149-52.\u003c/li\u003e\n\u003cli\u003eMo E, Feng K, Li Q, Xu J, Cen J, Li J, et al. Efficacy of corneal curvature on the accuracy of 8 intraocular lens po wer calculation formulas in 302 highly myopic eyes. J Cataract Refract Surg.49(12):1195-200.\u003c/li\u003e\n\u003cli\u003eKora Y, Yaguchi S, Inatomi M, Ozawa T. Preferred postoperative refraction after cataract surgery for high myo pia. J Cataract Refract Surg.21(1):35-8.\u003c/li\u003e\n\u003cli\u003eWan KH, Lam TC, Marco C, Chan TC. Accuracy and precision of intraocular lens calculations using the new Hill-RBF version 2.0 in eyes with high axial myopia. Am J Ophthalmol. 2019;205:66-73.\u003c/li\u003e\n\u003cli\u003eReinstein DZ, Yap TE, Carp GI, Archer TJ, Gobbe M. Reproducibility of manifest refraction between surgeons and optometrists in a clinical refractive surgery practice. Journal of Cataract \u0026amp; Refractive Surgery. 2014;40(3):450-9.\u003c/li\u003e\n\u003cli\u003eHayashi K, Hayashi H. Optimum target refraction for highly and moderately myopic patients af ter monofocal intraocular lens implantation. J Cataract Refract Surg.33(2):240-6.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"671\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"bottom\" style=\"width: 498px;\"\u003e\n \u003cp\u003eTable 1: Demographic Information and Ocular Biometric Parameters\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 237px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 261px;\"\u003e\n \u003cp\u003eMean\u0026nbsp;\u0026plusmn;\u0026nbsp;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003eRange\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 237px;\"\u003e\n \u003cp\u003eGender (Male/Female)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 261px;\"\u003e\n \u003cp\u003e74 males (64.3%) / 41 females (35.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003e/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 237px;\"\u003e\n \u003cp\u003eEyes (Right/Left)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 261px;\"\u003e\n \u003cp\u003e63 right (54.8%) / 52 left (45.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003e/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 237px;\"\u003e\n \u003cp\u003eAge (years)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 261px;\"\u003e\n \u003cp\u003e60.0\u0026plusmn;10.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003e25-80\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 237px;\"\u003e\n \u003cp\u003eAxial Length (mm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 261px;\"\u003e\n \u003cp\u003e29.0\u0026plusmn;2.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003e26.01-35.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 237px;\"\u003e\n \u003cp\u003eAnterior Chamber Depth (mm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 261px;\"\u003e\n \u003cp\u003e3.4\u0026plusmn;0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003e2.29-4.63\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 237px;\"\u003e\n \u003cp\u003eKmean (D)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 261px;\"\u003e\n \u003cp\u003e44.0\u0026plusmn;1.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003e38.31-49.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 237px;\"\u003e\n \u003cp\u003eIOL Power (D)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 261px;\"\u003e\n \u003cp\u003e8.0\u0026plusmn;5.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 173px;\"\u003e\n \u003cp\u003e-4-18✳\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"bottom\" style=\"width: 671px;\"\u003e\n \u003cp\u003eIOL = Intraocular lens; SD = Standard deviation; K\u003csub\u003emean\u003c/sub\u003e =Average keratometry\u003c/p\u003e\n \u003cp\u003e✳\u0026nbsp;Some highly myopic patients are targeted for a myopic postoperative refraction after cataract surgery\u003csup\u003e(\u003c/sup\u003e\u003csup\u003e30\u003c/sup\u003e\u003csup\u003e)\u003c/sup\u003e, resulting in a wider range of IOL power requirements.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"675\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"bottom\" style=\"width: 403px;\"\u003e\n \u003cp\u003eTable 2 Prediction error results for different IOL formulas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eFormula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003eME\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003eMAE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003eMedAE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e%\u0026plusmn;0.25D\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e%\u0026plusmn;0.50D\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e%\u0026plusmn;0.75D\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e%\u0026plusmn;1.00D\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eBUⅡ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.17\u0026plusmn;0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e29.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e67.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e80.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e84.35\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eSRK/T\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.41\u0026plusmn;0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e35.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e57.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e72.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e84.48\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eHaigis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.50\u0026plusmn;0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e25.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e53.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e65.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e80.87\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eHolladay 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.72\u0026plusmn;0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e25.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e41.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e53.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e72.17\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eHill-RBF 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.07\u0026plusmn;0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e37.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e67.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e80.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e89.57\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eKarmona\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.03\u0026plusmn;0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e32.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e63.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e78.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e87.83\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eHoffer QST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.12\u0026plusmn;0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e38.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e66.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e80.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e86.96\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003ePEARL-DGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.12\u0026plusmn;0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e34.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e65.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e78.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e86.96\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eLadas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.33\u0026plusmn;0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e24.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e56.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e73.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e83.48\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eKane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.06\u0026plusmn;0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e33.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e61.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e80.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e86.96\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 119px;\"\u003e\n \u003cp\u003eHM-ZL formula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 98px;\"\u003e\n \u003cp\u003e0.28\u0026plusmn;0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 46px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 65px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e41.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e63.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e75.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e82.61\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"8\" valign=\"bottom\" style=\"width: 675px;\"\u003e\n \u003cp\u003eBU II = Barrett Universal II, RBF = Radial Basis Function, Ladas = Ladas Super Formula, ME = mean prediction error; MAE = mean absolute error; MedAE = median absolute error, MAE = mean absolute error, MedAE = median absolute error, SD = standard deviation, D = diopter.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 3 Predict error in various axial length subgroups \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"674\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eFormula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003eME\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e\u0026nbsp;MAE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003eMedAE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e%\u0026plusmn;0.50 D\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eBU\u0026nbsp;Ⅱ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e72.92%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eSRK/T\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e68.75%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHaigis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e68.75%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHolladay 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e64.58%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHill-RBF 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e72.92%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eKarmona\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e68.75%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHoffer QST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e75.00%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003ePEARL-DGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e70.83%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eLadas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e68.75%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eKane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e66.67%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHM-ZL formula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e68.75%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eBU\u0026nbsp;Ⅱ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e70.97%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eSRK/T\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e64.52%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHaigis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e45.16%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHolladay 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e29.03%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHill-RBF 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e-0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e70.97%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eKarmona\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e-0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e70.97%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHoffer QST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e-0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e70.97%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003ePEARL-DGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e61.29%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eLadas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e64.52%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eKane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e-0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e67.74%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHM-ZL formula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e67.74%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eBU\u0026nbsp;Ⅱ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e58.33%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eSRK/T\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e1.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e36.11%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHaigis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e41.67%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHolladay 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e1.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e1.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e22.22%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHill-RBF 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e58.33%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eKarmona\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e-0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e50.00%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHoffer QST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e52.78%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003ePEARL-DGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e61.11%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eLadas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e33.33%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eKane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e-0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e50.00%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 163px;\"\u003e\n \u003cp\u003eHM-ZL formula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 97px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e52.78%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 4 Predict error in various K\u003csub\u003emean\u003c/sub\u003e subgroups\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"671\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eFormula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003eME\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;MAE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003eMedAE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e%\u0026plusmn;0.50 D\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"bottom\" style=\"width: 671px;\"\u003e\n \u003cp\u003eKmean＜42 D \u0026nbsp; \u0026nbsp;（n=16）\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eBU\u0026nbsp;Ⅱ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e68.18%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eSRK/T\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e61.62%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHaigis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e61.62%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHolladay 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e47.47%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHill-RBF 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e61.62%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eKarmona\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e-0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e63.64%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHoffer QST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e69.70%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003ePEARL-DGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e69.70%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eLadas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e62.63%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eKane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e-0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e61.62%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHM-ZL formula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e65.66%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"bottom\" style=\"width: 671px;\"\u003e\n \u003cp\u003e42.0 D\u0026nbsp;\u0026le;\u0026nbsp;Kmean\u0026nbsp;\u0026le;\u0026nbsp;46.0\u0026nbsp;（n=80）\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eBU\u0026nbsp;Ⅱ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e67.83%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eSRK/T\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e57.39%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHaigis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e53.91%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHolladay 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e41.74%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHill-RBF 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e67.83%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eKarmona\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e-0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e63.48%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHoffer QST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e66.96%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003ePEARL-DGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e65.22%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eLadas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e56.52%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eKane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e-0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e61.74%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHM-ZL formula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e63.48%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"bottom\" style=\"width: 671px;\"\u003e\n \u003cp\u003eKmean\u0026nbsp;＞\u0026nbsp;46.0 D \u0026nbsp;（n=19）\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eBU\u0026nbsp;Ⅱ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e70.53%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eSRK/T\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e62.11%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHaigis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e56.84%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHolladay 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e45.26%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHill-RBF 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e69.47%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eKarmona\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e-0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e62.11%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHoffer QST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e70.53%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003ePEARL-DGS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e68.42%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eLadas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e61.05%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eKane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e-0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e63.16%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHM-ZL formula\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e66.32%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-ophthalmology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"boph","sideBox":"Learn more about [BMC Ophthalmology](http://bmcophthalmol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/boph","title":"BMC Ophthalmology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"artificial intelligence, intraocular lens, high myopia, IOL power formula, prediction accuracy","lastPublishedDoi":"10.21203/rs.3.rs-6777942/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6777942/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose: \u003c/strong\u003eTo assess the accuracy of artificial intelligence (AI)-based intraocular lens (IOL) power calculation formulas compared with traditional methods in highly myopic eyes, and to evaluate their performance across varying axial lengths and corneal curvatures.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e This retrospective case series included 115 highly myopic eyes that underwent phacoemulsification with IOL implantation. IOL power was calculated using four conventional formulas (SRK/T, Haigis, Holladay 2, Barrett Universal II) and seven AI-based formulas (Hill-RBF 3.0, Karmona, Hoffer QST, PEARL-DGS, Ladas Super Formula, Kane, HM-ZL). The outcomes were evaluated using mean error (ME), mean absolute error (MAE), median absolute error (MedAE), and the percentage of eyes within ±0.25 D to ±1.00 D of the prediction error. Subgroup analyses were conducted based on axial length (AL) and corneal curvature (Kmean).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eAI-based formulas—especially Hill-RBF 3.0, Hoffer QST, and PEARL-DGS—demonstrated significantly higher accuracy than traditional formulas. Hill-RBF 3.0 achieved the lowest MAE (0.50 D) and MedAE (0.33 D) and the highest percentage of eyes within ±0.50 D (67.83%）and ±1.00 D (89.57%). Subgroup analyses showed that AI formulas maintained consistent performance across various AL and Kmean categories. Significant differences were noted between AI-based and traditional formulas, particularly in eyes with extreme biometric values.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eAI-based formulas provide superior refractive prediction in highly myopic eyes compared with traditional methods, particularly in cases of long axial length or steep corneal curvature. Tailored formula selection based on biometric profiles may enhance refractive outcomes in cataract surgery.\u003c/p\u003e","manuscriptTitle":"Comparative Evaluation of Traditional and AI-Based Intraocular Lens Power Calculation Formulas in Highly Myopic Eyes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-17 14:47:38","doi":"10.21203/rs.3.rs-6777942/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-06-23T06:36:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-19T23:17:18+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-18T10:59:39+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-15T13:03:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"322790147501019059777864036636443538107","date":"2025-06-13T12:45:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"25048441729562906721801817967608603832","date":"2025-06-12T23:31:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"191316920954679653364839524261759398540","date":"2025-06-12T21:15:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"6492601109820152603308236342103720702","date":"2025-06-12T20:23:47+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-06-12T18:04:01+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-06-06T12:53:41+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-04T03:02:32+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-06-04T02:59:45+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Ophthalmology","date":"2025-05-29T15:23:51+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-ophthalmology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"boph","sideBox":"Learn more about [BMC Ophthalmology](http://bmcophthalmol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/boph","title":"BMC Ophthalmology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"880ea1a6-3a9c-483a-881e-69797cd68e3a","owner":[],"postedDate":"June 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-09-29T16:08:03+00:00","versionOfRecord":{"articleIdentity":"rs-6777942","link":"https://doi.org/10.1186/s12886-025-04365-5","journal":{"identity":"bmc-ophthalmology","isVorOnly":false,"title":"BMC Ophthalmology"},"publishedOn":"2025-09-23 15:58:24","publishedOnDateReadable":"September 23rd, 2025"},"versionCreatedAt":"2025-06-17 14:47:38","video":"","vorDoi":"10.1186/s12886-025-04365-5","vorDoiUrl":"https://doi.org/10.1186/s12886-025-04365-5","workflowStages":[]},"version":"v1","identity":"rs-6777942","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6777942","identity":"rs-6777942","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}