Machine-learning technology for predicting intraocular lens power: Diagnostic data generalization

Abstract

BACKGROUND: The implantation of recent intraocular lens (IOLs) allows ophthalmologists to effectively solve the surgical rehabilitation problems of patients with cataracts. The degree of improvement in the patient’s visual function is directly dependent on the accuracy of the preoperative calculation of the optical IOL power. The most famous formulas used to calculate this indicator include SRK II, SRK/T, Hoffer-Q, Holladay II, Haigis, and Barrett. All these work well for an “average patient”; however, they are not adequate at the boundaries of input variable ranges.

AIM: To examine the possibility of using mathematical models obtained by deep learning of artificial neural network (ANN) models to generalize data and predict the optical power of modern IOLs.

MATERIALS AND METHODS: ANN models were trained on large-scale samples, including depersonalized data for patients in the ophthalmology clinic. Data provided in 2021 by ophthalmologist K.K. Syrykh reflect the results of both preoperative and postoperative observations of patients. The source file used to build the ANN model included 455 records (26 columns of input factors and one column for the output factor) for calculating IOL (diopters). To conveniently build ANN models, a simulator program previously developed by the authors was used.

RESULTS: The resulting models, in contrast to the traditionally used formulas, reflect the regional specificity of patients to a much greater extent. They also make it possible to retrain and optimize the structure based on newly received data, which allows us to consider the nonstationarity of objects. A distinctive feature of such ANN models in comparison with the well-known formulas SRK II, SRK/T, Hoffer-Q, Holladay II, Haigis, and Barrett, which are widely used in surgical cataract treatment, is their ability to consider a significant number of recorded input quantities, which reduces the mean relative error in calculating the optical IOL power from 10%–12% to 3.5%.

CONCLUSION: This study reveals the fundamental possibility of generalizing a significant amount of empirical data on calculating the optical IOL power using training ANN models that have a significantly larger number of input variables than those obtained using traditional formulas and methods. The results obtained allow the construction of an intelligent expert system with a continuous flow of new data from a source and a step-by-step retraining of ANN models.

Keywords

artificial intelligence / medical data / dataset / machine learning / intraocular lenses

References

[1] Fyodorov SN, Kolinko AI. Method of calculating the optical power of an intraocular lens. The Russian Annals of Ophthalmology. 1967;(4):27–31. (In Russ).

[2] Фёдоров С.Н., Колинко А.И. Методика расчета оптической силы интраокулярной линзы // Вестник офтальмологии. 1967. № 4. С. 27–31.

[3] Fyodorov SN, Kolinko AI. Method of calculating the optical power of an intraocular lens. The Russian Annals of Ophthalmology. 1967;(4):27–31. (In Russ).

[4] Balashevich LI, Danilenko EV. Results in application of the fyodorov’s iol power formula for posterior chamber lenses calculation. Fyodorov Journal of Ophthalmic Surgery. 2011;(1):34–38. EDN: PXRASV

[5] Балашевич Л.И., Даниленко Е.В. Результаты использования формулы С.Н. Фёдорова для расчёта силы заднекамерных интраокулярных линз // Офтальмохирургия. 2011. № 1. С. 34–38. EDN: PXRASV

[6] Balashevich LI, Danilenko EV. Results in application of the fyodorov’s iol power formula for posterior chamber lenses calculation. Fyodorov Journal of Ophthalmic Surgery. 2011;(1):34–38. EDN: PXRASV

[7] Sanders DR, Kraff MC. Improvement of intraocular lens power calculation using empirical data. American Intra-Ocular Implant Society Journal. 1980;6:263–267. doi: 10.1016/s0146-2776(80)80075-9

[8] Sanders D.R., Kraff M.C. Improvement of intraocular lens power calculation using empirical data // American Intra-Ocular Implant Society Journal. 1980. Vol. 6. P. 263–267. doi: 10.1016/s0146-2776(80)80075-9

[9] Sanders DR, Kraff MC. Improvement of intraocular lens power calculation using empirical data. American Intra-Ocular Implant Society Journal. 1980;6:263–267. doi: 10.1016/s0146-2776(80)80075-9

[10] Sanders DR, Retzlaff JA, Kraff MC. Comparison of the SRK II formula and other second-generation formulas. Journal of Cataract & Refractive Surgery. 1988;14(2):136–141. doi: 10.1016/s0886-3350(88)80087-7

[11] Sanders D.R., Retzlaff J.A., Kraff M.C. Comparison of the SRK II formula and other second-generation formulas // Journal of Cataract & Refractive Surgery. 1988. Vol. 14, N 2. P. 136–141. doi: 10.1016/s0886-3350(88)80087-7

[12] Sanders DR, Retzlaff JA, Kraff MC. Comparison of the SRK II formula and other second-generation formulas. Journal of Cataract & Refractive Surgery. 1988;14(2):136–141. doi: 10.1016/s0886-3350(88)80087-7

[13] Sanders DR, Retzlaff JA, Kraff MC. Development of the SRK/T IOL power calculation formula. Journal of Cataract & Refractive Surgery. 1990;16(3):333–340. doi: 10.1016/s0886-3350(13)80705-5

[14] Sanders D.R., Retzlaff J.A., Kraff M.C. Development of the SRK/T IOL power calculation formula // Journal of Cataract & Refractive Surgery. 1990. Vol. 16, N 3. P. 333–340. doi: 10.1016/s0886-3350(13)80705-5

[15] Sanders DR, Retzlaff JA, Kraff MC. Development of the SRK/T IOL power calculation formula. Journal of Cataract & Refractive Surgery. 1990;16(3):333–340. doi: 10.1016/s0886-3350(13)80705-5

[16] Hoffer KJ. The Hoffer Q formula: a comparison of theoretic and regression formulas. Journal of Cataract & Refractive Surgery. 1993;19(6):700–712. doi: 10.1016/s0886-3350(13)80338-0

[17] Hoffer K.J. The Hoffer Q formula: a comparison of theoretic and regression formulas // Journal of Cataract & Refractive Surgery. 1993. Vol. 19, N 6. P. 700–712. doi: 10.1016/s0886-3350(13)80338-0

[18] Hoffer KJ. The Hoffer Q formula: a comparison of theoretic and regression formulas. Journal of Cataract & Refractive Surgery. 1993;19(6):700–712. doi: 10.1016/s0886-3350(13)80338-0

[19] Holladay JT, Prager TC, Ruiz RS, et al. A three-part system for refining intraocular lens power calculation. Journal of Cataract & Refractive Surgery. 1988;14(1):17–24. doi: 10.1016/S0886-3350(88)80059-2

[20] Holladay J.T., Prager T.C., Ruiz R.S., et al. A three-part system for refining intraocular lens power calculation // Journal of Cataract & Refractive Surgery. 1988. Vol. 14, N 1. P. 17–24. doi: 10.1016/S0886-3350(88)80059-2

[21] Holladay JT, Prager TC, Ruiz RS, et al. A three-part system for refining intraocular lens power calculation. Journal of Cataract & Refractive Surgery. 1988;14(1):17–24. doi: 10.1016/S0886-3350(88)80059-2

[22] Pershin KB, Pashinova NF, Tsygankov AYu, Legkhih SL. Choice of IOL Optic Power Calculation Formula in Extremely High Myopia Patients “Excimer” Ophthalmology Centre, Moscow. Point of view. East - West. 2016;(1):64–67. EDN: WHCNPF

[23] Першин К.Б., Пашинова Н.Ф., Цыганков А.Ю., Легких С.Л. Алгоритм выбора формулы для расчета оптической силы ИОЛ при экстремальной миопии // Точка зрения. Восток - Запад. 2016. № 1. C. 64–67. EDN: WHCNPF

[24] Pershin KB, Pashinova NF, Tsygankov AYu, Legkhih SL. Choice of IOL Optic Power Calculation Formula in Extremely High Myopia Patients “Excimer” Ophthalmology Centre, Moscow. Point of view. East - West. 2016;(1):64–67. EDN: WHCNPF

[25] Buduma N, Lokasho N. Foundations of deep learning. Creating Algorithms for Next Generation Artificial Intelligence. Moscow: Mann, Ivanov i Ferber; 2020. (In Russ).

[26] Будума Н., Локашо Н. Основы глубокого обучения. Создание алгоритмов для искусственного интеллекта следующего поколения. Москва : Манн, Иванов и Фербер, 2020.

[27] Buduma N, Lokasho N. Foundations of deep learning. Creating Algorithms for Next Generation Artificial Intelligence. Moscow: Mann, Ivanov i Ferber; 2020. (In Russ).

[28] Foster D. Generative deep learning. Creative potential of neural networks. Saint Petersburg: Piter; 2020. (In Russ).

[29] Фостер Д. Генеративное глубокое обучение. Творческий потенциал нейронных сетей. Санкт-Петербург : Питер, 2020.

[30] Foster D. Generative deep learning. Creative potential of neural networks. Saint Petersburg: Piter; 2020. (In Russ).

[31] Ramsundar B, Istman P, Uolters P, Pande V. Deep learning in biology and medicine. Moscow: DMK Press; 2020. (In Russ).

[32] Рамсундар Б., Истман П., Уолтерс П., Панде В. Глубокое обучение в биологии и медицине. Москва : ДМК Пресс, 2020.

[33] Ramsundar B, Istman P, Uolters P, Pande V. Deep learning in biology and medicine. Moscow: DMK Press; 2020. (In Russ).

[34] Kharrison M. Machine learning: a pocket guide. A quick guide to structured machine learning methods in Python. Saint Petersburg: Dialektika LLC; 2020. (In Russ).

[35] Харрисон М. Машинное обучение: карманный справочник. Краткое руководство по методам структурированного машинного обучения на Python. Санкт-Петербург : ООО «Диалектика», 2020.

[36] Kharrison M. Machine learning: a pocket guide. A quick guide to structured machine learning methods in Python. Saint Petersburg: Dialektika LLC; 2020. (In Russ).

[37] Arzamastsev AA, Fabrikantov OL, Zenkova NA, Belousov NK. Optimization of Formulae for Intraocular Lenses Calculating. Tambov University Reports. Series: Natural and Technical Sciences. 2016;21(1):208–213. EDN: VNWHVZ doi: 10.20310/1810-0198-2016-21-1-208-213

[38] Арзамасцев А.А., Фабрикантов О.Л., Зенкова Н.А., Белоусов Н.К. Оптимизация формул для расчета ИОЛ // Вестник Тамбовского университета. Серия Естественные и технические науки. 2016. Т. 21, № 1. С. 208–213. EDN: VNWHVZ doi: 10.20310/1810-0198-2016-21-1-208-213

[39] Arzamastsev AA, Fabrikantov OL, Zenkova NA, Belousov NK. Optimization of Formulae for Intraocular Lenses Calculating. Tambov University Reports. Series: Natural and Technical Sciences. 2016;21(1):208–213. EDN: VNWHVZ doi: 10.20310/1810-0198-2016-21-1-208-213

[40] Yamauchi T, Tabuchi T, Takase K, Masumoto H. Use of a machine learning method in predicting refraction after cataract surgery. Journal of Clinical Medicine. 2021;10(5):1103. doi: 10.3390/jcm10051103

[41] Yamauchi T., Tabuchi T., Takase K., Masumoto H. Use of a machine learning method in predicting refraction after cataract surgery // Journal of Clinical Medicine. 2021. Vol. 10, N 5. P. 1103. doi: 10.3390/jcm10051103

[42] Yamauchi T, Tabuchi T, Takase K, Masumoto H. Use of a machine learning method in predicting refraction after cataract surgery. Journal of Clinical Medicine. 2021;10(5):1103. doi: 10.3390/jcm10051103

[43] Certificate of state registration of the computer program № 2012618141/ 07.09.2012. Arzamastsev AA, Rykov VP, Kryuchin OV. Artificial neural network simulator with implementation of modular learning principle. (In Russ).

[44] Свидетельство о государственной регистрации программы для ЭВМ № 2012618141/ 07.09.2012. Арзамасцев А.А., Рыков В.П., Крючин О.В. Симулятор искусственной нейронной сети с реализацией модульного принципа обучения.

[45] Certificate of state registration of the computer program № 2012618141/ 07.09.2012. Arzamastsev AA, Rykov VP, Kryuchin OV. Artificial neural network simulator with implementation of modular learning principle. (In Russ).

[46] Kolmogorov AN. On the representation of continuous functions of several variables by superpositions of continuous functions of fewer variables. Doklady Akademii nauk SSSR. 1956;108(2):179–182. (In Russ).

[47] Колмогоров А.Н. О представлении непрерывных функций нескольких переменных суперпозициями непрерывных функций меньшего числа переменных // Доклады Академии наук СССР. 1956. Т. 108, № 2. С. 179–182.

[48] Kolmogorov AN. On the representation of continuous functions of several variables by superpositions of continuous functions of fewer variables. Doklady Akademii nauk SSSR. 1956;108(2):179–182. (In Russ).

[49] Kolmogorov AN. On the representation of continuous functions of several variables as a superposition of continuous functions of one variable. Doklady Akademii nauk SSSR. 1957;114(5):953–956. (In Russ).

[50] Колмогоров А.Н. О представлении непрерывных функций нескольких переменных в виде суперпозиции непрерывных функций одного переменного // Доклады Академии наук СССР. 1957. Т. 114, № 5. С. 953–956.

[51] Kolmogorov AN. On the representation of continuous functions of several variables as a superposition of continuous functions of one variable. Doklady Akademii nauk SSSR. 1957;114(5):953–956. (In Russ).

[52] Arzamaszev AA, Kryuchin OV, Azarova PA, Zenkova NA. The universal program complex for computer simulation on the basis of the artificial neuron network with self-organizing structure. Tambov University Reports. Series: Natural and Technical Sciences. 2006;11(4):564–570. EDN: IRMPYX

[53] Арзамасцев А.А., Крючин О.В., Азарова П.А., Зенкова Н.А. Универсальный программный комплекс для компьютерного моделирования на основе искусственной нейронной сети с самоорганизацией структуры // Вестник Тамбовского университета. Серия: Естественные и технические науки. 2006. Т. 11, № 4. C. 564–570. EDN: IRMPYX

[54] Arzamaszev AA, Kryuchin OV, Azarova PA, Zenkova NA. The universal program complex for computer simulation on the basis of the artificial neuron network with self-organizing structure. Tambov University Reports. Series: Natural and Technical Sciences. 2006;11(4):564–570. EDN: IRMPYX

[55] Arzamastsev AA, Zenkova NA, Kazakov NA. Algorithms and methods for extracting knowledge about objects defined by arrays of empirical data using ANN models. Journal of Physics: Conference Series. 2021. doi: 10.1088/1742-6596/1902/1/012097

[56] Arzamastsev A.A., Zenkova N.A., Kazakov N.A. Algorithms and methods for extracting knowledge about objects defined by arrays of empirical data using ANN models // Journal of Physics: Conference Series. 2021. doi: 10.1088/1742-6596/1902/1/012097

[57] Arzamastsev AA, Zenkova NA, Kazakov NA. Algorithms and methods for extracting knowledge about objects defined by arrays of empirical data using ANN models. Journal of Physics: Conference Series. 2021. doi: 10.1088/1742-6596/1902/1/012097

RIGHTS & PERMISSIONS

Eco-Vector