Using Kalman Filtering to Forecast Disease Trajectory for Patients With Normal Tension Glaucoma

Overview

Journal Am J Ophthalmol

Specialty Ophthalmology

Date 2018 Oct 19

PMID 30336130

Citations 11

Authors

Gian-Gabriel P Garcia

Koji Nitta

Mariel S Lavieri

Chris Andrews

Xiang Liu

Elizabeth Lobaza

Mark P Van Oyen

Kazuhisa Sugiyama

Joshua D Stein

Affiliations

Soon will be listed here.

Abstract

Purpose: To determine whether a machine learning technique called Kalman filtering (KF) can accurately forecast future values of mean deviation (MD), pattern standard deviation, and intraocular pressure for patients with normal tension glaucoma (NTG).

Design: Development and testing of a forecasting model for glaucoma progression.

Methods: We parameterized and validated a KF (KF-NTG) to forecast MD, pattern standard deviation, and intraocular pressure at 24 months into the future using 263 eyes of 263 Japanese patients with NTG. We determined the proportion of patients with MD forecasts within 0.5, 1.0, and 2.5 dBs of the actual values and calculated the root mean squared error (RMSE) for each forecast. We compared KF-NTG with a previously published KF model calibrated using patients with high-tension open-angle glaucoma (KF-HTG) and to 3 conventional forecasting algorithms.

Results: The 263 patients with NTG had mean ± standard deviation age of 63.4 ± 10.5 years. KF-NTG forecasted MD values 24 months ahead within 0.5, 1.0, and 2.5 dBs of the actual value for 78 eyes (32.2%), 122 eyes (50.4%), and 211 eyes (87.2%), respectively. The proportion of eyes with MD values forecasted within 2.5 dB of the actual value for the KF-NTG (87.2%) were similar to KF-HTG (86.0%) and the null model (86.4%), and much better than the 2 linear regression-based models (72.7-74.0%; P < .001). When forecasting MD, KF-NTG (RMSE = 2.71) and KF-HTG (RMSE = 2.68) achieved lower RMSE than the other 3 forecasting models (RMSE = 2.81-3.90), indicating better performance.

Conclusion: As observed previously for patients with HTG, KF can also effectively forecast disease trajectory for many patients with NTG.

Citing Articles

Prediction of visual field progression in glaucoma: existing methods and artificial intelligence.

Asaoka R, Murata H Jpn J Ophthalmol. 2023; 67(5):546-559.

PMID: 37540325 DOI: 10.1007/s10384-023-01009-3.

Use of artificial intelligence in forecasting glaucoma progression.

Thakur S, Dinh L, Lavanya R, Quek T, Liu Y, Cheng C Taiwan J Ophthalmol. 2023; 13(2):168-183.

PMID: 37484617 PMC: 10361424. DOI: 10.4103/tjo.TJO-D-23-00022.

Deep Learning Approaches for Predicting Glaucoma Progression Using Electronic Health Records and Natural Language Processing.

Wang S, Tseng B, Hernandez-Boussard T Ophthalmol Sci. 2022; 2(2):100127.

PMID: 36249690 PMC: 9559076. DOI: 10.1016/j.xops.2022.100127.

Augmenting Kalman Filter Machine Learning Models with Data from OCT to Predict Future Visual Field Loss: An Analysis Using Data from the African Descent and Glaucoma Evaluation Study and the Diagnostic Innovation in Glaucoma Study.

Zhalechian M, Van Oyen M, Lavieri M, De Moraes C, Girkin C, Fazio M Ophthalmol Sci. 2022; 2(1):100097.

PMID: 36246178 PMC: 9560647. DOI: 10.1016/j.xops.2021.100097.

Current and Future Implications of Using Artificial Intelligence in Glaucoma Care.

Ahuja A, Bommakanti S, Wagner I, Dorairaj S, Ten Hulzen R, Checo L J Curr Ophthalmol. 2022; 34(2):129-132.

PMID: 36147268 PMC: 9486995. DOI: 10.4103/joco.joco_39_22.

References

McNaught A, Crabb D, Fitzke F, Hitchings R . Modelling series of visual fields to detect progression in normal-tension glaucoma. Graefes Arch Clin Exp Ophthalmol. 1995; 233(12):750-5. DOI: 10.1007/BF00184085. View

Anderson D, Drance S, Schulzer M . Natural history of normal-tension glaucoma. Ophthalmology. 2001; 108(2):247-53. DOI: 10.1016/s0161-6420(00)00518-2. View

Stein J, Kim D, Niziol L, Talwar N, Nan B, Musch D . Differences in rates of glaucoma among Asian Americans and other racial groups, and among various Asian ethnic groups. Ophthalmology. 2011; 118(6):1031-7. PMC: 3109193. DOI: 10.1016/j.ophtha.2010.10.024. View

Nitta K, Wajima R, Tachibana G, Inoue S, Ohigashi T, Otsuka N . Prediction of Visual Field Progression in Patients with Primary Open-Angle Glaucoma, Mainly Including Normal Tension Glaucoma. Sci Rep. 2017; 7(1):15048. PMC: 5678075. DOI: 10.1038/s41598-017-15267-y. View

Schell G, Lavieri M, Helm J, Liu X, Musch D, Van Oyen M . Using filtered forecasting techniques to determine personalized monitoring schedules for patients with open-angle glaucoma. Ophthalmology. 2014; 121(8):1539-46. PMC: 4495761. DOI: 10.1016/j.ophtha.2014.02.021. View

Heijl A, Bengtsson B, Hyman L, Leske M . Natural history of open-angle glaucoma. Ophthalmology. 2009; 116(12):2271-6. DOI: 10.1016/j.ophtha.2009.06.042. View

Sung M, Kang Y, Heo H, Park S . Optic Disc Rotation as a Clue for Predicting Visual Field Progression in Myopic Normal-Tension Glaucoma. Ophthalmology. 2016; 123(7):1484-93. DOI: 10.1016/j.ophtha.2016.03.040. View

Thonginnetra O, Greenstein V, Chu D, Liebmann J, Ritch R, Hood D . Normal versus high tension glaucoma: a comparison of functional and structural defects. J Glaucoma. 2009; 19(3):151-7. PMC: 2891909. DOI: 10.1097/IJG.0b013e318193c45c. View

Musch D, Gillespie B, Lichter P, Niziol L, Janz N . Visual field progression in the Collaborative Initial Glaucoma Treatment Study the impact of treatment and other baseline factors. Ophthalmology. 2008; 116(2):200-7. PMC: 3316491. DOI: 10.1016/j.ophtha.2008.08.051. View

10.

Sommer A . Ocular hypertension and normal-tension glaucoma: time for banishment and burial. Arch Ophthalmol. 2011; 129(6):785-7. DOI: 10.1001/archophthalmol.2011.117. View

11.

Gardiner S, Crabb D . Examination of different pointwise linear regression methods for determining visual field progression. Invest Ophthalmol Vis Sci. 2002; 43(5):1400-7. View

12.

Musch D, Lichter P, Guire K, Standardi C . The Collaborative Initial Glaucoma Treatment Study: study design, methods, and baseline characteristics of enrolled patients. Ophthalmology. 1999; 106(4):653-62. DOI: 10.1016/s0161-6420(99)90147-1. View

13.

Kazemian P, Lavieri M, Van Oyen M, Andrews C, Stein J . Personalized Prediction of Glaucoma Progression Under Different Target Intraocular Pressure Levels Using Filtered Forecasting Methods. Ophthalmology. 2017; 125(4):569-577. PMC: 5866175. DOI: 10.1016/j.ophtha.2017.10.033. View

14.

Bengtsson B, Patella V, Heijl A . Prediction of glaucomatous visual field loss by extrapolation of linear trends. Arch Ophthalmol. 2009; 127(12):1610-5. DOI: 10.1001/archophthalmol.2009.297. View

15.

Shields M . Normal-tension glaucoma: is it different from primary open-angle glaucoma?. Curr Opin Ophthalmol. 2008; 19(2):85-8. DOI: 10.1097/ICU.0b013e3282f3919b. View

16.

. [The Japan Glaucoma Society Guidelines for Glaucoma (3rd Edition)]. Nippon Ganka Gakkai Zasshi. 2012; 116(1):3-46. View

17.

Wang Q, Molenaar P, Harsh S, Freeman K, Xie J, Gold C . Personalized State-space Modeling of Glucose Dynamics for Type 1 Diabetes Using Continuously Monitored Glucose, Insulin Dose, and Meal Intake: An Extended Kalman Filter Approach. J Diabetes Sci Technol. 2014; 8(2):331-345. PMC: 4455398. DOI: 10.1177/1932296814524080. View

18.

Eberle C, Ament C . The Unscented Kalman Filter estimates the plasma insulin from glucose measurement. Biosystems. 2010; 103(1):67-72. DOI: 10.1016/j.biosystems.2010.09.012. View

19.

Drance S, Anderson D, Schulzer M . Risk factors for progression of visual field abnormalities in normal-tension glaucoma. Am J Ophthalmol. 2001; 131(6):699-708. DOI: 10.1016/s0002-9394(01)00964-3. View

20.

Ahrlich K, De Moraes C, Teng C, Prata T, Tello C, Ritch R . Visual field progression differences between normal-tension and exfoliative high-tension glaucoma. Invest Ophthalmol Vis Sci. 2010; 51(3):1458-63. DOI: 10.1167/iovs.09-3806. View