Artificial intelligence in ophthalmology. Do we need risk calculators for glaucoma development and progression?

Artificial intelligence (AI) is rapidly entering modern medical practice. Many routine clinical tasks, from imaging and automated diagnostics to robotic surgery, cannot be imagined without the use of AI. Neural networks show impressive results when analyzing a large amount of data obtained from standard automated perimetry, optical coherence tomography (OCT) and fundus photography. Currently, both in Russia and abroad mathematical algorithms are being developed that allow detection of glaucoma based on certain signs. This article analyzes the advantages and disadvantages of employing artificial intelligence in ophthalmological practice, discusses the need for careful selection of the criteria and their influence on the accuracy of calculators, considers the specifics of using mathematical analysis in suspected glaucoma, as well as in an already established diagnosis. The article also provides clinical examples of the use of glaucoma risk calculator in the routine practice of an ophthalmologist.

[1]  A. A. Antonov,et al.  A method for measuring intraocular pressure using artificial intelligence technology and fixed-force applanation tonometry , 2022, Russian Ophthalmological Journal.

[2]  P. Koulen,et al.  Novel Machine-Learning Based Framework Using Electroretinography Data for the Detection of Early-Stage Glaucoma , 2022, Frontiers in Neuroscience.

[3]  I. Simakova,et al.  Evaluation of the effectiveness of frequency doubling technology perimetry in the diagnosis of optic neuropathies , 2022, National Journal glaucoma.

[4]  J. Heier,et al.  Prospective, longitudinal study: daily self-imaging with home OCT in neovascular age-related macular degeneration. , 2022, Ophthalmology. Retina.

[5]  Ryan P. McNabb,et al.  Contactless optical coherence tomography of the eyes of freestanding individuals with a robotic scanner , 2021, Nature Biomedical Engineering.

[6]  Aletta Eikelboom,et al.  Human- versus Artificial Intelligence , 2021, Frontiers in Artificial Intelligence.

[7]  Orlaith Mc Grath,et al.  Clinical Utility of Artificial Intelligence Algorithms to Enhance Wide-Field Optical Coherence Tomography Angiography Images , 2021, J. Imaging.

[8]  Angelica I. Avilés-Rivero,et al.  Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans , 2020, Nature Machine Intelligence.

[9]  Chaz Firestone,et al.  Performance vs. competence in human–machine comparisons , 2020, Proceedings of the National Academy of Sciences.

[10]  Alejandro F Frangi,et al.  Development and clinical deployment of a smartphone-based visual field deep learning system for glaucoma detection , 2020, npj Digital Medicine.

[11]  John Paoli,et al.  Human–computer collaboration for skin cancer recognition , 2020, Nature Medicine.

[12]  F. Medeiros,et al.  Quantification of Retinal Nerve Fibre Layer Thickness on Optical Coherence Tomography with a Deep Learning Segmentation-Free Approach , 2020, Scientific Reports.

[13]  A. J. Mariano,et al.  Human–machine partnership with artificial intelligence for chest radiograph diagnosis , 2019, npj Digital Medicine.

[14]  Gian-Gabriel P. Garcia,et al.  Accuracy of Kalman Filtering in Forecasting Visual Field and Intraocular Pressure Trajectory in Patients With Ocular Hypertension. , 2019, JAMA ophthalmology.

[15]  Nara G. Ogata,et al.  Detecting Retinal Nerve Fibre Layer Segmentation Errors on Spectral Domain-Optical Coherence Tomography with a Deep Learning Algorithm , 2019, Scientific Reports.

[16]  Gian-Gabriel P. Garcia,et al.  Using Kalman Filtering to Forecast Disease Trajectory for Patients With Normal Tension Glaucoma. , 2019, American journal of ophthalmology.

[17]  Eui Jin Hwang,et al.  Development and Validation of Deep Learning-based Automatic Detection Algorithm for Malignant Pulmonary Nodules on Chest Radiographs. , 2019, Radiology.

[18]  Hiroshi Ishikawa,et al.  A feature agnostic approach for glaucoma detection in OCT volumes , 2018, PloS one.

[19]  Vasudevan Lakshminarayanan,et al.  Application of Deep Learning in Fundus Image Processing for Ophthalmic Diagnosis -- A Review. , 2019 .

[20]  Raphael Sznitman,et al.  A deep learning approach to automatic detection of early glaucoma from visual fields , 2018, PloS one.

[21]  Mark A. Christopher,et al.  Performance of Deep Learning Architectures and Transfer Learning for Detecting Glaucomatous Optic Neuropathy in Fundus Photographs , 2018, Scientific Reports.

[22]  M. Teus,et al.  Longitudinal Change in Central Corneal Thickness in the Tema Eye Survey. , 2018, American journal of ophthalmology-glaucoma.

[23]  M. He,et al.  Efficacy of a Deep Learning System for Detecting Glaucomatous Optic Neuropathy Based on Color Fundus Photographs. , 2018, Ophthalmology.

[24]  Pooyan Kazemian,et al.  Personalized Prediction of Glaucoma Progression Under Different Target Intraocular Pressure Levels Using Filtered Forecasting Methods. , 2017, Ophthalmology.

[25]  Scientific,et al.  Suspected glaucoma and early stage glaucoma: differential diagnostic criteria , 2017 .

[26]  D. Lovpache,et al.  Pharmacoeconomic expediency of the use of preservative-free therapy in the treatment of primary open-angle glaucoma in patients with the dry eye syndrome (intermediate results) , 2017 .

[27]  Hiroshi Murata,et al.  Detecting Preperimetric Glaucoma with Standard Automated Perimetry Using a Deep Learning Classifier. , 2016, Ophthalmology.

[28]  Martin Emiliano Leuci.  Machine to machine , 2016 .

[29]  G. Rebolleda,et al.  Re: Kim et al.: Diagnostic classification of macular ganglion cell and retinal nerve fiber layer analysis: differentiation of false-positives from glaucoma (Ophthalmology 2015;122:502-10). , 2015, Ophthalmology.

[30]  K. Park,et al.  Author reply: To PMID 25444638. , 2015, Ophthalmology (Rochester, Minn.).

[31]  Ki Ho Park,et al.  Diagnostic classification of macular ganglion cell and retinal nerve fiber layer analysis: differentiation of false-positives from glaucoma. , 2015, Ophthalmology.

[32]  Ryo Asaoka,et al.  Identifying "preperimetric" glaucoma in standard automated perimetry visual fields. , 2014, Investigative ophthalmology & visual science.

[33]  Shu Liu,et al.  Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: a prospective analysis of age-related loss. , 2012, Ophthalmology.

[34]  Michael V. Boland,et al.  Effect of patient's life expectancy on the cost-effectiveness of treatment for ocular hypertension. , 2010, Archives of ophthalmology.

[35]  J. Cairns,et al.  The clinical effectiveness and cost-effectiveness of screening for open angle glaucoma: a systematic review and economic evaluation. , 2007, Health technology assessment.

[36]  R. D'Agostino,et al.  Validated prediction model for the development of primary open-angle glaucoma in individuals with ocular hypertension. , 2007, Ophthalmology.

[37]  S. Miglior,et al.  Predictive factors for open-angle glaucoma among patients with ocular hypertension in the European Glaucoma Prevention Study. , 2007, Ophthalmology.

[38]  Tom Fawcett,et al.  An introduction to ROC analysis , 2006, Pattern Recognit. Lett..

[39]  Philip P. Chen,et al.  A multicenter, retrospective pilot study of resource use and costs associated with severity of disease in glaucoma. , 2006, Archives of ophthalmology.

[40]  Robert N Weinreb,et al.  Risk assessment in the management of patients with ocular hypertension. , 2004, American journal of ophthalmology.

[41]  M. Yablonski.  Ocular hypertension treatment study. , 2004, Archives of ophthalmology.

[42]  E. E. Hartmann,et al.  The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma. , 2002, Archives of ophthalmology.

[43]  M. Kass,et al.  The Ocular Hypertension Treatment Study: design and baseline description of the participants. , 1999, Archives of ophthalmology.

[44]  K. Smigel.  Breast Cancer Prevention Trial shows major benefit, some risk. , 1998, Japanese journal of clinical oncology.