Session-by-Session Prediction of Anti-Endothelial Growth Factor Injection Needs in Neovascular Age-Related Macular Degeneration Using Optical-Coherence-Tomography-Derived Features and Machine Learning

Background/Objectives: Neovascular age-related macular degeneration (nAMD) is a retinal disorder leading to irreversible central vision loss. The pro-re-nata (PRN) treatment for nAMD involves frequent intravitreal injections of anti-VEGF medications, placing a burden on patients and healthcare syste...

Full description

Saved in:
Bibliographic Details
Main Authors: Flavio Ragni, Stefano Bovo, Andrea Zen, Diego Sona, Katia De Nadai, Ginevra Giovanna Adamo, Marco Pellegrini, Francesco Nasini, Chiara Vivarelli, Marco Tavolato, Marco Mura, Francesco Parmeggiani, Giuseppe Jurman
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/14/23/2609
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846124404571897856
author Flavio Ragni
Stefano Bovo
Andrea Zen
Diego Sona
Katia De Nadai
Ginevra Giovanna Adamo
Marco Pellegrini
Francesco Nasini
Chiara Vivarelli
Marco Tavolato
Marco Mura
Francesco Parmeggiani
Giuseppe Jurman
author_facet Flavio Ragni
Stefano Bovo
Andrea Zen
Diego Sona
Katia De Nadai
Ginevra Giovanna Adamo
Marco Pellegrini
Francesco Nasini
Chiara Vivarelli
Marco Tavolato
Marco Mura
Francesco Parmeggiani
Giuseppe Jurman
author_sort Flavio Ragni
collection DOAJ
description Background/Objectives: Neovascular age-related macular degeneration (nAMD) is a retinal disorder leading to irreversible central vision loss. The pro-re-nata (PRN) treatment for nAMD involves frequent intravitreal injections of anti-VEGF medications, placing a burden on patients and healthcare systems. Predicting injections needs at each monitoring session could optimize treatment outcomes and reduce unnecessary interventions. Methods: To achieve these aims, machine learning (ML) models were evaluated using different combinations of clinical variables, including retinal thickness and volume, best-corrected visual acuity, and features derived from macular optical coherence tomography (OCT). A “Leave Some Subjects Out” (LSSO) nested cross-validation approach ensured robust evaluation. Moreover, the SHapley Additive exPlanations (SHAP) analysis was employed to quantify the contribution of each feature to model predictions. Results: Results demonstrated that models incorporating both structural and functional features achieved high classification accuracy in predicting injection necessity (AUC = 0.747 ± 0.046, MCC = 0.541 ± 0.073). Moreover, the explainability analysis identified as key predictors both subretinal and intraretinal fluid, alongside central retinal thickness. Conclusions: These findings suggest that session-by-session prediction of injection needs in nAMD patients is feasible, even without processing the entire OCT image. The proposed ML framework has the potential to be integrated into routine clinical workflows, thereby optimizing nAMD therapeutic management.
format Article
id doaj-art-f3ee2f0a607848e69dcbca5fd214a2e1
institution Kabale University
issn 2075-4418
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-f3ee2f0a607848e69dcbca5fd214a2e12024-12-13T16:24:23ZengMDPI AGDiagnostics2075-44182024-11-011423260910.3390/diagnostics14232609Session-by-Session Prediction of Anti-Endothelial Growth Factor Injection Needs in Neovascular Age-Related Macular Degeneration Using Optical-Coherence-Tomography-Derived Features and Machine LearningFlavio Ragni0Stefano Bovo1Andrea Zen2Diego Sona3Katia De Nadai4Ginevra Giovanna Adamo5Marco Pellegrini6Francesco Nasini7Chiara Vivarelli8Marco Tavolato9Marco Mura10Francesco Parmeggiani11Giuseppe Jurman12Data Science for Health Unit, Fondazione Bruno Kessler, 38123 Trento, ItalyData Science for Health Unit, Fondazione Bruno Kessler, 38123 Trento, ItalyData Science for Health Unit, Fondazione Bruno Kessler, 38123 Trento, ItalyData Science for Health Unit, Fondazione Bruno Kessler, 38123 Trento, ItalyDepartment of Translational Medicine and for Romagna, University of Ferrara, 44121 Ferrara, ItalyDepartment of Translational Medicine and for Romagna, University of Ferrara, 44121 Ferrara, ItalyDepartment of Translational Medicine and for Romagna, University of Ferrara, 44121 Ferrara, ItalyUnit of Ophthalmology, Azienda Ospedaliero Universitaria di Ferrara, 44100 Ferrara, ItalyDepartment of Translational Medicine and for Romagna, University of Ferrara, 44121 Ferrara, ItalyERN-EYE Network—Center Retinitis Pigmentosa of Veneto Region, Camposampiero Hospital, 35012 Padua, ItalyDepartment of Translational Medicine and for Romagna, University of Ferrara, 44121 Ferrara, ItalyDepartment of Translational Medicine and for Romagna, University of Ferrara, 44121 Ferrara, ItalyData Science for Health Unit, Fondazione Bruno Kessler, 38123 Trento, ItalyBackground/Objectives: Neovascular age-related macular degeneration (nAMD) is a retinal disorder leading to irreversible central vision loss. The pro-re-nata (PRN) treatment for nAMD involves frequent intravitreal injections of anti-VEGF medications, placing a burden on patients and healthcare systems. Predicting injections needs at each monitoring session could optimize treatment outcomes and reduce unnecessary interventions. Methods: To achieve these aims, machine learning (ML) models were evaluated using different combinations of clinical variables, including retinal thickness and volume, best-corrected visual acuity, and features derived from macular optical coherence tomography (OCT). A “Leave Some Subjects Out” (LSSO) nested cross-validation approach ensured robust evaluation. Moreover, the SHapley Additive exPlanations (SHAP) analysis was employed to quantify the contribution of each feature to model predictions. Results: Results demonstrated that models incorporating both structural and functional features achieved high classification accuracy in predicting injection necessity (AUC = 0.747 ± 0.046, MCC = 0.541 ± 0.073). Moreover, the explainability analysis identified as key predictors both subretinal and intraretinal fluid, alongside central retinal thickness. Conclusions: These findings suggest that session-by-session prediction of injection needs in nAMD patients is feasible, even without processing the entire OCT image. The proposed ML framework has the potential to be integrated into routine clinical workflows, thereby optimizing nAMD therapeutic management.https://www.mdpi.com/2075-4418/14/23/2609neovascular age-related macular degenerationoptical coherence tomographyanti-VEGF drugsartificial intelligencemachine learninginjection prediction
spellingShingle Flavio Ragni
Stefano Bovo
Andrea Zen
Diego Sona
Katia De Nadai
Ginevra Giovanna Adamo
Marco Pellegrini
Francesco Nasini
Chiara Vivarelli
Marco Tavolato
Marco Mura
Francesco Parmeggiani
Giuseppe Jurman
Session-by-Session Prediction of Anti-Endothelial Growth Factor Injection Needs in Neovascular Age-Related Macular Degeneration Using Optical-Coherence-Tomography-Derived Features and Machine Learning
Diagnostics
neovascular age-related macular degeneration
optical coherence tomography
anti-VEGF drugs
artificial intelligence
machine learning
injection prediction
title Session-by-Session Prediction of Anti-Endothelial Growth Factor Injection Needs in Neovascular Age-Related Macular Degeneration Using Optical-Coherence-Tomography-Derived Features and Machine Learning
title_full Session-by-Session Prediction of Anti-Endothelial Growth Factor Injection Needs in Neovascular Age-Related Macular Degeneration Using Optical-Coherence-Tomography-Derived Features and Machine Learning
title_fullStr Session-by-Session Prediction of Anti-Endothelial Growth Factor Injection Needs in Neovascular Age-Related Macular Degeneration Using Optical-Coherence-Tomography-Derived Features and Machine Learning
title_full_unstemmed Session-by-Session Prediction of Anti-Endothelial Growth Factor Injection Needs in Neovascular Age-Related Macular Degeneration Using Optical-Coherence-Tomography-Derived Features and Machine Learning
title_short Session-by-Session Prediction of Anti-Endothelial Growth Factor Injection Needs in Neovascular Age-Related Macular Degeneration Using Optical-Coherence-Tomography-Derived Features and Machine Learning
title_sort session by session prediction of anti endothelial growth factor injection needs in neovascular age related macular degeneration using optical coherence tomography derived features and machine learning
topic neovascular age-related macular degeneration
optical coherence tomography
anti-VEGF drugs
artificial intelligence
machine learning
injection prediction
url https://www.mdpi.com/2075-4418/14/23/2609
work_keys_str_mv AT flavioragni sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT stefanobovo sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT andreazen sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT diegosona sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT katiadenadai sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT ginevragiovannaadamo sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT marcopellegrini sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT francesconasini sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT chiaravivarelli sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT marcotavolato sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT marcomura sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT francescoparmeggiani sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning
AT giuseppejurman sessionbysessionpredictionofantiendothelialgrowthfactorinjectionneedsinneovascularagerelatedmaculardegenerationusingopticalcoherencetomographyderivedfeaturesandmachinelearning