AI and deep learning revolutionized ocular imaging in patient studies and continue to provide novel insights into the origins and management of progressive eye disease. The utilization of AI protocols for the detection and management of glaucoma is challenged by diversity in patient populations, access to imaging modalities, and variability between imaging platforms. Current efforts in AI for glaucoma seek to identify critical features, correction factors, and registration paradigms that will enable broad application of an AI approach to glaucoma care. In the preclinical space, development of AI approaches for detection and assessment of glaucoma pathology in animal models is a burgeoning frontier. This provides a unique environment within the glaucoma field, where advances in the clinic could drive advances in preclinical science and translation. Specifically, the simplification of AI protocols to address clinical challenges may also facilitate reverse translation of AI protocols to animal models.
Bench to bedside translation is inhibited by the relevance of preclinical outcomes to clinical manifestations. The development of preclinical outcomes with direct or strong associations to clinical outcomes is a means to overcoming this challenge. Could advances in AI technology that facilitate broad application to glaucoma patients also be used to derive AI-based approaches with high translational potential in animal models?
The goal of this Research Topic is to consider the potential for reverse translation of AI in glaucoma by providing an overview of current efforts to improve generalized application of AI in glaucoma care as well as current applications of AI to preclinical outcomes in animal models.
This article collection seeks to highlight the opportunity for translational advancement in glaucoma via integration of AI approaches in preclinical models and clinical populations. Thus, studies utilizing AI approaches in both preclinical and clinical arenas are applicable. Highly relevant studies will: 1) directly address AI translation between preclinical and clinical datasets, 2) utilize novel AI approaches in animal models, 3) develop AI approaches to simplify clinical data sets for broad application, or 4) identify disease features for AI development and translation between clinical outcome modalities.
We welcome articles addressing, but not restricted to, the following:
- Deep learning approaches to outcomes in animal models
- Deep learning approaches to clinical outcomes
- Transfer learning approaches in glaucoma detection and management
- AI-based registration between clinical outcome modalities, i.e. MRI and OCT
- Refinement paradigms for existing AI approaches in relevant clinical diagnostics
- AI approaches targeting variability between clinical populations or acquisition parameters
- Assessment of confounds for current AI approaches in preclinical or clinical outcomes
AI and deep learning revolutionized ocular imaging in patient studies and continue to provide novel insights into the origins and management of progressive eye disease. The utilization of AI protocols for the detection and management of glaucoma is challenged by diversity in patient populations, access to imaging modalities, and variability between imaging platforms. Current efforts in AI for glaucoma seek to identify critical features, correction factors, and registration paradigms that will enable broad application of an AI approach to glaucoma care. In the preclinical space, development of AI approaches for detection and assessment of glaucoma pathology in animal models is a burgeoning frontier. This provides a unique environment within the glaucoma field, where advances in the clinic could drive advances in preclinical science and translation. Specifically, the simplification of AI protocols to address clinical challenges may also facilitate reverse translation of AI protocols to animal models.
Bench to bedside translation is inhibited by the relevance of preclinical outcomes to clinical manifestations. The development of preclinical outcomes with direct or strong associations to clinical outcomes is a means to overcoming this challenge. Could advances in AI technology that facilitate broad application to glaucoma patients also be used to derive AI-based approaches with high translational potential in animal models?
The goal of this Research Topic is to consider the potential for reverse translation of AI in glaucoma by providing an overview of current efforts to improve generalized application of AI in glaucoma care as well as current applications of AI to preclinical outcomes in animal models.
This article collection seeks to highlight the opportunity for translational advancement in glaucoma via integration of AI approaches in preclinical models and clinical populations. Thus, studies utilizing AI approaches in both preclinical and clinical arenas are applicable. Highly relevant studies will: 1) directly address AI translation between preclinical and clinical datasets, 2) utilize novel AI approaches in animal models, 3) develop AI approaches to simplify clinical data sets for broad application, or 4) identify disease features for AI development and translation between clinical outcome modalities.
We welcome articles addressing, but not restricted to, the following:
- Deep learning approaches to outcomes in animal models
- Deep learning approaches to clinical outcomes
- Transfer learning approaches in glaucoma detection and management
- AI-based registration between clinical outcome modalities, i.e. MRI and OCT
- Refinement paradigms for existing AI approaches in relevant clinical diagnostics
- AI approaches targeting variability between clinical populations or acquisition parameters
- Assessment of confounds for current AI approaches in preclinical or clinical outcomes
