Artificial intelligence techniques in inherited retinal diseases: A review

AI shows promise for diagnosing inherited retinal diseases (IRDs), which are a major cause of blindness in working-age adults. The paper highlights that convolutional neural networks are particularly effective at detecting these complex genetic disorders, offering potential solutions to diagnostic challenges that currently exist in clinical practice.

Explainable AI is critical for clinical trust and adoption. The review emphasizes that transparency in how AI systems make decisions is essential in medical settings. Rather than treating AI as a "black box," the paper stresses the importance of integrating explainable AI so doctors and patients can understand why the system reached a particular diagnosis or recommendation.

AI can help predict disease progression and enable personalized treatment. Beyond diagnosis, the paper identifies AI's potential to forecast how IRDs will advance in individual patients and to tailor treatment plans accordingly—moving toward more personalized medicine rather than one-size-fits-all approaches.

Significant research gaps remain despite rapid AI development. The review notes that knowledge in this field is currently "fragmented," indicating that while AI techniques are advancing quickly, focused studies specifically examining their application to IRDs are still limited, requiring more targeted research efforts.

Successful AI deployment requires interdisciplinary collaboration. The paper concludes that advancing AI for IRDs will depend on partnerships between AI researchers, ophthalmologists, geneticists, and other specialists, rather than siloed technical development.