Related Experiment Video
Updated: Jan 18, 2026
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
3.3K
Artificial Intelligence in Ocular Transcriptomics: Applications of Unsupervised and Supervised Learning
Catherine Lalman1,2, Yimin Yang3, Janice L Walker1,2,4
1Department of Pathology and Genomic Medicine, Thomas Jefferson University, Philadelphia, PA 19107, USA.
Cells
|September 13, 2025
Summary
Artificial intelligence (AI) and machine learning (ML) are revolutionizing ophthalmology by analyzing complex gene expression data from transcriptomic profiling. These AI methods enhance biomarker discovery and disease modeling for various eye conditions.
Area of Science:
- Ophthalmology
- Genomics
- Bioinformatics
Background:
- Transcriptomic profiling offers deep insights into ocular tissue complexity.
- Advancements in RNA sequencing (RNA-seq) and single-cell RNA-seq generate high-dimensional gene expression data.
- Artificial intelligence (AI) is crucial for analyzing this complex data.
Purpose of the Study:
- To review AI-enabled transcriptomic studies in ophthalmology from 2019-2025.
- To highlight the role of machine learning (ML) in biomarker discovery, cell classification, and disease modeling.
- To discuss the application of AI in understanding eye development and ocular diseases.
Main Methods:
- Review of AI applications in transcriptomic analysis for ophthalmology.
- Discussion of unsupervised ML methods: Principal Component Analysis (PCA), t-distributed Stochastic Neighbor Embedding (t-SNE), Uniform Manifold Approximation and Projection (UMAP), Weighted Gene Co-expression Network Analysis (WGCNA).
- Discussion of supervised ML methods: Least Absolute Shrinkage and Selection Operator (LASSO), Support Vector Machines (SVMs), Random Forests (RFs), and deep learning frameworks (variational autoencoders, neural networks).
Main Results:
- AI and ML methods have significantly advanced biomarker discovery for ocular diseases like age-related macular degeneration (AMD), diabetic retinopathy (DR), and glaucoma.
- Unsupervised and supervised ML techniques are standard in single-cell RNA-seq workflows for cell type classification and disease modeling.
- AI facilitates multi-omics integration and supports the development of diagnostic and prognostic markers.
Conclusions:
- AI-driven transcriptomic analysis is transforming precision ophthalmology.
- Explainable AI and multimodal approaches are key to overcoming challenges in interpretability and standardization.
- Future research directions include advancing AI for personalized eye care and disease management.
Keywords:
artificial intelligencecorneaeyeocular transcriptomicsretinasupervised and unsupervised machine learning
