AI4VS Research

The Artificial Intelligence for Vision Science (AI4VS) Laboratory is led by Professor Kaveri Thakoor and is located at the Columbia University Irving Medical Center (CUIMC).

 

 

 

 

 

 

 

The goal of the AI4VS Lab is to develop human-vision-inspired AI to automate the analysis of biomedical images to augment and work in tandem with the abilities of medical experts to expedite the detection of eye diseases, such as glaucoma, age-related macular degeneration, and beyond.

Research Focus Areas

Robust

We develop AI that is robust to data collected at multiple locations. See some of our past projects which achieve accurate glaucoma and AMD detection from multiple locations or using multiple imaging modalities.

Explore Robustness

Interpretable

We develop AI in such a way that the mechanisms behind its decision-making can be understood by clinicians. See some of our past projects using expert eye tracking and statistical odds ratio analysis to corroborate AI mechanisms.

Uncover Interpretability

Portable

We develop AI systems which perform accurately even on data collected with low-cost, portable imaging systems. See our proof-of-concept work which may enable a broader population to gain access to sight-saving technology.

Discover Portability