Artificial Intelligence in Cornea, Refractive, and Cataract Surgery

Aazim A. Siddiqui; John G. Ladas; Jimmy K. Lee


Curr Opin Ophthalmol. 2020;31(4):253-260. 

In This Article


Over the years, the advent and integration of technology has fueled progress within medicine, ophthalmology, and many of its subspecialties. With its integration into the plethora of areas within ophthalmology, artificial intelligence is continuing to play a major role in not only the diagnostic but also therapeutic element of disease management.

In cornea and refractive aspects of ophthalmology, artificial intelligence could potentially play a substantial role in early identification of keratoconus and other ectatic disorders. In cataract surgery, it can improve refractive outcomes with its integration into IOL calculation formulas and automated refraction devices.

The super formula may be trained and developed further to yield increasingly accurate solutions. The next step in increasing the integration of artificial intelligence is to incorporate additional variables to achieve higher accuracy. These variables could include white-to-white, lens thickness, and posterior corneal power. Further, the best features of any existing IOL formula may be included into the super formula algorithm. Several aspects of the formula may be optimized further based on actual postoperative outcome data. This 'big data' and 'crowd-sourced' approach could eventually use millions of data points to achieve very high levels of accuracy. This approach can evolve over time and become a system or a 'living organism' that is constantly improving.

The future of artificial intelligence in ophthalmology is promising. The desire to achieve early diagnosis of debilitating corneal disorders, and achieve improved postoperative refractive outcomes is responsible for the continued growth in the industry of artificial intelligence and ophthalmology. In IOL calculations, the super formula framework may evolve in the near future to address the challenges of post-refractive, toric, and pediatric cataract surgeries. With continued deployment of advanced mathematical techniques and artificial intelligence, the process of IOL calculations can reach higher levels of accuracy.