New 3D lens geometry-based formula improves intraocular lens positioning, reducing refractive errors in cataract surgery
Institute Reference: 6-24053
In cataract surgery, selecting the correct intraocular lens (IOL) power is essential for achieving optimal visual outcomes. However, this process is often challenging due to the difficulty in predicting the postoperative position of the lens. Traditional IOL power estimation formulas rely on limited parameters, and typically do not consider the crystalline lens anatomy, where the IOL is implanted. Estimation errors in lens position result in IOL power selection errors that affect refractive accuracy. To address these limitations, researchers have developed a new AI-powered formula incorporating 3D crystalline lens geometry that offers a breakthrough in IOL power calculation.
This new method leverages the full three-dimensional geometry of the crystalline lens, captured non-invasively using optical coherence tomography (OCT). The formula integrates parameters such as lens volume, surface area, equatorial plane position, and diameter—data that previous formulas did not utilize. By doing so, it provides a more precise estimation of the lens's postoperative position and tilt angle, minimizing one of the primary sources of refractive error in cataract surgery.
Recent studies have demonstrated that the new approach significantly reduces lens positioning errors compared to standard methods. Testing on real patients has shown the mean absolute estimation error was reduced by more than 50%, showing clear potential to improve refractive outcomes. Moreover, this method offers an adaptable framework that can be applied to various IOL models and patient cases, including those with complex eye geometries or prior refractive surgeries. Additionally, the geometrical parameters used in the formulas can be obtained through quantitative analysis of images obtained with various commercially available OCT instruments.
The benefits of this innovation are far-reaching. It offers surgeons more reliable tools for lens positioning, which in turn improves visual outcomes for patients. By utilizing high-resolution OCT imaging, the approach is entirely non-invasive and provides rapid data collection for surgical planning. It also supports the development of personalized eye models, which can simulate individual anatomical characteristics for ray-tracing analysis—especially useful for patients with irregular corneas or prior LASIK surgeries.
The University of Rochester is open to exploring funded research collaborations, licensing agreements, and other partnership opportunities.