Customised Selection of the Haptic Design in C-Loop Intraocular Lenses Based on Deep Learning

Area of Science:

• Ophthalmology
• Biomedical Engineering
• Artificial Intelligence

Background:

• Successful cataract surgery requires optimal intraocular lens (IOL) performance.
• Patient-specific IOL customization is crucial for enhancing surgical outcomes.
• Current methods may not fully account for individual biomechanical characteristics.

Purpose of the Study:

• To develop deep neural network (DNN) models for predicting IOL biomechanical stability and optimal haptic design.
• To enable personalized IOL selection based on patient-specific biomechanical responses.
• To demonstrate the application of AI in optimizing IOL selection for cataract surgery.

Main Methods:

• Utilized a validated finite element model to generate biomechanical data for diverse IOL geometries.
• Tested IOLs according to ISO 11979-3 compression standards.
• Developed two DNN models: one for predicting biomechanical stability, another for designing optimal haptics.

Main Results:

• The biomechanical response model achieved high accuracy (Pearson's r = 0.995).
• The IOL haptic design model demonstrated that multiple designs can yield the same biomechanical response (Pearson's r = 0.992).
• A graphical user interface (GUI) was developed to showcase the practical application of these DNN models.

Conclusions:

• Deep learning models offer a powerful tool for analyzing IOL designs and predicting biomechanical performance.
• Personalized IOL selection based on predicted biomechanical response can improve cataract surgery success rates.
• This AI-driven approach facilitates informed decision-making for IOL manufacturers and ophthalmologists.