This technology introduces a portable and affordable way to perform visual field assessment using only a standard webcam and machine learning software. Unlike conventional systems that rely on costly specialized equipment and require constant supervision, this approach tracks eye movements in response to simple on-screen visual stimuli. The software then analyzes gaze patterns to infer potential visual impairments. By removing the need for internet connectivity, external devices, or technician oversight, the system provides a scalable and accessible solution that can be deployed in clinics, telehealth platforms, or underserved communities. Its offline capability makes it particularly valuable for rural or resource-limited settings, where early detection of vision problems is often inaccessible. Background: Standard automated perimetry, the current gold standard for visual field testing, demands bulky equipment, fixed head positioning, and manual patient responses. These requirements present barriers for elderly, pediatric, and disabled patients, while also creating cost and workflow burdens for clinics. Virtual reality and eye movement perimetry have emerged as alternatives, but they still require manual clicking, technician oversight, or expensive eye-tracking hardware. The invention overcomes these limitations by applying deep learning gaze estimation to ordinary video input, eliminating both specialized equipment and manual response requirements. This results in a more natural testing environment, improved patient accessibility, and significantly reduced costs compared to existing technologies. Applications:
Advantages: