AI Algorithm Shows High Accuracy in Longitudinal Home OCT Monitoring of Neovascular AMD
Overview
A 2025 ASRS study demonstrated that an AI-based algorithm for home OCT monitoring in neovascular AMD patients closely matches expert grader assessments over time. The personalized AI approach achieved 99.1% sensitivity and 89.4% specificity in detecting disease activity.
Background
Neovascular age-related macular degeneration (AMD) requires frequent monitoring to detect disease activity and guide treatment. Home optical coherence tomography (OCT) devices offer near-daily imaging outside the clinic, potentially improving disease management. Prior studies showed good agreement between AI and expert graders on home OCT scans, but longitudinal validation was lacking. This study evaluated the accuracy of an AI algorithm compared to expert graders over an extended period in 180 patients.
Data Highlights
| Metric | Value |
|---|---|
| Sensitivity (personalized thresholds) | 99.1% |
| Specificity (personalized thresholds) | 89.4% |
Key Findings
- This is the first longitudinal study validating AI-based home OCT technology against human experts in neovascular AMD.
- Personalized detection thresholds set by physicians yield very high sensitivity (99.1%) and good specificity (89.4%).
- AI detection of subretinal and intraretinal hyporeflective spaces correlates strongly with expert grader assessments.
- Near-daily home OCT scans analyzed by AI provide reliable monitoring of disease activity over time.
- The study included 180 patients monitored longitudinally, enhancing the robustness of findings.
Clinical Implications
The high sensitivity and specificity of AI algorithms with personalized thresholds support their use in home OCT devices for monitoring neovascular AMD. This technology can facilitate earlier detection of disease activity, potentially improving treatment timing and patient outcomes. Clinicians can consider integrating AI-assisted home OCT monitoring into routine care to complement in-office evaluations.
Conclusion
Longitudinal validation confirms that AI-based home OCT monitoring is a reliable tool for detecting neovascular AMD activity, closely matching expert human graders. This advancement promises to enhance patient management through more frequent, accurate home-based assessments.
References
- Mukkamala KM et al. 2025 ASRS Annual Meeting -- Longitudinal Validation of AI-Based Home OCT in Neovascular AMD
This content is an AI-generated, fully rewritten summary based on a published scholarly article. It does not reproduce the original text and is not a substitute for the original publication. Readers are encouraged to consult the source for full context, data, and methodology.